• eMBB: Enhanced Mobile Broadband

• URLLC: Ultra-reliable and Low-latency Communication

• mMTC: Massive Machine-Type Communications

• How can we tell people in remote locations about training in Huawei?

• We can call them using fixed-line phones or mobile phones, send text messages,
and send emails or WeChat messages over the Internet.
• In 1837, Samuel Morse from the US invented the telegraph. This marked the first
device to quickly transmit information to any corner of the world.

• In 1876, Alexander Graham Bell invented the telephone. In the 19th and early
20th centuries, telephones had no dial, so it was impossible to dial a number
directly. Instead, an operator would connect the calling party. A telephone with a
rotary dial was invented by the 1920s, and push-button phones went to market
in the 1970s and 1980s. These also had the CLIP function.

• However, users needed to be near a fixed-line phone to receive calls. As such,

people wanted a more convenient communication tool. First, pagers were
invented, becoming transition products. With a pager, the called party could see
the calling party's number, but not communicate with the calling party directly. A
pager was also known as a beeper, because it beeped after someone called.

• In 1973, Dr. Martin Cooper of Motorola in the US invented the first handheld
cellular phone.
• Wired communication: The transmission media is tangible, such as conducting
wires, cables, or optical cables.
• Wireless communication: The propagation medium uses electromagnetic waves.
Microwave, shortwave, mobile, and satellite communications are all of this kind.
The electromagnetic wave transmission should meet the following requirements:
c=λxf
▫ λ: wavelength of radio waves, in meters
▫ f: frequency of radio waves, in Hz
▫ c: light speed, fixed at 299792458 m/s
• Spectrum is a physical quantity that exists in nature and cannot be increased or
decreased. Therefore, spectrum needs to be carefully allocated. The International
Telecommunication Union (ITU) defines the usable frequency range of
electromagnetic waves as 3 kHz to 300 GHz, which are divided into eight
segments: very low frequency (VLF), low frequency (LF), medium frequency (MF),
high frequency (HF), very high frequency (VHF), ultra high frequency (UHF), and
tremendously high frequency (THF, that is, terahertz radiation).

• The propagation characteristics of electromagnetic waves vary across frequency

bands.

▫ Low frequency bands feature small propagation loss, long coverage

distance, and a strong diffraction capability. Low frequency resources are
limited, and the system capacity is small.

▫ High frequency bands come with large propagation loss, short coverage
distance, and poor diffraction capability, as well as difficult implementation
and high costs. High frequency resources are limited, and the system
capacity is large.

• Using the spectrum more efficiently and obtaining a higher transmission rate
with the limited spectrum becomes a goal for continuous breakthrough in
wireless communications.
• Radio communications use radio waves to transmit voice, data, text, and images
across a certain space.
• Very low frequency (VLF) and low frequency (LF): 3~300 KHz
▫ According to the International Telecommunication Union (ITU), the
frequency band with a range of 3 kHz to 300 kHz is called VLF and LF. Its
wavelength can reach tens of kilometers and has strong diffraction
capability to easily cover the entire earth. Therefore, this frequency band is
initially used for navigation of air and sea transportation.
• Medium frequency (MF): 300 kHz~3 MHz
▫ When the capability of radio waves to transmit information such as sound
was discovered, MF initially became the preferred frequency band for
regional radio stations. MF is also used in many navigation systems.
• High frequency (HF): 3~30 MHz
▫ In the radio broadcasting field, HF is called short wave. The first global
broadcasting station and global communications radio station were
implemented in the HF range as HF allowed for ultra-long-distance
transmission through ionospheric reflection without requiring the ultra-high
power of the transmit stations.
▫ Radio frequency identification (RFID) and Near Field Communication (NFC)
work in this frequency range, where NFC works in 13.56 MHz, and RFID
additionally uses 27.12 MHz. Selecting this frequency band reduces the
design difficulty and manufacturing costs of the receiver and transmitter,
rather than increases the transmission distance.
• Very high frequency (VHF): 30~300 MHz
▫ Since LF and HF have been used for transmitting sound globally, VHF was
then developed and utilized to achieve visualized, two-way communication.
FM radios, walkie-talkies, pagers, cordless phones, and wireless TVs all work
in this frequency band. The popularity of these products has profoundly
affected social development.
▫ In addition, VHF is also used in international maritime communications,
aviation navigation, aviation ground communications, and so on.
• Ultra high frequency (UHF): 300 MHz~3 GHz
▫ Most digital wireless communication technologies, are provided in this
frequency range. Given the massive applications, the use of this frequency
range must be licensed in most cases in countries around the world, in a
form such as a mobile phone operator-specific license.
▫ Countries also define unlicensed frequency bands (Wi-Fi and Bluetooth) in
this frequency band. As long as the device's power complies with
regulations, using the 2.4 GHz frequency band does not need to be licensed
by a country.
• Super high frequency (SHF): 3~30 GHz
▫ The UHF is rather congested with a large number of wireless
communications working in it. Therefore, a higher frequency band is
required to further improve the transmission rate. From 802.11n onwards,
the 5 GHz unlicensed frequency band is used to achieve above-Gigabit-level
Wi-Fi speeds. Due to its large bandwidth, the 5 GHz unlicensed frequency
band helps further improve the transmission rate.
▫ As for the 5G communications standard, the mmWave band 28 GHz, in
addition to the 2.4 GHz frequency band that has been used in LTE, is
utilized to achieve an ultra-high transmission rate.
• Extremely high frequency (EHF): 30~300 GHz
▫ In the next-generation Wi-Fi standard IEEE 802.11ad, the 60 GHz frequency
band is used to reach a maximum transmission rate of 7 Gbps. Before IEEE
802.11ad was released, Wireless HDMI implemented wireless transmission
of HDMI signals within 10 meters in the 60 GHz frequency band. Despite
many restrictions, the EHF is definitely another path of wireless
communications. To achieve the transmission rate of over 10 Gbps, the EHF
must be fully utilized.
• Tremendously high frequency: 300 GHz~3 THz
▫ The wavelength of the THF is between the EHF and the infrared ray. The
electromagnetic wave in the THF has various features of the light wave.
Therefore, the THF can scan an object like a ray. Although the imaging
quality is inferior to that of the X ray, the THF does not have a radioactive
effect on the object.
▫ Due to its exclusive features, the THF is used in imaging and security, but
does not lead to many breakthroughs in communications.
• The first generation (1G) of mobile communications system was born in Chicago,
USA, in 1986. The most typical representative of the 1G network is the popular
mobile telephone launched by Motorola in the 1990s.

• The second generation (2G) uses digital modulation technology and the main
communication standards include code division multiple access (CDMA, used by
Motorola) and global system for mobile communications (GSM, used by Nokia).
In the 2G era, mobile phones could access the Internet, but with a low data
transmission rat.

• The third generation (3G) substantially increased the transmission rate by

developing new electromagnetic spectrum and formulating new communication
standards. With the 3G-capable iPhone released in 2008, users could browse web
pages, send and receive emails, make video calls, and watch live videos from this
hand-held device, dawning the era of mobile multimedia.

• Supporting high-definition movies and fast transmission of large data, 4G ushers

in the era of the mobile Internet. Today, 4G is indispensable to our lives, like
water and electricity.

• 5G expands communication from just people-people to people-people, people-

things, and things-things, achieving interconnection of everything. 5G networks
will enable the digitalization of the entire industry, bring tremendous changes,
and reshape our life, work, and business models.
• Use viewing a TikTok video as an example:

▫ Step 1: A user turns on the mobile phone. The mobile phone automatically
accesses the network through the base station and transport network. After
performing authentication and registration, the core network allocates an
IP address and a bearer to the user.

▫ Step 2: When the user runs the mobile phone application, a service request
is sent to the corresponding port number and destination IP address
through the bearer.

▫ Step 3: The application platform responds to the request with the

corresponding data packet. The video then starts to play.
• DIS: Data Ingestion Service
• DAU: Data Access Unit

• DSA: Discrete Spectrum Aggregation

• Wireless mobile networks have gradually changed people's way of life, learning,
and entertainment: 1G enabled voice communications; 2G provided SMS and
MMS services; 3G opened mobile Internet; and 4G brought video and apps.
Overall, the evolution of wireless mobile networks has changed and most
importantly improved communication between people.
• The latest fifth generation mobile communications system is emerging, partly in
response to the explosive growth in mobile data traffic, massive device
connections, as well as new services and applications.
• In June 2015, ITU-R defined three types of 5G applications as well as eight types
of capability requirements for 5G networks, including improved throughput,
latency, connection density, and spectral efficiency improvement.

• 5G will support enhanced mobile broadband (eMBB), ultra-reliable and low-

latency connections (URLLC), and massive machine-type connections (mMTC) to
meet diversified service requirements of network capability extremism, network
capability differentiation, and network convergence, a new era of Internet of
Everything (IoE) is quickly approaching.
• Network slicing enables operators to slice a network into multiple virtual
networks on the same hardware infrastructure to allocate resources as required
and flexibly combine capabilities. This technology enables different requirements
of diverse services to be met. When new requirements are raised but the network
cannot meet them, operators only need to virtualize a new slice to launch
services quickly, without needing to deploy a new network.
• With 5G, new technologies, such as Internet of Things (IoT), artificial intelligence
(AI), cloud computing, and big data will flourish, enabling IoE where a massive
number of devices will connect to networks. These devices belong to various
industries and have different characteristics and requirements, posing different
requirements on network mobility, security, latency, reliability, and bandwidth.
▫ eMBB: includes service applications requiring high transmission rates, such
as AR, VR, and HD video live broadcast. The breakthrough in spectrum
utilization and bandwidth technologies on the wireless network enables 5G
to provide 10 times the transmission rates of 4G.
▫ mMTC: supports applications that consume little power but have a high
density of connections, such as service applications of water meters, street
lamps, and cameras in IoT. With multi-user shared access, ultra-dense
heterogeneous networks, and other technologies, 5G can support 1 million
devices per square kilometer, which is over 100 times that of 4G.
▫ URLLC: oriented toward service applications that feature quick response,
security, and high reliability, such as autonomous driving and telemedicine.
In this scenario, a large amount of data needs to be processed
instantaneously and decisions need to be made quickly. Therefore, the
network needs to provide high bandwidth, low latency, and high reliability
to ensure millisecond-level E2E latency and reliability of 99.999% or higher.
• Research at Huawei Wireless X Labs shows that Cloud VR will be the future trend
of VR.
• In the local VR mode, a VR terminal needs to be connected to a local server
through a cable, and the cost is high but user experience is poor. With Cloud VR,
the VR terminal is wireless and connected to the cloud server, which handles the
rendering. Cloud VR greatly reduces the cost and improves user experience.
• Cloud VR proposes higher requirements on mobile networks, mainly on the
bandwidth and latency. For example, entry-level experience requires a bandwidth
of 100 Mbps and a latency of 10 ms, while the ultra-high experience requires a
9.4 Gbps bandwidth and a 2 ms latency, which can only be achieved with 5G.
• VR applications mainly focus on videos and gaming currently and will be
expanded to more applications in the future.
• For low-latency service applications (such as autonomous driving, remote surgery,
and human-robot collaboration), the response time must be less than 10 ms or
even 1 ms. After being obtained through a terminal, information is sent to a
remote server on the network side for processing and then sent back to the
terminal. It is difficult to complete this task within 10 ms. As more and more
services require low latency, cloud computing lags behind in meeting these
requirements.

• With 5G, edge and cloud computing will collaborate. Cloud computing focuses on
big data analysis and processing of non-real-time and long-period data. Edge
computing is closer to application entities and can respond faster. Therefore, it
can better assure service quality when low latency is required.
• In terms of network construction, edge computing integrates traditional wireless
networks with services. Multi-access Edge Computing (MEC) is deployed in the
central equipment room of an enterprise or the equipment room of an edge
radio base station, which is located in the edge data center (DC).

• The MEC terminates services by steering and processing traffic locally. It also
hosts third party apps to provide tailored and differentiated services to enterprise
customers, and agile, smart, tailored service to end users in real time.

Edge DC Local DC Center DC

Access Aggregation Inter-city Backbone

< 30 km 50–100 km > 200 km
< 10 ms 15–30 ms > 30 ms
• The latency in 4G is less than 50 ms, which is half of that in 3G. However,
applications such as autonomous driving still require much lower latency. 1 ms
ultra-low latency can be achieved with 5G, achieving a 50-fold higher response
speed than that in 4G.
• In a 4G network, a car traveling at a speed of 100 km/h still travels for 1.6 m
when the driver stops the car in a case of an emergency. This distance could still
endanger peoples' lives. In a 5G network, the car traveling at the same speed
only travels for 3.3 cm, much like the antilock braking system (ABS).
• Currently, devices such as video cameras, radar sensors, and laser rangefinders
are used to assist single-vehicle automation. The 5G Internet of Vehicles (IoV)
will significantly enhance the perception of the surrounding environment of
vehicles.
▫ Level 0: manual driving; no automation, just notification generation.
▫ Level 1: assisted manual driving. Automatic control of single-vehicle speed
and steering can be implemented, but a human still controls the vehicle
(such as cruise control and ACC).
▫ Level 2: partial autonomous driving. The vehicle speed and steering control
can be automated. The driver must ensure that the vehicle remains in the
correct lane.
▫ Level 3: autonomous driving under certain circumstances. The driver
monitors the system with hands off and intervenes when necessary.
▫ Level 4: advanced autonomous driving. The driver can be eyes off and does
not need to intervene in some predefined scenarios.
▫ Level 5: fully autonomous driving without drivers
• mMTC (large-scale machine-to-machine communications service): By using
technologies such as multi-user shared access and the ultra-dense heterogeneous
network, 5G can support access of 1 million devices per square kilometer, which
is 100 times that of 4G.
• With the rapid development of smart cities, public facilities such as street lamps,
manhole covers, and water meters have been connected to the network and can
be remotely managed, but 5G will trigger more innovations.
• The powerful connection capability of 5G enables all public devices of various
industries in a city to access the intelligent management platform, coordinate
with each other, and be centrally managed by a small number of maintenance
personnel, greatly improving the efficiency of city operation.
• Today's 4G networks cannot meet the requirements of future application
experiences in terms of latency, throughput, and the number of connections.

• Technically speaking, 5G is the next-generation mobile broadband

communications technologies, and it is the main enabler for a wide range of
industries.
• Answer：ABC
• Mobile communications protocols are pre-defined, and the products only perform
operations as defined in the protocols.

• Small modifications in the protocols are implemented through software upgrades;

big changes to protocols may require redesigning hardware and developing
software.
• ITU is an international organization responsible for establishing international
radio and telecommunication management rules and standards. The ITU
formulates standards, allocates radio resources, and develops international toll
interconnection solutions between countries.

• 3GPP is an international standards organization. Its members include European

Telecommunications Standards Institute (ETSI) in Europe, Association of Radio
Industries and Businesses (ARIB) and Telecommunication Technology
Commission (TTC) in Japan, China Communications Standards Association (CCSA)
in China, Telecommunications Technology Association (TTA) in South Korea,
Alliance for Telecommunications Industry Solutions (ATIS) in North America, and
Telecommunications Standards Development Society, India (TSDSI) in India. It
originally stipulated a global (third generation) mobile telephone system
standard in accordance with the ITU International Mobile Telecommunications
(IMT)-2000 plan. Similarly, it also formulated a global (fourth generation)
mobile communications system standard in compliance with the International
Mobile Telecommunications-Advanced (IMT-Advanced) plan. During the World
Radiocommunication Conference 2015 held in Geneva, Switzerland, ITU
Radiocommunication Sector (ITU-R) officially approved the resolution on
promoting future 5G research and formally decided on "IMT-2020" as the legal
name for 5G. Unlike ITU, 3GPP elaborates on technical details, but its standard
plans must comply with ITU requirements.
• ATIS: Alliance for Telecommunications Industry Solutions

• ETSI: European Telecommunications Standards Institute

• ARIB: Association of Radio Industries and Businesses

• TTC: Telecommunication Technology Committee

• CCSA: China Communications Standards Association

• TTA: Telecommunications Technology Association

• TSDSI: Telecommunications Standards Development Society, India

• Strictly speaking, 5G includes LTE evolution and new 5G technologies which are
introduced since 3GPP Release 15. This course focuses on new 5G technologies,
including 5G New Radio (NR) and 5G Next Generation Core.

• LTE was introduced since 3GPP Release 8, LTE-A since Release 10, and 4.5G (LTE-
A Pro) since Release 12. 5G was defined in 3GPP Release 15.
• At MWC 2016, Verizon, KT, SKT, and DCM announced the establishment of OTSA
to formulate unified specifications for 5G tests, promote the allocation of 28 GHz
spectrum resources, develop and discuss 5G use cases, and promote the
development of the 5G industry.

• In 2017, multiple operators and equipment vendors, such as DCM, KT, SKT,
Vodafone, AT&T, BT, DT, Qualcomm, Intel, Nokia, Ericsson, and Huawei,
announced their support for accelerating the standardization of 5G, preventing
OTSA from undermining the unified global 5G standards and interrupting the
balance in the industry.

• Verizon in the US verified key 28 GHz fixed wireless access technologies based on
OTSA test specifications. However, Verizon later shifted to 3GPP and the other
three vendors also announced that they would no longer provide products based
on OTSA specifications. Today, OTSA is no longer functioning.

• 5G networking mode: Phase 1.1 launched the 5G NSA networking architecture.

This architecture worked with the Multiple Stream Aggregation (MSA)
technology to implement collaboration between LTE and 5G. Phase 1.2 launched
the 5G SA networking architecture.
• NB-IoT/eMTC technologies are still evolving. mMTC standards, technologies, and
deployment specifications will be completed in versions later than 3GPP Release
17. According to 3GPP, the NB-IoT & eMTC system can be embedded into the 5G
system and evolve to the NR system in the future.

• Key technologies

▫ New waveform: 5G uses the F-OFDM technology.

▫ Numerology: The slot lengths, frame structures, and the like vary with
subcarrier spacing.

▫ Massive MIMO: 64T64R is supported.

▫ NSA/SA: non-standalone/standalone

▫ D2D: Device-to-device communication is available without using a network.

▫ V2X: vehicle-to-everything
▫ Unlicensed: Unlicensed common frequencies can be used to transmit data.
Currently, unlicensed spectrum resources are mainly distributed in
frequency bands 2.4 GHz, 5 GHz, 6 GHz, and 60 GHz around the world, and
regulations on different frequency bands vary slightly across regions. The
2.4 GHz frequency band is widely used in the Industrial, Scientific, and
Medical (ISM) field and indoor Wi-Fi. The spectrum bandwidth is only 83.5
MHz, resulting in limited frequency resources. In the 5 GHz frequency band,
580 MHz is allocated for the United States and Canada, 455 MHz for
Europe and Japan, and 325 MHz for China. The frequency resources are
relatively idle. Regarding the 6 GHz and 60 GHz frequency bands, the US
FCC released the 6 GHz frequency band for unlicensed use in April 2020
(abundant frequency resources with a total bandwidth of 1200 MHz), and
spectrum authorities in Europe, Japan, and China are actively exploring the
use of the band.
• Answer：C
• Faster commercial use of 5G
▫ Network: It only took six months to release the first commercial network
after the standard was frozen.
▫ Maturity of the terminal industry chain: Terminals and networks were ready
at the same time. It took only two years to release 5G mobile phones worth
about CNY1,000 after launching commercial 5G networks.
▫ User base: It took 10 years for 3G and 5 years for 4G to attract 500 million
users. 5G is expected to take just 3 years to attract 500 million users.
• Simplification is pivotal for 5G, meaning both the ecosystem and network
deployment must be simplified.
▫ A simplified chip is used to support 2G, 3G, 4G, and 5G.
▫ A simplified network adopts the co-site and co-antenna modes.
• The number of 5G terminals is increasing rapidly. These devices include
smartphones, CPEs, industrial modules, mobile Wi-Fi, dongles, UAVs, robots, and
tablets.

• Currently, most 5G terminals on the market support these frequency bands: n78,
n41, n79, n77, n1, and n3.
• Category: indicates the capability level of a UE, such as the data processing
capability (download and upload rates), maximum spatial multiplexing, and
modulation and coding capability.

• Currently, mainstream 5G smartphones support downlink rates of category 22,

4x4 MIMO, 256QAM and even higher modulation schemes, and a maximum of
24 data streams. The LTE download rate can reach 2.5 Gbps, and even 7.5 Gbps
in the downlink when coordinating with 5G.
• In 2020, 20% of new mobile phones will support 5G, and mid-range and low-end
5G mobile phones will be launched in large quantities.
• 5G MiFi can convert ubiquitous 5G network signals into high-speed personal
hotspot signals. It can be used for personal travel, the home, SOHO mobile office,
mobile video live broadcasting, Internet celebrity live streaming, 5G remote live
classes, and UAV inspection backhaul.

• 5G MiFi enables mobile Internet access of multiple devices. The 5G CPE (user-end
device) uses the 5G network to provide stable Wi-Fi coverage in specific
scenarios, meeting the requirements of multi-user data access.

• TD Tech released 5G industrial-grade CPEs to meet requirements in the industry.

▫ Industrial-grade protection: IP65 protection level, shockproof, anti-salt spray

corrosion, and low power consumption.

▫ VPN secure connection, ensuring digital security

▫ Multiple scenarios: CPE+IoT gateway, CPE+enterprise routing

▫ Simplified installation, saving 50% installation time, no need for grounding

or antenna adjustment.
• As the core component for 5G network connections of industrial products, 5G
industrial modules encapsulate hardware such as 5G baseband chips, radio
frequency, storage, and power management units and provide standard software
and hardware interfaces.

• Huawei launches the first commercial 5G industrial module that encapsulates

hardware such as the 5G baseband chip, RF, storage, and power management
units. It provides standard software and hardware interfaces towards external
controllers, sensors, storage devices, Wi-Fi, Bluetooth, and Ethernet.

• Industrial modules have a long life cycle and usually support either both NSA and
SA networking or only SA networking. With the maturity of the 5G terminal
ecosystem and large-scale deployment of 5G networks, the price of 5G modules
will further decrease.
• The candidate frequency bands of mmWave mainly include 24.25–27.5, 37–40.5,
42.5–43.5, 45.5–47, 47.2–50.2, 50.4–52.6, 66–76, and 81–86 GHz. 27.5–29.5 GHz
(28 GHz) is the industrial frequency band.
• Considering coverage and capacity, regular mobile phones prefer to adapt to C-
band networks. The global C-band ecosystem is relatively complete, while
mmWave is used as a supplementary frequency band for hotspots.

• The Ministry of Industry and Information Technology (MIIT) of China took the
lead in releasing the frequency usage plan for the 5G system in the 3000–5000
MHz frequency band. The plan designates the 3300–3400 MHz (for indoor use
only), 3400–3600 MHz, and 4800–5000 MHz frequency bands for 5G.
• When the C-band spectrum is unavailable, 2.6 GHz is selected as the first
frequency band of 5G. In addition, dual connectivity with LTE 2.1 GHz/1.8 GHz
can be used to improve user experience on 5G.

• MM: massive MIMO

• AAU: active antenna unit

• In different countries, frequency resources are allocated to different operators,
and each operator may own only a part of a frequency band. The following uses
spectrum allocation to operators in China as an example:

▫ 900 MHz: 24 MHz for China Mobile and 6 MHz for China Unicom, both for
GSM

▫ 1800 MHz: 25 MHz for China Mobile for GSM; 30 MHz for China Unicom,
with 10 MHz for GSM and 20 MHz for LTE; 20 MHz for China Telecom for
LTE

▫ 2.1 GHz: used by China Mobile and China Unicom for UMTS; 20 MHz for
China Telecom for LTE

▫ 2.6 GHz: The frequency resources originally used by China Telecom and
China Unicom will be reallocated to China Mobile, which exclusively
occupies 160 MHz for 5G network construction.

▫ 3.5 GHz: 100 MHz for China Telecom and China Unicom separately, for 5G
network construction

▫ 4.9 GHz: 100 MHz for China Mobile and 60 MHz for China Broadcast &
Television, for 5G network construction.

▫ 700 MHz: owned by China Broadcast & Television, for 5G network

construction
• The MIIT has licensed the 5G frequency bands for the three major operators in
China:

▫ China Telecom: 3400 ~ 3500 MHz

▫ China Unicom: 3500 ~ 3600 MHz

▫ China Mobile: 2515 ~ 2675 MHz, 4800 ~ 4900 MHz

• Multi-spectrum full-service 5G construction ensures a continuous leading position
in wireless networks.
• Hierarchical networking with the macro base station, pole site, and indoor
distributed system achieves full 5G coverage. The legacy networks can be
upgraded to inherit the existing advantages.

• Macro base station

▫ 64T/32T hybrid networking implements wide, continuous, and shallow

coverage.

▫ Existing 8T sites are reused in rural areas and upgraded to support NR,
effectively protecting investment.

• Pole site

▫ Legacy networks can be upgraded to quickly inherit advantages.

▫ New site deployment improves netwok performance.

• Metro

▫ High-power 2T devices are used with leaky cables to cover lines.

▫ The DIS meets the large-capacity requirements of the platform.

• High-speed railway

▫ 8T RRUs are used to cover the lines, providing optimal user experience.

▫ 2T RRUs are mainly used for tunnel coverage, and existing 4G devices can
be reused for line coverage, which is the most cost-effective.

• Indoor distributed system: intended to fill coverage holes, provide deep

coverage, and offload traffic sharing in areas of high value

• DIS: Digital Indoor System

• DAS: distributed antenna system

• Simulation: In common urban areas, 32T accounts for approximately 40%. The
network performance does not deteriorate, reducing investment by 13%.

• Live network test: The 64T/32T combination of macro base stations can meet the
requirements for coverage and capacity in all urban scenarios.
• Planning objectives

▫ In the initial phase, focus the construction on city clusters, key cities, and
key scenarios to achieve continuous coverage in urban areas of prefecture-
level cities, key coverage in counties and towns, and indoor coverage in key
scenarios. In three years, aim to provide basically continuous outdoor
coverage nationwide and improve indoor coverage.

• Key points of planning:

▫ Focus on high-traffic and high-value areas and provide differentiated

coverage based on the 5GtoC/5GtoB customer distribution. Implement
precise network construction based on the distribution of high-ARPU users,
high-value terminals, and high-traffic users. Deploy indoor distribution
systems in key indoor scenarios.
• Currently, NB-IoT and eMTC are in the initial stage of incubation. Inventory IoT
mainly lies in 3G and 4G networks. 4G IoT evolution is an important step in 5G
IoT.

• The market space of vertical industries is huge, but the industry maturity is low
and needs to be cultivated for a long time.
• By 2025, the digital economy, vitalized by new technologies, will reach USD23
trillion.
• Answer：ABD
• 5G networks are built based on existing 4G networks. To protect investment,
eNodeBs and gNodeBs may coexist. In this case, 5G networks may have multiple
network architectures.
• As specified in the 3GPP specifications, there are two standards for 5G network
architectures: standalone (SA) mode and non-standalone (NSA) mode. In NSA
mode, eNodeBs and gNodeBs coexist on the radio access network (RAN), and the
EPC or 5GC is used as the core network. In SA mode, gNodeBs are used on the
RAN, and 5GC is used as the core network. This architecture is the ultimate goal
of 5G network evolution.
 Applicable Data Split Core
Option Option Characteristic
Scenario Anchor Network
3 eNodeB The number of eNodeBs is greater than that
of gNodeBs in the early stage of 5G
3a EPC
deployment.
These options are standardized early, and
relevant products have been well developed,
Early stage of facilitating market promotion.
3 5G 4G NSA DC data split is supported, ensuring
deployment service continuity.
3x gNodeB
Small reconstruction to the existing live
network and small investment
New services and functions related to 5GCs
are not supported.
Only eMBB services are supported.
eLTE The 5GC is introduced in the early and
7
eNodeB middle stages of 5G deployment. eNodeBs
co-exist with gNodeBs.
Early and 7a 5GC
eNodeBs cover a wide range, and DC
middle stages
7 5G ensures good user mobility.
of 5G
New services and functions related to 5GCs
deployment
7x gNodeB are supported for better user experience.
The upgrade and reconstruction workload
for eLTE eNodeBs is overwhelming.
4 gNodeB In the middle and later stages of 5G
deployment, gNodeBs gradually replace
Middle and
eNodeBs on live networks.
later stages
4 5G Complete new 5G functions are supported,
of 5G 4a 5GC and the traffic gain is noticeable.
deployment
It takes a long time to maturate the
industry.
Option Characteristic
The existing EPC network architecture cannot meet the requirements of
1 ultra-low latency services, such as autonomous driving and industrial
control.
The ultimate goal of the 5G network architecture. It supports eMBB,
mMTC, and URLLC scenarios, facilitating the expansion of vertical
2
industries. Simple networking but with long construction period. The
investment on the live network cannot be protected.
The peak rate, latency, and capacity of eLTE eNodeBs cannot match
5 those of gNodeBs. The reconstruction of the live network is large, and
the commercial value is low.
The 4G core network restricts the advantages of 5G base stations, such
6 as high throughput and high capacity. The possibility of commercial use
is low.
• Among the NSA networking modes, only the 5G radio system needs to be
introduced, and the EPC needs to be upgraded to support 5G services in Option 3
series, which helps accelerate 5G deployment and enables flexible network
transition. Therefore, Option 3x is preferred by many operators at the initial stage
of 5G deployment. Option 3 series networking modes require fewer changes on
the terminal, RAN, and core network sides, and enable users to experience 5G
services as soon as possible. Option 2 is the ultimate networking mode for 5G.
Options 7 and 4 are also possible candidates.
• The core network is responsible for:

▫ Providing user connections

▫ Mobility management and session management

▫ User authentication

▫ Charging management

▫ Providing interfaces for connecting to external networks

• RAN: Radio Access Network

• CN: Core Network

• RNC: Radio Network Controller

• BSC: Base Station Controller

• SGSN: Serving GPRS Support Node

• GGSN: Gateway GPRS Support Node

• MSC: Mobile Switching Center

• VLR: Visitor Location Register

• E-UTRAN: Evolved Universal Terrestrial Radio Access Network

• EPC: Evolved Packet Core

• MME: Mobility Management Entity

• S-GW: Serving Gateway

• P-GW: Packet Data Network Gateway

• Modularized network functions facilitate function decoupling and integration.
Decoupled network functions are abstracted as network services, which can be
independently expanded, evolved, and deployed on demand.

• All network functions on the control plane interact with each other through
service-based interfaces. A service can be invoked by multiple network functions,
reducing the coupling degree defined for interfaces between network functions.

• In addition, network-wide functions can be customized on demand to support

different service scenarios and requirements.
• The network functions defined in the 5GC are basically inherited from those of
the EPC. Most network functions in the 5GC have corresponding NE entities in
the EPC.

• Main network functions of the 5GC are as follows:

▫ AMF:

▪ Termination of uplink NAS signaling

▪ NAS signaling security

▪ AS security control

▪ Signaling node for interoperability in 3GPP systems

▪ Reachability management for UEs in idle mode

▪ UE location area management

▪ UE access authentication
▫ SMF:

▪ Session management

▪ UE IP address allocation

▪ User-plane function selection and control

▪ UPF service control

▪ QoS and policy execution

▪ Downlink data arrival notification

▫ UPF:

▪ Anchor point for mobility management on the user plane

▪ PDU session node

▪ Downlink data routing and forwarding

▪ User-plane data detection and policy execution

▪ Uplink service type identification

▪ QoS-based packet filtering and processing

▪ Data packet flag

▫ PCF:
▪ QoS policy control

▪ Charging control
• The centralized CP simplifies O&M, improves O&M efficiency, and accelerates
new service deployment.

• Distributed UP:

▫ The UP location is determined based on service delay requirements (for

example, IoV services)

▫ User experience + bandwidth saving

• DC: Data Center

• Full-stack cloudification and services: Full-stack cloudification (NFVI and VNF)
enables quick resource orchestration, and full-stack services (AMF, SMF, UPF,
PCF, UDM and IMS) enable quick function orchestration.

• Microservice-based software architecture: Lego-style service combination is more

suitable for flexible 5G services.
• Multi-access edge computing (MEC) was proposed by ETSI. It offers application
developers and content providers cloud-computing capabilities and an IT service
environment at the edge of the network. With the MEC system, operators are
able to deploy applications, content, and a part of MBB core network functions
(such as service logic and resource orchestration) at the edge of radio access
networks. Unlike conventional central DCs, MEC nodes are located close to users
and can provide better reliability and service experience.
• MEC Hardware/MEC IaaS: provides MEC hardware and IaaS.

• 5GC UP: The SGW-U/PGW-U/UPF is deployed at the network edge.

• MEP: MEC platform, which provides service registration, discovery, and

deregistration, and platform capability openness

• VAS: Value-added service, which is integrated on the MEP and is provided by

operators for apps, such as TO, NAT, and FW

• ME APP: an edge application developed by OTT vendors and integrated on the

MEP platform, for example, V2X server/CDN/AR VR

• API: network capability openness and platform capability openness

• Video optimization: Wireless analysis applications are deployed at the network
edge to assist TCP congestion control and bit rate adaptation.

• AR: Edge applications quickly process user locations and camera images to
provide users with real-time auxiliary information.

• Enterprise traffic steering: User-plane traffic is distributed to enterprise networks.

• IoV: The MEC analyzes data of vehicles and roadside sensors and sends latency-
sensitive information such as hazards to surrounding vehicles.

• IoT: MEC applications aggregate and analyze messages generated by devices and
make decisions in a timely manner.

• Video stream analysis: Videos are analyzed and processed at the edge to reduce
the number of required video collection devices and the traffic sent to the core
network.

• Assistance for intensive computation: The MEC provides high-performance

computing, processes latency-sensitive data, and feeds back the results to
devices.
• OTT CDN nodes, for example, those provided by Baidu, Tencent, and Alibaba,
have been deployed in cities. If mobile gateways are still deployed on the
backbone network, transmission resources will be wasted and user experience
will be poor. MECs can be deployed in cities to further reduce transmission paths,
thereby reducing latency and improving user experience.
• Backhauling surveillance videos consume a large amount of bandwidth, but most
of the video images are static. Therefore, deploying an MEC server on the RAN to
analyze and process the video content and to transmit only changed video
images can effectively reduce transmission resource consumption.
• Cloud VR extends terminal capabilities and accelerates the development of the
VR industry. High bandwidth and low latency features of 5G can meet VR
requirements. VR applications in 2019 had been expected to require a bandwidth
about 50 Mbps and a latency about 20 ms. In the future, the required bandwidth
will exceed 200 Mbps and the latency will be about 5 ms.

• An example of VR application is a live broadcast of a football game, where

panoramic cameras capture high-definition real-time video streams. The
audience can replay a scene from multiple close-up angles through their VR
devices. With high bandwidth and low latency, MEC is an effective solution for
eliminating dizziness caused by bandwidth and latency limitations.
• What is a transport network? A transport network is a network that is
responsible for transmitting data.

• If we compare the core network to the human brain and the access network to
the limbs, the transport network is the neural network that connects the brain
and limbs and transmits information and instructions.

• The communications network is a pipe, and the transport network is an inner

pipe.
• L3 on the 5G transport network must be moved to at least the aggregation layer
for flexible service scheduling. It is better to move L3 to the access layer.
• AU: Antenna Unit

• RU: Radio Unit

• DU: Distributed unit

• CU: Central unit

• RRU: Remote Radio Unit

• BBU: Baseband Unit

• The solution where the fronthaul optical fiber for the AAU/RRU and the BBU is
shorter than 100 m is used in a D-RAN site. Applicable countries: Europe (99%),
China Mobile (90%), and China Telecom and China Unicom (50%)

• The typical fronthaul optical fiber between the AAU/RRU and BBU is within 10
km (80%), and some optical fibers are between 10 km and 20 km (20%).

• In C-RAN sites, BBU cases are installed in a centralized manner. There are two
modes: 1. BBUs are placed in a centralized manner, and BBPs in the BBUs are not
interconnected. 2. BBUs are installed in a centralized manner, and BBPs in the
BBUs are interconnected through USUs.
• According to the RAN function splitting options proposed by different vendors, a
PDCP-RLC split is the mainstream. It has clear advantages in PDCP data split
scenarios, reduces data bypassing, and saves transmission bandwidth. In addition,
PDCP can be flexibly deployed on DUs and CUs for future MEC deployment.

• After CU/DU separation, the service data and control messages of the DU are still
transmitted to the CU over the F1 interface.

• RRC: Radio Resource Control

• PDCP: Packet Data Convergence Protocol

• RLC: Radio Link Control

• MAC: Medium Access Control

• PHY: Physical Layer

• Compared with the D-RAN and C-RAN, the CU/DU separation solution features
high resource elasticity and is ideal for load sharing and maximizing resource
utilization. This solution can also be used to solve the tidal traffic effect.
• The preceding figure shows the 5G network slicing architecture. Operators can
virtualize slices as needed on physical resources, for example, an eMBB slice for
common Internet services and an mMTC slice for meter-reading services in
vertical industries.
• A network slice is an E2E isolated logical network. Each network slice includes a
RAN sub-slice, a transport network sub-slice, and a core network sub-slice.

• Network slicing features on-demand customization, E2E, and isolation. E2E

network services can be deployed on demand (location and specifications) and
slices have independent lifecycles and are isolated from each other.

• RAN slicing: QoS-based scheduling or RB resource reservation is used to

guarantee services on the RAN side.

• Core network slicing: Microservices are flexibly orchestrated based on service

requirements. Different microservice functions are deployed for different slices.
Microservices of slices can be flexibly deployed at different locations on the
network.

• Transport network slicing: The transport network can implement VPN-based soft
slicing and FlexE-based hard slicing.
• FlexE is used to build pipe slicing. It uses the TDM-based slot allocation mode to
completely isolate services physically, implementing E2E hard pipe slicing. Each
slice corresponds to a type of service with the same or similar requirements. In
each slice, different users may be isolated based on the VPN or QoS. In the
management and control planes, independent views and resource allocation are
available for each slice.
• Air interface technologies, the jewel on the crown of mobile communications,
distinguish each generation of mobile communications. New technologies must
be introduced to the air interface in 5G to adapt to service requirements such as
large bandwidth and low latency.
• Interference: Spectrum fragmentation and radar/satellite interference.

• Secondary harmonic: If C-band is selected, secondary harmonic interference may

occur in the case of uplink and downlink decoupling, DC, or CA.

• Uplink and downlink decoupling: The uplink 1.8 GHz is preferred in terms of the
industry chain, inter-site distance on live networks, user experience, and evolution
of existing devices.

• Roaming frequency: In terms of the 5G primary frequency in major regions, C-

band is likely to become a global roaming frequency.
• High bandwidth is a typical feature of 5G:

▫ The maximum cell bandwidth of sub-6 GHz is 100 MHz.

▫ The maximum cell bandwidth of mmWave is 400 MHz.

▫ The bandwidths below 20 MHz are defined to meet the requirements for
the evolution of the existing spectrum.
• Layer 1 and Layer 3 functions of the NR network are similar to those of the LTE
network. Layer 2 of the NR network includes the SDAP, PDCP, RLC, and MAC
layers.

• NSA is used for initial 5G deployment and the control plane of the air interface
depends on the RRC connection of the LTE network. There is no control plane
over the NR air interface.
• Modulation refers to the information processing of a signal source and adding
the information to a carrier wave. Generally, different information is carried using
different frequencies, phases, and amplitudes of a carrier wave.

• Higher QAM efficiency requires higher SNR. However, the number of phases
cannot be increased infinitely. In later versions, new modulation technologies
may be introduced to increase spectral efficiency.
• Similar to LTE, the lengths of the radio frame and the subframe are fixed, so that
LTE and the NR can better co-exist. The difference is that 5G NR defines a flexible
sub-architecture. The slot and symbol lengths can be flexibly defined according
to the subcarrier spacing.
• D2D allows direct user-plane data transmission between UEs using the operator's
spectrum with the assistance of cellular networks. It can improve spectral
efficiency of a communications system, and ease limited radio spectrum
resources in a wireless communications system to some extent. In addition, it can
effectively reduce the terminal transmit power and battery consumption, as well
as prolong the battery life of UEs. Compared with NFC and WiFi-Direct, D2D
covers a longer distance (more than 1 km). It is a technology for operators to
implement social networking or short-distance communication.

• Characteristics of D2D communication:

▫ Reduces the pressure on the base station backhaul network and reduces the
network delay.

▫ Reduces the terminal transmit power and increases the standby duration.

▫ Improves spectral efficiency and enables sufficient radio spectrum resources.

▫ Local data applications: emergency communication, public security, IoT

• Band combination:

▫ DL 3.3 GHz–3.8 GHz + SUL 700/800/900/1800/2100 MHz

▫ DL 4.4 GHz–5.0 GHz + SUL 900/1800 MHz

• C-band spectrum features large bandwidth, making it perfect for 5G eMBB

services. A vast majority of global operators have selected C-band spectrum as
the preferential 5G frequency band. DL coverage is favored over UL coverage on
C-band for its large DL transmit power and disproportion in the UL and DL slot
configuration of NR. The application of technologies such as beamforming and
cell-specific reference signal (CRS)-free reduces downlink interference and further
increases the difference between C-band uplink and downlink coverage.

• This feature defines new paired spectrum, with C-band for DL transmission and a
sub-3 GHz band (for example, 1800 MHz) for UL transmission, improving UL
coverage. This addresses the issues related to C-band UL coverage.
• Benefits and impacts of Super Uplink

• Capacity gain

▫ After super uplink is enabled, sub-3 GHz uplink spectrum and C-band
spectrum are both utilized for uplink transmission in different slots,
increasing the uplink spectrum resources and the uplink capacity of 5G UEs.

▫ The LTE uplink spectrum resource usage on the live network is relatively
low. With uplink dynamic spectrum sharing, the idle LTE low-frequency
spectrum resources are shared with the NR network, increasing the usage
of low-frequency spectrum resources.

• Coverage gain

▫ The propagation loss of C-band is large, and the coverage at the cell edge
is limited. However, the propagation feature and the uplink user-perceived
rate at the cell edge of sub-3 GHz is superior to that of C-band.
• Massive MIMO is also called full-dimension (FD) MIMO. Generally, antenna
arrays with more than 16 antennas are regarded as massive MIMO arrays.
Compared with traditional MIMO, massive MIMO introduces the beamforming
technology, which is essentially an antenna technology. Massive MIMO is
decoupled from air interface technologies. Therefore, massive MIMO can be used
in both 4G and 5G networks.

• The number of antennas and ports of 5G base stations will greatly increase. A
large-scale antenna array with hundreds of antennas and dozens of antenna
ports can be configured. In addition, the multi-user MIMO technology is used to
support spatial multiplexing transmission of more users, improving the spectral
efficiency of the 5G system by several times. This feature is used to improve user
experience in high-capacity scenarios with a large number of users.
• Massive MIMO provides receive diversity gains of massive antenna arrays. More
antennas indicate stronger interference rejection combining (IRC) capabilities of
the receive diversity. In addition, the receive diversity effectively suppresses severe
attenuation and improves reception and demodulation performance.
• Narrow beams are good for precise coverage. However, not all beam scanning
uses narrow beams.

• Beam optimization is introduced based on traditional RF optimization by using

the beam sweeping technology.
• MU-MIMO has been introduced in the LTE phase. Before massive MIMO is
introduced, the number of antenna dipoles is small. As a result, the proportion of
UEs that can be paired is low, and the gains are limited. Massive MIMO uses a
large number of antenna dipoles to reduce channel correlation between different
UEs and facilitate uplink pairing. In addition, the total number of multiplexing
layers increases with more antenna elements. This way, more UEs can be paired
at the same time.
• A DDoS attack uses the client/server technology to combine multiple computers
as an attack platform and launch DoS attacks on one or more targets.

• SQL injection refers to the situation where the original SQL semantics are
changed by importing parameters using a program. SQL injection occurs when
statements that are not parameterized are concatenated or parameters are not
parameterized during stored procedure invoking. Executing malicious SQL
statements may cause user data leakage and unauthorized data access.

• EMS: Element Management System

• UPF: User Plane Function

• AUSF: Authentication Server Function

• AMF: Access and Mobility Management Function

• SMF: Session Management Function

• NSSF: Network Slice Selection Function

• NEF: Network Exposure Function

• NRF: Network Repository Function

• UDM: Unified Data Management

• PCF: Policy Control Function

• AF: Application Function

• VNF: Virtual Network Function

• MEC: Multi-access Edge Computing

• MEP: Multi-access Edge Platform

• IMSI: International Mobile Subscriber Identity

• TMSI: Temporary Mobile Subscriber Identity

• SUPI: Subscription Permanent Identifier

• SUCI: Subscription Concealed Identifier

• RRC: radio resource control

• UP: user plane

• NAS: non-access stratum

• HTTPS is a common protocol in the industry. Huawei follows the industry policy
to fix vulnerabilities in a timely manner and has complete vulnerability response
mechanisms.
• SEPP: security edge protection proxy

• TLS: transport layer security

• Answer:

▫ True

▫ ABCD

▫ True
• Throughout history — from the agricultural and industrial ages to the
information age — every technological or industrial revolution has promoted a
great leap forward in productivity, driving human civilization to the next level.
Technological forces have continuously driven humans to create a new world,
leading us to the precipice of a new era. The digital economy — a new form of
economic and social development in the information age — is becoming a new
driving force for economic development worldwide as it facilities achieving
economies of scale. It uses digital knowledge and information as a key
production element, and driven by digital technology innovation, it relies on
modern information networks. Deep integration of digital technologies and the
real economy continues to improve the level of digitization and intelligence in
traditional industries while also accelerating the reconstruction of economic
development and government governance modes.
• "5G + Cloud + AI" has become an essential engine that drives the digital
economy development. Reliable 5G networks, strong cloud computing power,
and AI application are collaborating with each other in various industries to
create new service experience, new industry applications, and new industry layout.
Converged and innovative development of "5G + Cloud + AI" will open up new
space for the development of various industries, from digital government to
smart city, from industrial automatic control to agricultural smart management.
This will inject new momentum into government and enterprise transformation
and industry upgrade. — Source: "5G + Cloud + AI": Engine for the New Era of
Digital Economy, China Academy of Information and Communications
Technology (CAICT)
• At present, the digital economy is transforming from the consumer Internet to
the industrial Internet. As the global digital economy further develops into a new
era, the digital economy has pivoted from the consumer Internet to the industrial
Internet. The industrial Internet refers to emerging digital technologies, such as
5G, cloud computing, AI, big data, and IoT, used by traditional industries to
improve internal efficiency and external service capabilities and achieve
leapfrogged development. Its essence is to promote the use of digital
technologies by enterprises to improve efficiency and optimize configuration,
while connecting enterprises and data in the industry chain. — Source: "5G +
Cloud + AI": Engine for the New Era of Digital Economy, CAICT
• Over the next two to three decades, an intelligent society will become reality,
where all things are aware, connected, and intelligent. Everything in the physical
world will be sensed and transformed into digital signals. Multi-sensory channels
(such as temperature, space, touch, hearing, and vision) will enable situational
awareness and interaction to deliver an immersive user experience. The Internet
of everything will bring all data online, providing wide-ranging connectivity
across cities, mountains, and even space to enable intelligence. Everything will
become intelligent thanks to big data and AI, and individuals, families, industries,
and cities will gradually embrace digital twins to enhance the physical world. A
second world — a digital one — will emerge to augment the physical world,
enriching life . All of this will be made possible as ICT technologies continue to
advance. ICT infrastructure will be the foundation of an intelligent world.

• Between 2020 and 2025, CAICT predicts that the commercial use of 5G in China
will directly drive a total economic output of CNY10.6 trillion and generate an
additional CNY3.3 trillion in economic value. Indirectly, these values will increase
to about CNY24.8 trillion and CNY8.4 trillion, respectively. By 2025, 5G is expected
to have directly created over 3 million jobs. This shows that 5G will be a major
contributor to economic growth. 5G technologies will change people's lives and
production methods, and will even bring fundamental changes in society. In the
future, 5G will become the key infrastructure for comprehensive economic and
social digital transformation.
• 5G features large bandwidth, high data rate, low latency, high reliability, and
massive connectivity. A higher data rate is vital for improving user experience. 5G
provides a significant improvement over 4G, offering peak downlink and uplink
rates of 20 Gbps and 10 Gbps, respectively. In the 5G era, services that require
high data rates can be widely adopted. For example, cloud VR requires high data
rates for video transmission and instant interaction, but because 4G cannot
achieve such rates, users still need to rely on expensive local devices for
processing. 5G's high data rate will enable rapid development of cloud VR. 5G
supports connections with a minimum unidirectional air-interface latency of 1 ms
and 99.999% reliability in high-speed mobility scenarios. This ultra-low latency
supports scheduling of sensitive services and provides more secure and reliable
network connectivity for vertical industries such as Internet of Vehicles (IoV),
industrial control, and smart grid. In addition, it enables application scenarios
such as autonomous driving and telemedicine. 5G networks not only provide
services such as network slicing and edge computing for individuals, enterprise
users, and intelligent industrial devices, but also provide millions of connections
per square kilometer in multiple connection modes, bringing things closer
together and achieving intelligent interconnection between people and
everything.
• 5G ushers in new opportunities and an era for smart city development. Each
person, thing, and organization in a city will become an intelligent entity. 5G
provides connectivity for intelligent city entities anytime and anywhere.

• People, things, and organizations are connected and exchange data and
requirements in real time in digital twin cities. Digital twin cities are seamlessly
integrated and interchanged with physical cities, driving the connection of
individual entities in a city into a distributed superbrain. The city will become
more intelligent, fully meeting the personalized requirements of each intelligent
entity in the city.
• The term "Internet of Things (IoT)" was first coined by Massachusetts Institute of
Technology (MIT) in 1999. The original concept of IoT is radio frequency
identification (RFID)-enabled technologies and devices, which interoperate within
the Internet based on agreed communications protocols. IoT aims to intelligently
identify and manage objects as well as interconnect and exchange information.

• IoT is envisaged as a network of devices all connected to the Internet with the
support of sensor technologies such as QR code readers, RFID chips, infrared
sensors, global positioning systems, and laser scanners to realize information
exchange and communication, thereby enabling smart tagging, positioning,
tracking, monitoring and management. Source: International Telecommunication
Union (ITU)

• The Internet of things (IoT) describes the network of physical objects—“things”

or objects—that are embedded with sensors, software, and other technologies
for the purpose of connecting and exchanging data with other devices and
systems over the Internet.— Source: https://en.wikipedia.org/
• The concept of IoT dates back to Bill Gate's 1995 book titled The Road Ahead, in
which he mentions the idea about the Internet of Things. It attracted little
attention due to the development of wireless networks, hardware, and sensor
devices. In 1998, MIT creatively proposed an IoT-like concept, which was then
called the EPC system. In 1999, the Auto-ID Center in the US first proposed the
concept of IoT based on item coding, RFID technology, and the Internet.

• On August 7, 2009, the then Chinese Premier, Wen Jiabao, delivered an important
speech during his inspection tour in Wuxi. He proposed the strategic concept of
"Sensing China", saying that China should seize opportunities and vigorously
develop IoT technologies. On November 3, 2009, he delivered a speech titled
"Enabling Science and Technology to Lead China's Sustainable Development" to
the science and technology community in Beijing. He again emphasized the
importance of scientifically selecting emerging strategic industries and pushed for
relevant players to make breakthroughs in key technologies such as sensor
networks and IoT.
• The "Industry 4.0" program proposes intelligent logistics, which integrates
logistics resources through the Internet, IoT, and logistics networks to fully utilize
the efficiency of existing logistics resource suppliers. Demanders can quickly
obtain service matching and logistics support.
• The "Made in China 2025" initiative aims to further develop China's
manufacturing sector over a 10-year period. By 2020, China aims to achieve
industrialization, further consolidate its status as a manufacturing powerhouse,
and significantly improve the informatization level of its manufacturing sector. It
also aims to master key technologies in major fields, further enhance its
competitiveness in key fields, and greatly improve product quality. The main
objective is for the manufacturing sector to achieve significant progress in
digitization, networking, and intelligence, and dramatically reduce the energy
consumption, material consumption, and pollution of key industries.
• At the end of 2018, China's Central Economic Work Conference defined IoT as a
new type of infrastructure. IoT is becoming the infrastructure of the digital
economy, moving toward the fourth phase. The development of IoT will continue
to be driven by external forces such as digital and intelligent transformation of
industries and consumption upgrade requirements, as well as internal driving
forces such as maturing technologies and ecosystem developments.
• China's manufacturing sector has made significant progress in digitization,
networking, and intelligence, and key industries have achieved dramatic
reductions of energy consumption, material consumption, and pollution.
• Bluetooth is a large-capacity wireless digital communication technology
standard for exchanging data over a short range. It aims to achieve a maximum
data transmission rate of 1 Mbit/s and a maximum transmission range of 10 cm
to 10 m. The transmission distance can reach 100 m by increasing the transmit
power.

• Advantages: high speed , low power consumption, and high security

• Disadvantages: not suitable for multi-point deployment due to limited network

nodes

• Bluetooth is a wireless technology standard for exchanging data over short

distances among fixed and mobile devices, and building personal area networks
using UHF radio waves in the ISM band from 2.4 GHz to 2.485 GHz. The
technology was invented by the telecom giant Ericsson in 1994 as an alternative
to the RS232 data line. Bluetooth is used to connect to multiple devices,
overcoming the challenges of data synchronization.

• Bluetooth is currently managed by the Bluetooth Special Interest Group (SIG),

which is a global community of over 25,000 companies spanning multiple fields
such as telecommunications, computers, networks, and consumer electronics.
• Uu: the interface between UEs and the UTRAN (base stations in the network).
• Evolved universal terrestrial radio access network (E-UTRAN): a new radio access
architecture defined in 3GPP Release 8. It features high transmission rate, low
latency, and optimized data packets. The E-UTRAN contains several eNodeBs and
provides the E-UTRA user plane (PHY/MAC) and control plane (RRC) protocols
that are terminated at the UE.
• User equipment (UE)
• Non-access stratum (NAS): a functional layer between the UEs and the core
network. The NAS supports transmission of services and signaling messages
between the UEs and core network.
• Evolved packet core (EPC): an evolved 3GPP system architecture that focuses on
higher data rates, lower delay, and packet optimization, and supports multiple
radio access technologies (RATs).
• Mobility management entity (MME): a key control node on the 3GPP LTE access
network responsible for signaling processing. Specifically, it is responsible for the
positioning and paging of UEs in idle mode, including the relay. It involves the
bearer activation/deactivation process and selects a serving gateway (SGW) for a
UE when the UE is initialized and connected. The MME interacts with the home
subscriber server (HSS) to authenticate subscribers and allocates them temporary
IDs. It also supports lawful interception and listening. The MME provides the
control function interface for 2G/3G access networks through the S3 interface
and provides the S6a interface towards the HSS for roaming UEs. The MME
manages mobility and bearers, authenticates users, and chooses the SGW or
PGW.
• Low costs: Huawei SingleRAN solution supports upgrade and reconstruction of
legacy network devices, thereby reducing network construction and maintenance
costs. NB-IoT chips are specifically designed for IoT devices. These chips apply
only to narrowband and low rates, and they support only single-antenna
transmission and half-duplex mode in compliance with IoT requirements. In
addition, the signaling processing of NB-IoT chips is simplified. Due to these
factors, NB-IoT chips cost only a few dollars.
• Low power consumption: NB-IoT uses the power saving mode (PSM) and
extended discontinuous reception (eDRX) features for IoT services where small
packets are occasionally transmitted. With these features, a device enters the
dormant state immediately after sending data packets and wakes up only when
data reporting is required again. An IoT device can be in the dormant state for up
to 99% of its operational lifetime, achieving ultra-low power consumption.
• Wide coverage: NB-IoT is designed for IoT, especially LPWA connections. It uses
retransmission over the air interface and ultra-narrow bandwidths to provide
gains of over 20 dB compared with GSM. This means that fewer sites can cover
wider areas with strong signal penetration (down to basements). Devices such as
electricity and water meters in hard-to-reach areas can be covered, and pet
tracking and other services that require broad coverage can be used.
• Massive connectivity: The low prices of NB-IoT terminals mean that such
terminals can be widely deployed across various industries, especially industries
using instruments. For the same eNodeB, using NB-IoT can provide 50 to 100
times the number of connections provided by existing wireless technologies. One
sector can support hundreds of thousands of connections, with support for low
delay sensitivity, ultra-low device costs, low device power consumption, and
optimized network architecture.
• IoT devices have different communication requirements from mobile phones. In
most cases, an IoT device only sends uplink data packets. The device itself
determines whether to send data packets, and it does not need to remain in
standby mode waiting for calls from other devices. In contrast, a mobile phone
must wait for network-initiated call requests.

• If the design principles used in 2G/3G/4G were applied to IoT communication, the
behavior of IoT devices would be the same as that of mobile phones. In that case,
IoT devices would consume a large amount of power while monitoring requests
that may be initiated by the network at any time. Consequently, it would be
impossible to achieve low power consumption.

• Based on the NB-IoT technology, an IoT device enters the dormant state
immediately after sending data packets and wakes up only when data reporting
is required again. An IoT device can be in the dormant state for up to 99% of its
operational lifetime, achieving ultra-low power consumption.
• Benefits:

▫ Extended discontinuous reception (eDRX) enables UEs to remain in the

sleep state for a longer period than DRX, achieving greater power savings.

• Application scenarios:

▫ This feature applies to scenarios where LTE UEs are used to perform IoT
services and low power consumption is required (for example, smart meters,
sewer monitoring, and protection for children and the elderly).

▫ This feature applies to cells with eDRX-capable UEs.

• MMMB: Multimode Multiband

• MB: Multiband

• BB: Baseband

• PMU: Power Management Unit

• PA: Power Amplifier

• SoC: System on a Chip

• Maximum coupling loss (MCL) : used to evaluate the extent of coverage
(penetration range). A larger value indicates a greater extent of coverage.

• The existing access technologies cannot meet IoT's requirements for wide
coverage. 3GPP TS 45.820 proposes that the LPWA technology provides an MCL
20 dB higher than GSM and LTE to support the deployment of IoT services such
as reading water and electricity meters.

• With this 20 dB gain, NB-IoT base stations can cover underground garages,
basements, underground pipes, and other places where signal coverage was
extremely weak.

• The frequency spectrum density is improved.

▫ The uplink and downlink physical channel formats and modulation

specifications have been redefined so that uplink and downlink control
information and service information can be transmitted in narrower
bandwidths compared with LTE. At the same transmit power, the PSD gain
is increased, reducing demodulation requirements on the receiver.

▫ Repetition of transmission: The encoding scheme used for repeated

transmission is introduced, promoting transmission reliability when the
channel condition is extremely unfavorable.
• The breakthroughs of 5G technologies mean new opportunities for the IoT
industry. Compared with 4G networks, 5G networks offer more powerful
communication and bandwidth capabilities, and can meet the requirements of
high-speed stability and wide coverage of IoT applications. 5G enables many IoT
applications that are still in the theoretical or experimental phases to be quickly
applied.

• With all things connected, machine-type, large-scale, and mission-critical

communications impose higher requirements on the network speed, stability, and
latency. People have strong demands for mobile Internet applications with heavy
traffic and connectivity of everything. New applications, including autonomous
driving, AR, VR, and touch-sensitive Internet, urgently need 5G. This means that
5G will bring significant business opportunities to IoT industries.
• 5G eliminates the problems of IoT transmission rate and huge data connection
sets.
• The IaaS layer provides basic computing, storage, and network services. Typical
IaaS cloud services include Elastic Cloud Server (ECS).

• The PaaS layer provides an environment for development, as well as components

for development, and an environment for running applications. Typical PaaS
cloud services include database services.

• The SaaS layer provides software-related functions through web pages. Typical
SaaS cloud services include Office 365.
• Virtualization technology is the cornerstone of cloud computing. Virtualization
enables one physical server to run multiple VMs, which share the CPU, memory,
and I/O hardware resources of the physical server. However, the VMs are
logically isolated from each other.

• In computer science, virtualization refers to the creation of virtual computing

resources, that is, physical computing resources are virtualized into one or more
operating environments. Virtualization implements simulation, isolation, and
sharing of resources.

• Virtualization is a process in which a lower-layer software module provides a

virtual software or hardware interface to an upper-layer software module. This
interface is completely consistent with an expected operating environment. In this
way, the upper-layer software module can directly run in the virtual environment.
Virtualization abstracts a resource into one or more parts by means of space
division, time division, and simulation.
• Common virtualization technologies include:

▫ Memory virtualization: Page File

▫ Disk virtualization: RAID and Volume

▫ Network virtualization: VLAN

• Virtualization creates an isolation layer to separate hardware from upper-layer

applications so that multiple logical applications can run on one hardware
resource.

• FusionSphere virtualizes x86 servers, including:

▫ Computing capability: CPU/memory virtualization

▫ Storage: VIMS file system

▫ Network: distributed virtual switch

• First, cloud services will be fully upgraded in the 5G era. In the 4G era, cloud
computing is widely adopted and brings convenience to many enterprise users.
However, for individual users, there are limited opportunities to access and use
the cloud. Given 5G's improved performance, more and more cloud services will
be upgraded, which in turn will have a direct impact on people's lives. 5G will be
deeply integrated with IoT, IoV, smart city, industrial Internet, and smart
healthcare, enabling people to enter the smart life era.
• Second, the advent of the 5G era will promote the comprehensive upgrade of
cloud vendors. Network construction will be greatly enhanced, and the rapid
improvement of such construction will drive the overall development of cloud
infrastructure. Cloud service providers need to upgrade and reconstruct their
network architectures, infrastructure, service models, and operation systems to
accelerate the adoption of cloud solutions for vertical industries and keep pace
with the development of the cloud computing era.
• Third, in the 5G era, cloud computing will shift from the network center to the
network edge. As the network improves, more and more devices will be
connected to the network, and user requirements for obtaining data will increase.
If user-side data acquisition is performed at the data center each time, 5G
application experience will be significantly affected. As edge computing develops,
users only need to transmit data to the closest edge data center for processing,
further reducing the network latency and meeting the delivery requirements of
real-time 5G response services in the future. In addition, edge computing can
further accelerate the integration of the industry ecosystem, identify new service
scenarios, and explore cloud service models for vertical industries.
• Applications in the 5G era will be predominantly based on device-network-cloud
synergy in mobile scenarios. Huawei defines these applications as Cloud X
services.

• 5G brings new eMBB networks and edge computing closer to users. This is
expected to change the entire service chain.

• On the device side, the ubiquitous 5G connection and edge cloud capabilities can
migrate the computing, storage, and even rendering capabilities from the client
to the cloud. The device will become very "thin" with lower costs and stronger
mobility. More importantly, these capabilities can reduce the threshold for
service deployment and popularization, enhancing the service vitality.

• On the cloud side, a large number of services are integrated on the cloud,
increasing dependence on the network, edge computing, and future network
slicing capabilities. This further highlights the importance of network capabilities
and strengthens operators' control.

• In the 4G era, videos are used as the benchmark of network capabilities, with
operators playing a role in only the network. In the 5G era, Cloud X will be used
as the benchmark of network capabilities, bringing opportunities for operators to
reshape service models and network benefits.
• Research by Huawei Wireless X Labs shows that Cloud VR rendered on the cloud
will be the developmental trend of VR in the future. In local VR mode, VR
terminals need to be connected to local servers through cables, resulting in poor
user experience and high costs. Cloud VR implements wireless terminals and
image rendering through cloud servers, greatly reducing terminal costs and
improving user experience.

• Cloud VR poses higher requirements on mobile networks, predominantly on

bandwidth and latency. Only 5G networks can meet the requirements of ultimate
VR experience.

• Currently, VR is mainly used in videos and gaming. In the future, it will be used in
more scenarios.
• In addition to the three basic "Vs", the industry proposes "value" as a fourth
dimension, while IBM adds "veracity" (uncertainty of data) as another dimension
of big data.
• To date, there is no universally agreed definition of big data. Most of its
definitions are based on its characteristics.
• The first "V" is variety. There are two aspects. One is that the data sources are
diversified, meaning that the data collected through different channels and
platforms is diversified. The other is that there are various data structures, such
as structured and non-structured data.
• The second "V" is massive volume. As the Internet develops, the volume of data
generated by the Internet is increasing day by day. The volume of data generated
in one year now may be comparable to all the data generated in the past.
• The third "V" is velocity. This involves the entire process of big data, such as the
growth rate and processing speed of data. We have been able to provide
feedback on many types of data in real time. Impacts are made on our life as
soon as the data is collected.
• The last "V" is low value density. Although there is a large amount of big data,
more data does not indicate a better meaning. A great deal of data is
meaningless, and useful data is submerged in the massive amount of useless
data. This is one of the difficulties in big data technologies. We need to perform
in-depth analysis on massive amounts of useless and complex data, and mine
data that is of value.
• The minimum data storage unit is bit. All units are listed in ascending order of
size: bit, byte, KB, MB, GB, TB, PB, EB, ZB, YB, BB, NB, and DB.

• They are calculated at the rate of 1024 (2 to the 10th power):

▫ 1 byte = 8 bits

▫ 1 KB = 1,024 bytes = 8192 bits

▫ 1 MB = 1,024 KB = 1,048,576 bytes

▫ 1 GB = 1,024 MB = 1,048,576 KB

▫ 1 TB = 1,024 GB = 1,048,576 MB

▫ 1 PB = 1,024 TB = 1,048,576 GB

▫ 1 EB = 1,024 PB = 1,048,576 TB

▫ 1 ZB = 1,024 EB = 1,048,576 PB

▫ 1 YB = 1,024 ZB = 1,048,576 EB

▫ 1 BB = 1,024 YB = 1,048,576 ZB

▫ 1 NB = 1,024 BB = 1,048,576 YB

▫ 1 DB = 1,024 NB = 1,048,576 BB
• At present, computers are binary. To facilitate computer calculation, only the
integer power of 2 is used. However, because people are used to using the
decimal system, memory manufacturers use 1,000 as the progress rate. This
means that the actual capacity is less than the nominal capacity. 1,024 is the 10th
power of 2, because if it is set to a larger value, it is not close to the integral
power of 10, which is inconvenient for people to calculate. If the value is too
small, the progress rate is too low and more units are required to meet the
requirements. Therefore, the value is 2 to the power of 10.

• For example, if the nominal capacity of a hard disk is 100 GB, its actual capacity
is calculated as follows: 100 x 1,000 x 1,000 x 1,000 bytes/1,024 x 1,024 x 1,024 ≈
93.1 GB.
• Changes in data structures, data processing methods, and enterprise services lead
to the emergence of big data and its related products. Compared with
conventional data analysis, big data analysis has the following changes:

• Data format

▫ Structured data: Traditional data sets are usually small. Most data
warehouses have a refined extraction, transformation, and loading (ETL)
process and database restrictions. This means that data loaded to data
warehouses is easy to parse with a clean data format.

▫ Non-structured or semi-structured data: The biggest advantage of big data

is that it can capture non-structured data in addition to traditional data.
This means that the data format is broader and the analysis method is
more challenging.
• Data relationships

▫ Relational model: Traditional analysis is based on relational databases.

Relationships between topics have been created in the system, and analysis
is performed based on the relationships.

▫ Uncertain relationship: It is difficult to establish a correct relationship

between all information in the form of non-structured data (such as video
and audio). Therefore, it is difficult to establish a definite relationship
between data.

• Handling method

▫ Directional batch processing: Traditional data analysis is directional batch

processing. We need to wait for the extraction, transformation, and loading
to complete before obtaining the required insight.

▫ Non-directional batch processing or real-time processing: Big data analysis

uses software that is meaningful to data to process non-directional batch
processing or real-time analysis of data.
• When people talk about big data technologies, they often refer to not only data
but also the combination of data and big data technologies. Big data
technologies refer to technologies used for big data collection, storage, analysis,
and application. They use non-traditional tools to process a large amount of
structured, semi-structured, and non-structured data for analysis and prediction.
• During the Dartmouth Summer Research Project on Artificial Intelligence in 1956,
MIT proposed that AI would allow machines to behave like human beings.
However, the definition of AI has been gradually extended since that time. In the
future, super-AI will exhibit more intelligent behaviors than human beings.
• AI is a new technical science that studies and develops theories, methods,
techniques, and application systems to simulate and extend human intelligence.

• Machine learning is a core research field of AI. It studies how computers acquire
new knowledge or skills by simulating or performing the learning behavior of
human beings, and how they reorganize the existing knowledge structure to
improve its performance.

• Deep learning is derived from the research of artificial neural networks (ANNs),
and multilayer perceptron is a type of deep learning structure. Deep learning is a
new field in machine learning that simulates the human brain to interpret data
such as images, sounds, and texts.
• Deep learning is a specific branch of machine learning. To understand deep
learning, it is necessary to fully understand the basic principles of machine
learning.

▫ Task 𝑇: how should the machine learning system process an example. A

sample is a collection of features that have been quantitatively measured
from some objects or events that we want the machine learning system to
process, such as classification, regression, and machine translation.

▫ Performance measure 𝑃: evaluates the abilities of a machine learning

algorithm, such as accuracy and error rate.

▫ Experience 𝐸: Most learning algorithms are allowed to accumulate

experience on an entire data set, whereas some machine learning
algorithms are not trained on a fixed data set. For example, reinforcement
learning algorithms interact with an environment, so there is a feedback
loop between the learning system and its training process. Machine
learning algorithms can be broadly categorized as unsupervised or
supervised based on the learning process.
• To learn Go:

▫ Experience 𝐸1 : playing with itself — Unsupervised and indirect learning

▫ Experience 𝐸2 : inquiring humans when playing with itself — Semi-

supervised learning

▫ Experience 𝐸3 : historical human games — Supervised and direct learning

• Handwriting recognition issue

▫ Task T: identify handwriting texts

▫ Performance measure 𝑃: classification accuracy

▫ Experience 𝐸: classified example library (supervised and direct learning)

• Robots' desire to advance: Look for new games and practice their skills through
tiny changes in the same situation, enriching their training examples.
• AI is experiencing the third wave of development triggered by deep learning. It
has made significant progress in data, computing power, algorithms, and
platforms.

• Mutual enablement between 5G and AI: 5G is the key infrastructure for digital
transformation of various industries in the future. It features high bandwidth,
massive connectivity, and low latency, and will enable AI development in terms of
data, computing power, and application scenarios.

• Massive 5G connections facilitate data collection: IMT-2020 White Paper on 5G

Vision and Requirements predicts that the total number of connected mobile
network devices worldwide will exceed 100 billion by 2030. 5G-enabled
connectivity of everything will bring explosive growth in data volume, types, and
forms, and will collect massive high-quality data for AI training and modeling.
• Answer: C
• Current primary uses: HD video surveillance, VR/AR, remote control and private
networks.
• NaaS-based basic toB service capabilities, like computing resources, are carried
on the public cloud as IaaS.

• Basic service applications and industry solutions can provide PaaS and SaaS
services. However, not all products and functions are carried on the public cloud.

• SME: small and medium enterprise

• Currently: The QoS of original BE services cannot be guaranteed.

• Preferred minimum bit rate (PMBR): Assume that the QCI is set to 6 and PMBR
to 10 Mbps on the base station side. The base station ensures private line
bandwidth based on the PMBR. When implementing in real-life toB scenarios, it
is recommended that dedicated QCIs be planned for the enterprise private line
for the optimal results.

• GBR: The private line is configured to provide GBR services. The base station
preferentially schedules high-priority GBR services to ensure the private line
bandwidth. In this case, high-priority GBR services bypass the default bearer with
QCI 9. This solution gives positive results but consumes a lot of base station
resources.

• Resource reservation: Extended QCI 100 is configured for private line services. The
base station ensures private line services in terms of admission, mobility,
interference reduction, and RB resource reservation. (2021 Q2)
• URLLC is introduced in 3GPP Release 15 and enhanced in 3GPP Release 16. 3GPP
Release 15 was frozen in 2019 Q1, and 3GPP Release 16 in July 2020. 3GPP
Release 15 defines the basic URLLC functions. The technologies noted in 3GPP
Release 15 meet the requirements of some commercial applications. 3GPP
Release 16 further improves reliability and capacity through lower latency.
• When user plane functions are deployed closer to users and there is no
retransmission, the average E2E RTT (excluding applications) is 8 ms and the
maximum is 12 ms based on the current capabilities of the air interface.

• Air interface: pre-scheduling, Grant-free, short TTI

• Microwave: less than 25 μs per site

• Router: single-hop 10-15 μs, leading performance in the industry

• Core network: forwarding latency < 0.1 ms

• Core network: The user plane is deployed closer to users, and MEC is deployed.
The CDN and servers are moved downwards at the same time.

• Bearer: from L3 to edge, shortest path forwarding

• WDM: moved downwards to the convergence layer, direct pass-through at the

optical layer, service accessed within one hop

• Service: E2E extended QCI

• MEC helps move the UPF and application servers downwards.

• Enterprise requirements:

▫ Data processed within campus

▫ Low latency

▫ Third-party application loading

• Operators' requirements:

▫ Enter the toB market for big enterprise applications

▫ Retain control over the network

• In 4G, subcarrier bandwidth is fixed at 15 kHz.

• RAN RTT= Uplink OTT + Downlink OTT

• Uplink OTT = UE processing time + Waiting time in the uplink + Air interface TTI
+ Time used for data reception and processing at the base station

• Downlink OTT = Time for data transmission by the base station + Waiting time in
the downlink+ Air interface TTI + Time used for data reception and processing at
the UE
• CSG: cell site gateway

• ASG: access service gateway

• MASG: mobile aggregation site gateway

• CPE: customer-premises equipment

• UPF: user plane function

• Interference coordination in the time and frequency domains reduces the
capacity of the system.
• Other advantages of 5G networks:

▫ 5G guarantees lower latency.

▫ 5G spectrum is exclusive to operators and delivers stronger anti-

interference capabilities. It can also guarantee better data rates.

▫ 5G base stations have high transmit power. Far fewer 5G base stations are
required to achieve the same coverage as a large number of Wi-Fi 6 base
stations.

▫ 5G networks guarantee a better single-user experience.

• Main problems of using Wi-Fi

▫ Unstable network, which causes the likes of collisions and shutdown. This
could be due to:

▪ Signal blocking. There is no idea about how to deploy multiple APs to

ensure signal quality.

▪ Environment interference being too strong

▪ Single-AP capability being limited, with a packet loss rate of 20%

▫ Poor system reconfigurability. The path has poor flexibility and cannot be
changed once planned. As a result, vehicles wait for too long at the traffic
lights, reducing efficiency.

▪ Current closed-loop control latency: 160 ms (30 + 50 + 20 + 50 + 10),

where the communication time is 100 ms.

▪ Closed-loop control is predominantly implemented locally, and the

system changes will become complex.
• RAN slice management is implemented in the cloud (MCE platform) at the
access layer.

• MANO: Management and Orchestration

• 4K applications will first enter the maturity stage, whereas 5G modules, 8K
applications, and some industry applications need to be continuously developed
with the industry chain. Currently, 5G UHD applications are mainly used for
backhaul in programs such as live broadcast backpacks and 5G + 4K OB vans.
Mature live broadcast backhaul technology improves the immersive experience of
videos. Technologies such as 5G + 4K terminals, cloud-based news collection and
compiling, as well as 4K OB vans are developing rapidly and will enjoy large-scale
application in the next couple of years. Due to the restrictions of device-
technology convergence and internal industry specifications, UHD application
products in healthcare and security protection, such as the 4K endoscope, 4K
operating room display, and 4K monitor, may reach maturity 2 to 10 years later.
5G + 8K terminals and 8K OB vans are still in the exploration stage. It is a slow
process to improve the industry chain and popularize them. Source: CAICT's
White Paper on the 5G Application Innovation, October 2019
• The formula used for calculating the bit rate of the signals collected by original
cameras is:

▫ Pixels x Number of bits per pixel x Frame rate (= 4096 x 2160 x 24 x

30/(10^9) = 6.37 Gbps)
• 5G will empower the UHD industry in three phases. These phases are video
collection and backhaul, video material production on the cloud, and UHD video
program broadcasting.

• 4K/8K cameras convert original video streams into IP data streams through
encoding and stream pushing devices, and then forward the video data to the 5G
base station through the 5G CPE or the encoding and stream pushing device
integrated with the 5G module. During large-scale activities, tens of thousands of
connections are required, and a large number of videos generated through screen
recording using HD cameras or terminals need to be transmitted. The 5G
network helps handle the challenges with its ultra-high network speed, ultra-low
latency, and massive connections. Integrated with 5G modules, the encoding and
stream pushing devices and camera backpacks can provide stable and real-time
video transmission. Compared with traditional cable transmission, the
transmission mode fulfills UHD video backhaul requirements more flexibly
without space limitations.
• Integrated editing and broadcasting at the edge provides several advantages.
These include onsite short video production, real-time analysis of match
information and historical data, and AR-based interaction and customized
spectator viewing for fans in stadiums. It reduces the costs in broadcast trucks
and multi-channel backhaul, improves user experience, and adds new value-
added services.

• The 5G base station transmits video data to the video playback, storage, and
distribution ends through the core network. The video data is then sent to the
video display terminal in multiple modes.

• After UHD video materials are sent to the cloud, corresponding video production
software is deployed on the cloud to produce video materials through desktop
applications and HTML5 pages, and content is then distributed through the 5G
network to implement UHD video production and broadcasting based on the 5G
network.
• High cost of remote and local live broadcast

▫ Satellite van: SD/HD (CNY20,000–30,000 for 8 hours) + CNY40/minute x 9

MHz bandwidth

▫ Ground station: CNY1,200 (10 minutes) + CNY80/minute x 9 Mbps

bandwidth

• 5G video collection backpack

▫ Low cost of 5G backpacks: The cost of the 5G live broadcast solution is far
lower than that of the traditional live broadcast mode (live broadcast
vehicle + satellite).

▫ The 5G live broadcast solution is flexible and free from site restrictions. It is
ideal for supporting natural disaster relief and reporting breaking news.
Journalists are able to broadcast live news with just a phone.

▫ First generation: encoder + CPE; second generation: encoder + mobile

phone; third generation: encoder + 5G module
• Live outdoor broadcasting is limited by the high cost of equipment, 4G
bandwidth, and single live broadcast angle. The combination of 5G mobile live
broadcast and UAV-based broadcasting helps address the challenges.

• Protocol conversion: for example, conversion between RTSP and RTMP

• Intelligent broadcast control: reduces manual dependency, prevents broadcast

collaboration errors, and implements intelligent and automatic broadcast control
in complex live broadcast scenarios that require collaboration of multiple roles
and professionals.
• The live broadcast network of a fixed stadium is overloaded due to the high user
density, causing transmission quality to deteriorate. 5G micro base stations can
be used to provide large-scale coverage inside the stadium for stable and high-
performance transmission.

• 5G enables quick onsite audio and video processing as well as live broadcast of
panoramic VR videos, enhancing the viewing experience for those outside the
stadium.
• Four types of industry video services: video surveillance, live video, remote control,
machine vision.

• Video surveillance does not pose strict requirements on uplink bandwidth and
latency. It has loose requirements on network KPIs and the service SLA assurance
is relatively controllable.

• With video surveillance, numerous HD IP cameras are deployed, generating

explosive media data and requiring efficient network pipes.

• 4K video surveillance is required for important occasions, imposing higher

requirements on the 4K E2E service chain, such as terminals, network bandwidth,
backend storage, and AI computing power, and increasing costs. Therefore, for
4K video surveillance to be put into large-scale commercial use, costs must still
be cut across the entire industry chain.
• 1. Campus LAN: The UPF/MEC is deployed at the network edge, and traffic is
distributed to the video platform and AI applications locally, without bypassing
the public network.

• Applicable to local networking scenarios such as industrial parks, construction

sites, and port mines.

• 2. WAN (slicing): 5G E2E network slicing for the RAN, bearer network, and
core network help meet SLA and security isolation requirements such as
bandwidth.

• Applicable to scenarios with high security requirements, such as smart policing,

inspection, and environment monitoring.

• 3. 5G LAN: The enterprise-built private network is connected to the 5G virtual

private network, implementing 5G-based multi-network Layer 2 and Layer 3
interworking, dedicated data, and network isolation.

• Applicable to wide-range multi-point connection scenarios such as campuses and

complexes.
• Video surveillance technologies are increasingly becoming HD-based, digital,
networked, and intelligent. Video surveillance has changed from visual discovery
to automatic control, representing a technical leap and inevitable market trend.
After HD videos are widely used across the industry, intelligent video surveillance
will take off and embrace a wide range of market applications.
• Road and bridge surveillance vehicles use 5G to continuously transmit 4K video
surveillance data to the cloud in real time, automatically identify and handle
eight types of road distresses (such as subsidence, potholes, cracks, net-shaped
cracks, and upheavals), and automatically dispatch maintenance and acceptance
to replace manual inspection and analysis.
• Virtual reality (VR), augmented reality (AR), and mixed or mediated reality (MR)

• VR is a closed experience of virtual scenarios, while AR superimposes digital

elements on objects and backgrounds of the real world. As an upgrade to AR, MR
can combine virtual and real scenarios, and differs in the strict requirements on
the authenticity of virtual images. Therefore, it can be viewed as a type of AR.

• AR enables interaction and integration of a virtual world (on a screen) with a

real-world scenario by using actuarial position and angle calculation of camera
images and an image analysis technology.

• MR is the combination of real and virtual worlds to create a new environment

and visualization. Physical entities and digital objects coexist and interact with
each other in real time to simulate real objects. It combines reality, AR,
augmented VR, and VR technologies.
• 5G + AR/VR transforms a wide range of industries. The focus is shifting from VR
to AR, with the growth rate of the latter expected to surpass that of the former.

▫ According to IDC's latest report, the global AR/VR expenditure is expected

to reach USD160 billion by 2023, a 10-fold increase of the expenditure in
2019 (US$16.8 billion). The five-year compound annual growth rate (CAGR)
is 78.3%.

• VR is entertainment-oriented. With AR, toC may focus on big entertainment,

while toB focuses on problem solving. The future market scale is promising.

• toB: The largest investments in 2023 will be training (USD8.5 billion), industrial
maintenance (USD4.3 billion), and retail display (US$3.9 billion).

• toC: The market space for VR games, VR videos, and AR games will reach
USD20.8 billion by 2023.

• During the forecast period, hardware will account for over half of all AR/VR
expenditures, followed by software and services.

• AR HMDs have the fastest growth rate, with a five-year CAGR of 189.2%,
followed by VR devices.

• AR software will become the second fastest-growing field, and its expenditure
will exceed that of VR software by 2022.
• The VR/AR industry is ubiquitous and developing rapidly. 5G-enabled high rates
and low latency make cloud-based VR a reality, significantly improving the uplink
convenience and mobility of VR/AR.

▫ In 2017, Facebook released the AR-capable Camera Effects Platform.

▫ Microsoft and other companies have launched next-generation AR smart

glasses with better experience.

▫ Fast chip iteration improves the appliance performance by 30% annually.

• Microsoft Teams-based AR remote guidance can effectively complete overhauling
through a series of instructions such as video calls and photographing.

• Applicable to industrial, post-sales, and other overhauling scenarios

• The AR technology and remote experts are connected in real time to improve the
overhauling efficiency and optimize the work process.
• UAVs were first used in the military field. During World War I, they were used as
trailer aircrafts. After 9/11, large-scale applications (such as movie shooting)
appeared.

• According to flight platform, UAVs can be classified into fixed-wing UAVs, rotary-
wing UAVs, unmanned airships, parawing UAVs, and ornithopter UAVs.

• According to usage, UAVs can be classified into military UAVs and civil UAVs.

• According to size (based on civil aviation regulations), UAVs can be classified into
micro UAVs, light UAVs, small UAVs, and large UAVs.

• According to activity radius, UAVs can be classified into ultra-short-distance

UAVs, short-distance UAVs, mid-distance UAVs, and remote UAVs.

• According to task height, UAVs can be classified into ultra-low-altitude UAVs,

low-altitude UAVs, medium-altitude UAVs, high-altitude UAVs, and ultra-high-
altitude UAVs.
• Currently, UAVs are mainly controlled manually by using a remote control
system. Data is transmitted between the remote control system and an UAV
through Wi-Fi or Bluetooth, allowing for a very limited communication distance.
Let's take Wi-Fi as an example. The LOS distance typically ranges from 300 m to
500 m (and reaches over 1 km in certain conditions). Currently, the maximum
distance is 7 km (DJI-provided UAVs that use the Ocusync image transmission
technology). UAVs that fly long distances may fail to communicate with the
remote control system and even crash.

• Networked UAVs are connected and controlled through cellular communication

networks. More simply, cellular base stations with wider coverage are used to
connect UAVs, providing more flexible and reliable communication for UAVs than
Wi-Fi or Bluetooth.
• In 2019, there were over 100 UAV enterprises and over 1400 enterprises in the
UAV ecosystem in China.

• The UAV industry chain in China has been fast maturing since 2014.
• 5G networks feature the following benefits: low-altitude network coverage and
low-latency operations

• More secure, large bandwidth, and multiple connections (4G networks cannot
connect numerous UAVs.)

• Inspection case:

▫ Traditional river course inspection: a maximum of 20 km/day/person; UAV-

based inspection: 600 km/day/UAV, 30-fold efficiency (Zhejiang, China)

• Pesticide spraying case:

▫ Mercury 1 UAV is designed with the 25 kg load-bearing capability. Its

standard loading capability is 20 kg, which is two to three times that of
mainstream models. (Source: Wuxi Hanhe Aviation, China's first pesticide
spraying UAV enterprise)
• Air-ground converged cellular communication:

▫ At an altitude below 300 m above ground (4G)

▫ At an altitude of 300–1000 m above ground (5G full coverage)

• Dedicated air-to-air cellular communication:

▫ Specific air routes at an altitude below 10,000 m above ground

• Currently, the mobile cellular network can meet the requirements of UAV
industry applications in most scenarios at an altitude of below 120 m above
ground, and the requirements to ensure safe flight services for UAVs in most
areas at an altitude below 300 m above ground.
• There are two types of coverage problems: poor wanted signal strength (such as
weak coverage and no coverage), and low SINR due to strong interference signal
despite adequate wanted signal strength.

• In most scenarios where low-altitude UAVs (< 300 m) are used, coverage
problems are mainly caused by interference.

• Massive MIMO provides the vertical beam capability to enhance low-altitude

coverage. With this technology, the number of horizontal and vertical TX
channels is increased to enable the horizontal and vertical beams to accurately
point to UEs. Narrower beams help prevent interference and improve the UEs'
signal-to-noise ratio (SNR). In addition, spatial multiplexing can be performed for
more UEs to improve the downlink capacity.
• EPC: performs session, access management, policy control, and user plane
functions.
• Remote control process: The remote control process continues until the target is
achieved.

▫ The controlled party needs to send status information to the controller

through a communication network based on remote sensing.

▫ The controller then analyzes the received information, makes a decision,

and sends a corresponding instruction to the controlled party through the
communication network.

▫ The controlled party executes the instruction to implement remote control.

• The communication network is mainly used to transmit status information and

instructions and ensures the accuracy and reliability of the transmission.
• Fixed line: low-speed sightseeing buses and JD express vehicles, etc.

• Harsh environment: mining machines in mining areas, remote driving of logistics

vehicles, gantry cranes at ports, mobile cranes, etc.
• Insufficient medical resources and service capabilities in medical institutions:
remote B-mode ultrasonic examination and remote surgery (for Parkinson's
disease)

• Infectious disease: remote consultation robot and remote medical protection

(anti-epidemic robot, disinfecting robot, and unmanned delivery van)
1: Laser rangefinder

2: 3D scanner

3: Vehicle-mounted camera
• Due to the insufficient development of factors such as algorithms, sensors,
infrastructures, and the network environment, full-scenario autonomous driving
cannot be implemented currently. Therefore, the implementation of autonomous
driving will be mainly distributed based on application scenarios. From the low-
speed scenario to the restricted scenario to the complex urban road scenario, an
increasingly high requirement is being imposed on the autonomous driving
algorithm and sensor combination. The implementation time of autonomous
driving also varies with the scenario complexity.
• To achieve complete autonomous driving, driving mileage tests of billions of
kilometers are required. During these tests, many traffic incidents may occur.
Here, we use the case of one of Tesla's vehicles colliding into another vehicle to
see the defects of intelligent single-vehicle braking.
• On a highway in northern Florida, Tesla Model S, which was operating in
autopilot mode, failed to detect an overturned large tractor trailer and collided
into the trailer without implementing any deceleration measures. The vehicle
passed under the trailer, ripping off its roof and killing the driver, and finally
stopped hundreds of feet from the collision site. The police officers found a
portable DVD player in the vehicle's wreckage, which is in line with the trailer
driver's report that the Model S driver was not watching the road before the
incident. According to current data, Joshua Brown, the driver of the Tesla vehicle,
used the Autopilot function of the vehicle and watched a movie while driving on
the highway, but the vehicle passed under the large trailer that was turning left.
Due to various reasons, the driver and the Autopilot system of Model S failed to
detect the large trailer turning left. Image recognition had blind spots, and the
database samples of mmWave radar sensors were insufficient.

▫ When the left-turning trailer entered the detection area of Model S's
camera, the long-focus camera used by the Tesla Autopilot system captured
only part of the trailer body instead of the entire trailer. Due to the white
color and high chassis of the trailer, the image recognition system of Model
S misjudged the trailer chassis as a white cloud in the sky and did not send
an alarm to the autonomous driving system. That is, the visual sense of the
vehicle was deceived.

▫ The front mmWave radar of the vehicle had an accurate detection range of
20 m. However, lacking data samples of the side faces of long vehicles, the
mmWave radar could not accurately detect the obstacles (such as some
vehicle boxes) in front and failed to alert the autonomous driving system,
that is, the vehicle's touch sense was also compromised. Ultrasonic radars,
the detection range of which is only 5 m, can only assist in automatic
parking and lane merging.

• According to the preceding figure, the Tesla vehicle's camera view was blocked
by the turning of the white trailer. In addition, the strong light distorted the view
of the camera. Strong light or a large area of obstacles may affect the image
recognition of the front-view camera. As a result, autonomous driving became
unreliable.
• radar, which should have been able to identify any other vehicle within 150 m
and provide warnings in advance. The traffic collision analysis shows that the
white trailer was moving in the opposite direction on a four-lane bidirectional
highway. The mmWave radar could not identify another vehicle in the opposite
direction, and therefore the trailer was not detected 150 m away. When the
trailer turned, the mmWave radar's view was blocked by the large-sized vehicle
carriage and unable to detect the number of vehicles in front. Due to the
echowaves of the signals sent by the internal antenna matrix, the mmWave radar
could not measure the number of other vehicles, and therefore did not display
related content.

• In the United States, a traffic collision with casualties occurred over an average of
94,000,000 miles (150,000,000 kilometers) of motor vehicles. However, the Tesla
vehicle collision occurred after the Autopilot function had been running for
130,000,000 miles (208,000,000 kilometers). This shows that the safety of
autonomous driving is high. In addition, due to underdeveloped technology, the
Tesla Autopilot did not have the key features of autonomous driving such as V2V
and V2I. If the two vehicles had the V2V technology for inter-vehicle
communication, this incident would have been easily avoided.
• What is the key control point of autonomous driving? Security will be an eternal
goal. To improve the reliability of autonomous driving decision-making, what are
the requirements of IoV for networks? 5G features high bandwidth and low
latency (real-time download of HD maps and sharing of vehicle environment
information) that can effectively guarantee the security and efficiency of high-
order autonomous driving.
• In addition to the increase of wireless spectrum bandwidth and air interface
performance, 5G mainly involves network architecture changes, E2E flexible
networking, and network cloudification. 5G can flexibly support edge computing
and network slicing to better achieve autonomous driving.

• Edge computing moves more data computing and storage from the core to the
edge, effectively reducing latency and ensuring higher reliability. In addition, 5G
supports E2E network slicing to ensure the resource and service priority of the
Internet of Vehicles (IoV), greatly improving the security of autonomous driving.

• The control and data planes of the 5GC are completely separated. NFV increases
the flexibility of network deployment, enabling distributed edge computing
deployment. Edge computing not only moves more data computing and storage
from the core to the edge, and but it is also deployed near the data source. Some
data does not need to be processed on the cloud through the network, reducing
the latency and network load and improving data security and privacy.

• Unlike best-effort data transmission on the Internet, network slicing can provide
consistent low-latency and high-speed service assurance, which is especially
critical to autonomous driving that has extremely high security requirements. For
example, when a vehicle travels in a congested network area, network slicing can
preferentially guarantee high-rate and low-latency performance of vehicle
communication.
• T, B, B
• Network slicing enables flexible network configuration, allowing for precise
monetization of network resources.

• Low-latency and high-throughput service assurance truly enhances user

experience and improves control sensitivity.

• Combining a mature industry and 5G is the focus of 5G industry development

and is gradually expanding to more industries.
• As traffic congestion increases in many big cities, improving traffic efficiency has
now become more pressing. In addition, the rapid increase in the number of
vehicles in cities, road user violations (such as drunk driving and pedestrian
violations), and fatigued driving have led to a high number of traffic accidents.

• As such, using new ICT technologies has become an important development

direction of the transportation industry to improve traffic efficiency, ensure safety,
and achieve green travel.
• Intelligent Transport System (ITS) is a system that effectively integrates
information, communications, sensing, control, and computer technologies into a
transportation management system. It enables precise, efficient, and
comprehensive transportation and management in real time.

• It improves transportation efficiency and the road network capacity, while

reducing traffic congestion, accidents, energy consumption, and pollution through
close collaboration among people, vehicles, and roads.

• Objectives: safe, unobstructed, convenient, efficient, and eco-friendly, promoting

function transformation and structure expansion of cities

▫ Comprehensive supervision of 3D transportation

▫ Visualized traffic operation and scheduling

▫ Efficient emergency handling

▫ Efficient and convenient transportation management

▫ Door-to-door traffic information services

▫ Efficient and accurate decision analysis

▫ Integrated public transportation

▫ IoV with vehicle-road synergy and autonomous driving

• The Electronic toll collection (ETC) system is an automated traffic system for
information exchange through wireless communications (Dedicated Short Range
Communications, DSRC). It mainly comprises the automated vehicle identification
(AVI) system, information base management system, and corresponding auxiliary
devices. An on-board unit (OBU), also called an electronic label or transponder, is
usually mounted on the windshield of a vehicle, similar to a card reader. It stores
the identity information of a vehicle, which is used when a vehicle passes through
a toll station.
• In addition, the AVI system includes hardware devices such as a Road Side Unit
(RSU) installed at a toll station and a loop sensor buried under lanes. The
information base management system is the "brain" of the ETC system: it stores
a large amount of information about registered vehicles and users, which
matches the information in OBUs to accurately verify identities. Auxiliary devices
include electronic barrier gates, an automatic tolling system, alarm sirens, and
traffic lights.
• Upon detection of a vehicle arriving at the toll station, the loop sensor under the
lane will transmit a signal to the RSU. The RSU then sends an inquiry signal to
the vehicle, and the OBU transmits back vehicle information, such as the vehicle
registration plate and vehicle type. The information base identifies the vehicle by
matching the received information with the data registered in the library. If it
matches, the electronic barrier gate will raise, allowing the vehicle pass, and the
RSU will interact with the IC card to charge the user. If it does not, an alarm will
be triggered until the vehicle exits the range of the loop sensor. It seems to be a
complex process. However, the system response time is short. Drivers do not have
to stop, greatly reducing traffic delays.
• Smart car is a comprehensive system that integrates environmental awareness,
planning and decision-making, and multiple levels of assisted driving. It uses
technologies such as computing, modern sensing, information convergence,
communication, AI, and automatic control.

• Advanced Driver Assistance Systems (ADAS) cannot work effectively in poor

weather conditions, such as rain, snow, and fog, and non-line-of-sight (NLOS)
scenarios. Therefore, it cannot ensure vehicle safety.
• The smart transportation platform provides road digitalization and vehicle-road
synergy to enable transportation operations.

• Smart transportation evolves from single-scenario management to multi-scenario

smart transportation services.
• The V2X interaction modes are as follows.

▫ Vehicle-to-vehicle (V2V) enables vehicles to communicate with each other

using in-vehicle devices.

▫ Vehicle-to-pedestrian (V2P) enables vulnerable road users (such as

pedestrians and cyclists) to use their devices (such as mobile phones and
laptops) to communicate with in-vehicle devices.

▫ Vehicle-to-infrastructure (V2I) enables in-vehicle devices to communicate

with roadside infrastructure (such as traffic lights, traffic cameras, and
RSUs).

▫ Vehicle-to-network (V2N) enables in-vehicle devices to connect to the cloud

platform using the access or core networks.
• Smart connected automobiles are an overwhelming trend and IoV/vehicle-road
synergy is a necessary stage toward autonomous driving. Massive information
exchanges of V2X/X2V are a key support for autonomous driving.
• IoV helps promote the innovation and development of automobile industry and
create new models and services. IoV also boosts the innovation and application of
assisted and autonomous driving.
• Full automation is the highest level of IoV. According to the standards of
autonomous driving defined by SAE International, smart vehicles, networks, and
infrastructure are critical for IoV in terms of in-vehicle information services and
smart transportation services. L0 is traditional driving; L1 and L2 are assisted
driving; L3, L4, and L5 are autonomous driving. L5 is also referred to as
"unmanned driving".
• Currently, autonomous driving vehicles are mainly equipped with video cameras,
radar sensors, and laser range finders. In the future, a higher level of
autonomous driving will be implemented based on IoV technologies.
▫ Level 0 means that drivers need to fully drive the vehicle without any
assisted driving systems. The automated system only issues warnings.
▫ Level 1 refers to assisted driving. Automated systems can control either
vehicle speed or steering, while drivers must be ready to control the vehicle
themselves at any time (such as fixed-speed cruise and ACC).
▫ Level 2 is partial automation allowing automated systems take full control
of the vehicle's speed and steering. Drivers must monitor the driving at all
times (for example, lane keeping).
▫ Level 3 ("hands off") refers to conditional automation. Drivers need to
monitor the system and be prepared to intervene when necessary.
▫ Level 4 ("eyes off") refers to high automation, which eliminates the need
for manual intervention in some predefined scenarios.
▫ Level 5 ("driverless") refers to full automation, in which no human
intervention is required at all.
• With multiple connection applications, C-V2X applies to complex security
applications. C-V2X has made a late surge, and the industry chain has been
gradually optimized to support more complex security applications, meeting the
requirements for low latency, high reliability, and large bandwidth.

• IEEE 802.11p (also called WAVE, wireless access in the vehicular environment) is
a communication protocol extended from IEEE 802.11. This protocol is mainly
used in the wireless communication of vehicle electronic devices. It defines the
enhancements to IEEE 802.11 required to support Intelligent Transportation
Systems (ITS) applications. This includes data exchange between high-speed
vehicles and between the vehicles and the roadside infrastructure that works in
the licensed ITS band of 5.9 GHz (5.85–5.925 GHz). IEEE 802.11p is applied to
vehicle telematics (or dedicated short range communication, DSRC). It is used as
the basis for WAVE/DSRC, and IEEE 1609 is a higher layer standard.
• By September 2016, 3GPP had formulated the first C-V2X technology standard
and incorporated it into LTE Release 14 at the 3GPP RAN meeting. C-V2X mainly
focuses on vehicle-to-vehicle (V2V) communication, which refers to device-to-
device (D2D) direct communication based on LTE Release 12 and 13. A new D2D
interface (designated as PC5) was introduced, and now as part of the 3GPP V2V
WI, is mainly used to implement C-V2X communication, especially for high-speed
(up to 250 km/h) and high-density (thousands of nodes) applications. Currently,
the specifications of LTE-V2X laid out in 3GPP Release 14, including the
application, network, and access layers, have been formulated.
• In December 2019, the Federal Communications Commission (FCC) voted
unanimously to advance a proposal which enabled the deployment of C-V2X in
the 20 MHz channel located at the 5.905–5.925 GHz band, and asked whether to
designate the 10 MHz (5.895–5.905 GHz) for C-V2X. The proposal would
reallocate large portions of the 5.9 GHz band, dedicating spectrum to both
unlicensed and C-V2X technologies. In the past two decades, 75 MHz of the 5.9
GHz band has been designated to DSRC, but FCC is looking to revise the rules,
noting that progress on DSRC has been stagnant for multiple years. In April 2019,
Toyota paused its deployment of the DSRC V2X communication technology,
severely hindering the development of this technology.
• Automatic Vehicle Identification (AVI) uses Radio-frequency identification (RFID)
to automatically identify goods in the production line, customs clearance vehicles,
and containers.
• Relying only on sensors, ADAS cannot accurately detect road conditions at
anywhere and anytime. C-V2X however can expand the sensing scope and
provide reliable and intelligent decision-making and collaborative control based
on cloud computing and big data platforms to ensure safety.

• 3GPP TS 22.185: [R-5.2.4-001] The E-UTRAN shall be capable of supporting a

communication range sufficient to give the driver(s) ample response time (e.g.
4s).
• Compared with C-V2X, DSRC has fully-developed standards and solid network
stability. However, as a newcomer, C-V2X is gradually overtaking DSRC. In terms of
availability, with self-networking, DSRC does not depend on the network
infrastructure (such as its security management and Internet access), providing high
stability. Therefore, the system cannot be interrupted due to transmission bottlenecks
or faults on a single node.
• Advantage 1: C-V2X supports 5G-oriented evolution.
▫ C-V2X has a clear evolution roadmap, while IEEE 802.11p does not.
• Advantage 2: C-V2X supports V2P services.
▫ The IEEE 802.11p standard is not supported by smartphones, making it
impossible to support V2P services. In contrast, smartphones can be integrated
with C-V2X (using LTE/5G chipsets) to support V2P services (Uu or PC5).
• Advantage 3: C-V2X supports more business models.
▫ C-V2X supports both PC5 and Uu interfaces and various services such as
infotainment, telematics, traffic safety and efficiency, dynamic maps, and big
data analysis.
• Strength 4: The chipsets of C-V2X are simpler.
▫ A single chipset can support both short-distance (PC5) and long-distance
communications (Uu) simultaneously, while IEEE 802.11p requires dual chipsets.
• Advantage 5: C-V2X supports larger economies of scale.
▫ The C-V2X provides clearer network deployment and reduces chipset costs for
fast improvement of the penetration rate (with the support of vehicles for
4G/5G communications).
• Founded in September 2016, 5GAA consists of a large number of members,
including eight founding members: Audi AG, BMW Group, Daimler AG, Ericsson,
Huawei, Intel, Nokia, and Qualcomm.
• Since its inception, 5GAA has rapidly expanded to include key players with a
global footprint in the automotive, technology and ICT industries. This includes
automotive manufacturers, tier-1 suppliers, chipset/communication system
providers, mobile operators, and infrastructure vendors. More than 130
companies have now joined 5GAA.

• For the latest information, visit the official website: https://5gaa.org/

• 3GPP specifications define the basic applications of C-V2X, including 27 basic
functions. For details, see 3GPP TR 22.885.

• IoV services can be applied in many scenarios, including safety, entertainment,

connection management, and smart driving.

• According to a survey from Accenture, 70% of consumers are willing to pay more
for their favorite IoV services (information, safety, and entertainment).
• The 3GPP specifications also define four advanced application scenarios of C-V2X.
For details, see 3GPP TR22.886.

▫ Vehicle platooning enables vehicles to dynamically form an ordered group.

All vehicles in the platoon receive information from the leading vehicle to
perform platooning operations. The information enables vehicles to travel
cooperatively in a very short distance and in a same direction.

▫ Cooperative driving enables each vehicle and/or Road Side Unit (RSU) to
share data obtained from sensors with vehicles within their proximity,
enabling vehicles to coordinate their trajectories. In addition, each vehicle
can also share its driving plan with vehicles within their proximity.

▫ Information sharing enables the exchange of information gathered through

local sensors or live videos. Devices of pedestrians and V2X application
servers can enhance the perception of their environment beyond what the
vehicle sensors can detect and have a more holistic view of the local
situation.

▫ Remote driving enables a driver to drive a vehicle remotely due to not

being able to drive themselves or dangerous environments. In cases in
which variation is limited and routes are predictable, such as public
transportation, high reliability, and low latency are the main requirements.
• With LTE extending to enable new services in vertical industries, 3GPP has
developed corresponding standards for enhanced V2X communications. Currently,
C-V2X standardization is implemented in three phases: 3GPP Release 14 that
supports LTE-V2X was officially released in 2017; 3GPP Release 15 that supports
LTE enhanced V2X (LTE-eV2X) was frozen in June 2018; 3GPP Release 16 and
beyond that support 5G-V2X were kicked off in June 2018, and would further
complement LTE-V2X and LTE-eV2X.

• C-V2X evolution

▫ In 2016, 3GPP Release 12 and Release 13 defined device-to-device (D2D)

and Proximity Services (ProSe).

▫ In 2017, 3GPP Release 14 defined the PC5 interface based on D2D and
ProSe.

▫ 3GPP Release 15 released LTE-eV2X which meets the high requirements on

services.

▫ 3GPP Release 16 released 5G-V2X which meets the requirements for high
density, vehicle platooning, and collaborative lane switching, providing a
maximum bandwidth of 10 Gbps with a latency ranging from 1 to 5 ms.
• 3GPP defines the spectrum for NR-V2X, namely the frequency band n47 (5855–
5925 MHz) which is used for the PC5 interface.
• LTE-V2X is currently the mainstream for existing V2X networks, utilizing the Uu
and PC5 interfaces. It is applicable to 27 scenarios laid out in 3GPP TR 22.885,
including proactive safety, traffic efficiency, and infotainment.
• LTE-eV2X aims to be compatible with LTE-V2X, but outperforms V2X in reliability,
data rate, and latency in direct mode to meet the requirements for some
advanced V2X services. It is designed to improve PC5 performance, adopting the
same resource pool design concept and resource allocation format as LTE-V2X.
Because of this, it can prevent resource collisions and interference. Enhanced
technologies in LTE-eV2X mainly include carrier aggregation, high-order
modulation, transmit diversity, low latency, and resource pool sharing.

• 5G-V2X is applicable to 25 scenarios related to autonomous driving as laid out in

3GPP TR 22.886, including platooning, advanced driving, sensor information
exchange, and remote driving.

• In the future, LTE-eV2X will co-exist with 5G-V2X.

• D2D refers to direct communication between two devices, and includes
Bluetooth, Wi-Fi Direct, and so on.

• D2D aims to directly establish a path between devices without needing of a

medium; however, it involves base stations at the least in practice.

• D2D is not a new concept introduced in 5G. The LTE D2D concept has existed
since the introduction of Proximity-based Services (ProSe) in LTE R12. However,
D2D technology was not applied during the 4G era; instead, devices
communicated through a complete network path.

• ProSe often refers to D2D in protocols.

• Sidelink (SL) refers to a link used for D2D, that is, a link used for direct
communication between devices. It corresponds to UL and DL in current
communication systems. In protocols, sidelink often refers to D2D. Unlike uplink
and downlink, sidelink is a new link introduced to support direct communication
between devices. It was first introduced in the D2D application scenarios, and
later extended and enhanced in V2X based on the original protocol.
• Benefits of D2D communication: reduces the load of cellular networks and
battery consumption of mobile devices, and increases bit rates and robustness of
network infrastructure
• In V2V communication scenarios: D2D communication can be used to warn
drivers when a near vehicle moving at a high speed, changes lanes, decelerates,
or performs other operations.
• Advantages of D2D:

• Improves the efficiency of radio spectrum resources

▫ In D2D communication, user data is directly transmitted between devices

without routing through a cellular network, resulting in link gains.

▫ Resources can be reused between D2D users and between D2D and cellular
networks, generating resource reuse gains. The link and resource reuse
gains improve the efficiency of radio spectrum resources and increase
network throughput.

• Improves user experience

▫ A new link mode is provided, enabling near-distance data sharing, as well

as the small-scale social and business activities of users in adjacent areas.

• Extends communication applications

▫ Traditional wireless communication networks have high requirements on

infrastructure. Damage to core network facilities or access network devices
may lead to breakdowns in communication systems.

▫ When wireless communication infrastructure is damaged or devices are in

blind coverage areas of wireless networks, the devices can implement E2E
communication or even access to cellular networks by using D2D.
• C-V2X has two communication modes based on different interfaces: V2X-Direct
and V2X-Cellular. V2X-Direct uses the PC5 interface and the IoV dedicated
frequency band (for example, 5.9 GHz) to implement direct communication
between vehicles, vehicles and roads, as well as vehicles and pedestrians, with
low latency and high moving speeds. V2X-Cellular uses the Uu interface of the
cellular network to forward data, and the cellular network frequency band (for
example, 1.8 GHz).

• Distributed mode: The PC5 interface is developed based on the D2D ProSe in the
3GPP R12/R13 (2016), implementing high-speed and high-density
communication. In V2V/V2I scenarios, adjacent devices can also directly
communicate with each other without needing LTE network coverage.

• Centralized mode: The LTE-Uu interface is used to transmit voice and video data
to a different node through the V2X server and LTE base station network.
• QCI: QoS Class Identifier.

• SPS: Semi-Persistent Scheduling.

• SPS is a scheduling mode introduced in 3GPP specifications. When system

resources (uplink and downlink) are allocated or specified only once through the
PDCCH, the same time-frequency resources can be periodically used.
• V2V: The consensus in the industry is that the PC5 interface must be used to carry
V2V services. Automotive enterprises hope that V2V services can be provided
through the PC5 interface in areas without cellular network coverage, improving
traffic safety and efficiency.

• V2P: The industry believes that the PC5 interface can quickly penetrate into
common smartphones by utilizing the upgraded LTE chips, thereby providing V2P
services.

• V2I: Currently, the industry does not have a specific carrying technology, and
both Uu and PC5 interfaces serve as network communication channels. The
following factors affect actual deployment:

▫ Industry requirements: Public security and transportation departments have

requirements on the use of the 5.9 GHz private network to carry V2I
applications (vehicle control and scheduling by roadside transportation
infrastructure). As such, the PC5 interface needs to carry V2I services.
▫ Network coverage: In areas without public network coverage (such as
remote, dangerous roads), RSUs (PC5) can be easily and quickly deployed
to provide V2I services.

▫ Transmission efficiency: The Uu interface mainly uses unicast transmission

(due to limited commercial applications, network construction is required to
deploy LTE broadcast/multicast). PC5 uses broadcast/multicast
transmission. The Uu unicast resource overhead increases linearly with the
number of vehicles. In scenarios with high vehicle density, the network
capacity may not meet requirements.

▫ Inter-operator support: When V2I services are carried over the Uu interface,
only users in the network that sends the services can enjoy them. When V2I
services are carried by the PC5 interface, all users can enjoy them. This
improves the radio frequency utilization efficiency.

▫ Communication latency: Compared with base stations (Uu), RSUs (PC5)

carry the V2I service with a shorter latency. However, the latency of base
stations with mobile edge computing (MEC) can be reduced significantly,
depending on the network-side availability of MEC nodes.
• V1: The reference point between the V2X Application in the UE and in the V2X
Application Server. This reference point is not within the scope of 3GPP
specifications.
• V2: The reference point between the V2X Application Server and the V2X Control
Function in the operator's network. The V2X Application Server may connect to
V2X Control Functions belonging to multiple PLMNs.

• V3: The reference point between the UE and V2X Control Function in the
operator's network. It is based on the service authorization and provisioning part
of the PC3 reference point defined in clause 5.2 of TS 23.303, and is applicable to
both PC5 and LTE-Uu-based V2X communication and optionally eMBMS and
LTE-Uu-based V2X communication.

• V4: The reference point between the HSS and V2X Control Function in the
operator's network.

• V5: The reference point between the V2X Applications in the UEs. This reference
point is not specified in the 3GPP specification.
• V6: The reference point between the V2X Control Function in the HPLMN and
V2X Control Function in the VPLMN.

• For details about each reference point in the V2X network architecture, see 3GPP
TS 23.285.
• V2X scenarios:

▫ Scenarios based on cellular network coverage (served by E-UTRAN):

Services can be provided through the Uu interface, implementing large-
bandwidth and large-coverage communication, or through the PC5
interface, implementing low-latency and high-reliability direct
communication between vehicles and nodes in the surrounding
environment.

▫ Scenarios independent of the cellular network (not served by E-UTRAN): In

areas without deployed networks, IoV road services are provided through
the PC5 interface, meeting requirements for driving safety.
• User Centric No Cell Radio Access (UCNC)

▫ Virtual cell: Transmission reception points (TRPs) surrounding the UE form

a virtual cell that moves with the UE.

▫ Hyper cell: The continuous coverage areas of multiple independent TRPs

can be combined into one hyper cell. As each TRP uses the same PCI and
CGI, UEs cannot perceive the existence of multiple TRPs while moving
among them. As such, handovers do not occur and user experience is
improved in areas between TRPs.

• Dual connectivity: In 3GPP R16, NR-V2X supports both NSA and SA. The NR-V2X
architecture is classified into standalone and multi-RAT dual connectivity (MR-
DC), including six deployment modes.
• Time difference of arrival (TDOA)

• Observed time difference of arrival (OTDOA)

• Angle of arrival (AoA): indicates the angle between the wave ray and a certain
direction (horizontal plane or normal line of the horizontal plane) when the wave
reaches the observation point. The transmitter uses a single antenna to send a
reference signal, which the receiver uses multiple antennas to receive. Phase
difference is generated as the distance varies between antennas and the
transmitter. The angle relationship is calculated based on the phase difference
and distance between antennas.

• NR V2X supports two resource allocation solutions:

▫ Mode 1: The base station schedules sidelink resources for UEs to perform
sidelink transmission.

▫ Mode 2: UEs determine the sidelink resources configured (pre-configured)

by the base station/network.
• NR-V2X is applicable to more service scenarios based on multicast and unicast
communication scenarios. In addition, in multicast and unicast scenarios, hybrid
automatic repeat request (HARQ) is supported to improve communication
reliability.

• The NR sidelink consists of the physical sidelink control channel (PSCCH),

physical sidelink shared channel (PSSCH), physical sidelink broadcast channel
(PSBCH), and physical sidelink feedback channel (PSFCH). The first three
channels are introduced in LTE-V2X, while PSFCH is introduced in NR-V2X to
support HARQ transmission.

• NR-V2X-based sidelink supports unicast, multicast, and broadcast. Devices may

participate in multiple unicast, multicast, and broadcast communications
simultaneously to transmit different information.

▫ Unicast: implements reliable transmission with 3 ms E2E latency.

▫ Multicast: implements periodic transmission within the limited transmission

range. The 5G sidelink can dynamically form a multicast group for vehicles
that use the same service within a certain range.

▫ Broadcast: implements periodic transmission with a coverage area that is as

large as possible.
• The control and data planes of the 5G core network are completely separated.
NFV facilitates more flexible network deployment, which in turn enables
deployment of distributed edge computing. Edge computing moves data
computing and storage operations from the core to edge near the data source.
Some data does not need to be processed on the cloud through the network,
reducing latency and network load, while also improving data security and
privacy.

• Unlike best-effort data transmission on the Internet, network slicing can provide
consistent low-latency and high-rate services, which are especially critical for
autonomous driving with extremely high security requirements. For example,
when a vehicle travels in a congested network area, the network slicing can still
preferentially ensure a high rate and low latency for its communication.

• The core network functions such as mobility management and user plane
functions are moved to the edge, significantly reducing the processing latency.
• MEC cloud computing moves the local cloud platform to the network edge,
providing large-bandwidth and low-latency services for mobile devices. MEC can
be flexibly deployed at different locations on the network based on requirements
for service deployment and coverage. In IoV applications, the main scenarios for
the MEC system are intersections, industrial campuses, and highways. Vehicle
speed guidance, traffic light control, and high-precision map download can be
implemented in intersections; remote driving, remote monitoring, and
autonomous parking can be implemented in campuses; and platooning and
autonomous driving can be implemented on highways.
• Real-time kinematic (RTK) is also referred to as the carrier phase differential
technique. It is a technology that improves the precision of GNSS systems,
providing up to centimeter-level accuracy. A GNSS monitoring receiver is installed
in a precise location (Reference station), and a distance correction between the
reference station and GNSS satellite is calculated. Differential signals are
classified into position differential signals and distance differential signals. The
RTK can achieve centimeter-level positioning.

• Progress of cellular-aided RTK high-precision positioning in the specifications

(completed in R15): 36.305 (overall), 36.331 (Uu interface), and 36.355 (LPPa
interface)

• LTE Positioning Protocol (LPP) is used for the UE and the positioning computing
center, whereas LTE Positioning Protocol Annex (LPPa) is used for the base
station and the positioning computing center.

• Positioning reference signal (PRS)

• Interconnection between newly-developed and less-advanced vehicles, ensuring
basic security service communication.
• By building a cooperative intelligent transportation system (C-ITS) based on the
C-V2X, we can reduce vehicle-road information asymmetry, enable bi-directional
interaction between vehicles and roads to form the standard configuration, and
improve traffic safety and efficiency. With the evolution to cooperative
autonomous driving, the cost of a single autonomous driving vehicle can be
further reduced. Higher driving safety and traffic efficiency enable collaborative
autonomous driving.

▫ Real-time vehicle/road information interaction and sensing

▫ Global traffic data analysis and traffic flow guidance

▫ Lower technical requirements on intelligent vehicle in autonomous driving

▫ Globally unified standards, supporting the evolution to 5G

• BeiDou Navigation Satellite System (BDS)
• Telematics box (T-Box) is also called IoV control unit. (Telematics is a
portmanteau of telecommunications and informatics) Simply put, it is a
computer (embedded) installed on a vehicle to control and track the vehicle's
state. The T-Box is factory-installed, meaning the vehicle already has a
networked controller before reaching the market.
• On-board diagnostics (OBD) is an on-board automatic diagnosis system. It is a
device installed after the vehicle is launched to obtain real-time vehicle data
through the OBD port. The OBD box is a customer-installed product. An extra
small box is inserted into the vehicle's OBD port to connect to the cloud.
• 5G base station: small coverage area for 5G IoV services

• LTE-V2X: wide coverage area for Uu-based data forwarding, PC5-based resource
allocation in mode 3 and pre-configuration modification in mode 4
• Roadside unit (RSU): small coverage area for PC5 communication and as a bridge
to V2X infrastructure for V2X data sending and transmission
• MEC server: Uu-based low-latency services and local processing of V2X data

• IoV service platform: network management (including service and connection

management, with general IoV service analysis components); vehicle-vehicle and
vehicle-cloud synergy; and network openness (service expansion, access from
third-party apps)

• Suggestions on IoV deployment: 1. Policy support: The government releases

supporting policies to attract more investment, to increase IoV penetration rates,
and to promote corresponding development. 2. Standard leading: Promote
organizations such as IMT-2020/5GAA/C-ITS to lead application standardization
and streamline industry chain standards and application scenarios. 3. Application
encouragement: Expand city-level demonstration pilots in more cities, improve
the value of V2X technologies. Expand the application to road networks and build
smart highways and roads to reduce traffic accidents.
• Intelligent device modules (CPE to provide 5G signals) are installed on electric
self-operated mining trucks. Vehicle-mounted device modules obtain vehicle and
battery data through the CAN bus and receive GPS positioning information (with
modules provided by CHCNAV). Then, the data is transmitted to the cloud
computing platform (including the central control and security computing units)
through the 5G network. The platform monitors the working conditions in real
time to effectively control the vehicle running status, send vehicle warning
information, improve vehicle running safety, and provide core data support.
• A typical power system comprises power generation, transmission,
transformation, distribution, and consumption.
• Meeting the requirements of intelligent power distribution and consumption in
the last few kilometers is currently a major challenge.
• Many power grid devices still use GPRS due to safety concerns.
• Smart grid leverages integrated and high-speed bidirectional communication
networks, using a variety of advanced technologies to build reliable, secure,
economical, efficient, and eco-friendly power grids. Such advanced technologies
include those for sensing and measurement, devices, control methods, and decision
support systems.
• With the rapid development of new technologies in various fields, smart grids are
facing new challenges and opportunities.
▫ New energy: To address global warming and achieve sustainable development,
the industry urgently needs to adopt renewable energy in the production of
electricity. However, using multiple sources of renewable energy will bring new
challenges to grid operation and management. For example, power balance
and operation control are more difficult due to the intermittent and random
nature of renewable energy. In addition, the deep penetration of distributed
energy resources (DERs) means that the distribution network has to change
from passive one-way power flow to active two-way power flow.
▫ New users: Electric vehicles are becoming more and more popular, meaning
that there is increasing demand for charging capacity. New modes for
managing power consumption (for example, balanced power supply) are
therefore required to meet this demand. For example, instead of charging
vehicles only when they are connected to the power supply, users can charge
them interactively at any time.
▫ New requirements: New devices and scenarios pose higher requirements on
power quality. For example, some high-tech digital devices require an
uninterrupted power supply. In addition, grid operation requires high asset
utilization, ranging from higher utilization of equipment to lower capacity
ratios and line losses. This means that the load and power supply of the grid
need to be adjusted more accurately.
• Advanced metering infrastructure (AMI) is an architecture for automated,
bidirectional communication between power companies and smart meters with
IP addresses.
• Wireless private networks for power grids are still in the early stage of
construction and therefore do not provide wide coverage. Areas covered by
private networks use these networks to carry power services, whereas uncovered
areas use the public network to carry non-control power services. Such services
can gradually be migrated to private networks as private network coverage
increases.

• With the maturity and large-scale deployment of the eMBB network on 5G public
networks, the single-device bandwidth is greatly increased to carry non-key high-
bandwidth services, such as HD video and VR services. Because the 230 MHz
private network lacks sufficient frequency resources, it cannot adequately support
high-bandwidth services. Consequently, it is used to construct a full-coverage
power IoT infrastructure network.
• Based on the requirements of the power grid for wireless communication, typical
5G smart grid applications in the future will include smart automation of
distributed power distribution, millisecond-level precise load control, collection of
low-voltage power consumption data, and distributed power supply.

• The key requirements of smart automation of distributed power distribution for

communication networks include:

▫ Ultra-low latency: milliseconds

▫ High level of isolation: Automated power distribution is a service in the grid

production area and therefore must be completely isolated from services in
management areas.

▫ High reliability: 99.999%

• The key requirements of millisecond-level precise load control for communication

networks include:

▫ Ultra-low latency: milliseconds

▫ High level of isolation: Precise load control is a service in the grid

production area and therefore must be completely isolated from services in
management areas.

▫ High reliability: 99.999%

• The key requirements of low-voltage power consumption data collection for
communication networks include:

▫ Massive connectivity: tens of millions of devices

▫ High frequency and high concurrency: second-level to near-real-time data

reporting

• The key requirements of distributed power supply for communication networks

include:

▫ Massive connectivity: millions to tens of millions of devices

▫ Low latency: Distributed power supply management includes uplink data

collection and downlink control. The latency of downlink control flows
should be measured in seconds.

▫ High reliability: 99.999%

• In addition to the preceding two typical types of slices, the power industry may
also require eMBB and voice slices. The typical service scenario of the eMBB slice
is remote UAV inspection, and that of the voice slice is manual maintenance and
inspection.
• Based on the application scenarios of smart grids and the architecture of 5G
network slicing, slices for data collection, automated power distribution, and
precise load control are applied for 5G smart grids. Different slices meet the
technical specifications in various scenarios. Domain-based and E2E slice
management is implemented to meet service requirements.
• This pilot project involved a power supply company, a telecom company, and
Huawei. The test focused on the E2E latency of precise load control based on
eMBB defined in 3GPP Release 15.
• Test environment: 5G network in 3400 to 3500 MHz; 100 MHz bandwidth; 2 x
eMBB antennas; 0.5 ms air interface scheduling period. A 100 Mbps FTTH private
channel was used to connect the 5G core network and master station of the
electric power precision control system. Huawei 5G eMBB home broadband CPE
was used, and E2E network slicing was deployed on the 5G core network.
• System networking solution: Power slices are deployed on the 5G eMBB SA core
network, and 5G indoor distributed sites and macro sites are deployed on the
user side. A precise control master station is set up in the power supply company,
a 100 Mbps FTTH private channel is used to transmit data between the 5G core
network and master station, and IP RAN technology is used on the backhaul
network between the 5G base station and core network. To measure the latency
between load control devices and precise load control slave stations, the CTD-1
electric power communication tester is used.
• The average E2E latency measured during this test was 37 ms, including 4.5 ms
between the 5G core and the CPE and 0.5 ms between the 5G core and the
precise load control slave station. The remainder is between the control
command delivery by the slave station and the command processing by the load
control device.
• USD230 billion is expected to be invested into the smart healthcare market by
2025, and 5G will provide the connections required for smart healthcare.

• Remote diagnosis is a special application that particularly relies on 5G's low

latency and high QoS.

• A hospital in France provides remote B-mode ultrasonography and remote access

to medical consultation, reducing the cost of medical treatment. This remote B-
mode ultrasonic robot is ready for commercial use, representing a typical
application of force feedback and "tactile Internet". Force feedback increases the
accuracy of remote operations, mitigating patients' pain during health checkups.
The force feedback signal requires an E2E latency of 10 ms.

• The information system of hospitals is evolving from siloed toward digital,

intelligent, efficient, and collaborative industry interconnection that best
incorporates service values.
• 5G will bring new experiences in all scenarios and create new business value.

• The remote multimedia conference system expands the hospital boundary to

benefit more groups.
• 5G makes remote healthcare workshops, teaching, and surgery demonstrations
more convenient and intuitive.
• 5G provides better technical enablers for remote surgery, which is a crucial part
of remote healthcare.

• 5G wireless application scenarios in hospitals mainly include the following:

patient positioning, wireless infusion, wireless monitoring, robotic ward round,
emergency medical treatment, remote consultation, remote ultrasonography, and
remote surgery.
• 5G provides ultra-high bandwidth (10 Gbps), ultra-low latency (milliseconds),
and massive connections (1 million per km2).
• [Pain point]: There is a shortage of professional workers available to provide
emergency medical treatment. For example, city J has 13 districts and 501
ambulances but only 252 first aid specialists. In addition, general medical
personnel may provide unsuitable treatment without guidance.

• [Solution]: Real-time data (such as ambulance location, electrocardiograms,

ultrasound images, blood pressure, heart rate, oxygen saturation, and body
temperature) is synchronized to the 5G remote command center where doctors
can guide on-site treatment through real-time audio and video communications.

• [Value]: Medical resources are efficiently utilized to improve the overall

efficiency of pre-hospital emergency treatment.
• Remote healthcare not only fully utilizes medical resources, but also reduces
cross-infection during referrals and inspections. Remote consultation services
allow patients in designated hospitals in remote areas to obtain diagnosis and
treatment services from experienced doctors. This frees patients from time and
space restrictions, reduces direct contact between doctors and patients, and
significantly improves the efficiency of epidemic prevention and control.

• The remote healthcare system supports 1080p HD videos. Because medical

experts in different locations may need to share medical archives such as CT
images through auxiliary streams for diagnosis, HD videos are critical for remote
healthcare. In addition, medical carts equipped with cameras can take photos of
patients in wards. These images can also be uploaded for remote consultations.
• Education networks currently face multiple challenges:
▫ Resource sharing: Information systems (such as teaching, scientific research,
management, technical service, and life service) are siloed, hindering the
integration of services and processes.
▫ Insufficient capacity to bear new services: New services (such as 4K/8K live
classroom, AR/VR classroom, holographic education, 4K HD video, and mobile
patrol vehicles) pose higher requirements on network bandwidth.
▫ Data security: Data leakage may occur when sharing resources with parents or
between campuses. This is made worse by the aggregation of big data.
▫ High construction and O&M costs: Education information systems and multi-
network convergence are costly to build, operate, and maintain.
• 5G smart classroom uses 5G-ready devices and leverages the following advantages of
5G to deliver the optimal teaching experience for school users:
▫ With the unified network, schools do not need to deploy multiple networks.
▫ Ultra-high bandwidth ensures that interactive displays and devices for signal
transmission and processing in smart classrooms can perfectly reproduce 4K
images, and supports the upcoming 8K interactive devices.
▫ High speed and low latency ensure normal streaming in smart classrooms.
During distance teaching, school users at remote sites can access 4K or higher-
resolution images immediately.
▫ 5G smart classroom provides capacity to foster new application scenarios, such
as game-like courses, VR lab environments, HD 3D display, remote exam
monitoring, learning behavior tracking, smart lab systems, and smart teaching
systems.
• 5G can help build smart campuses and develop diversified education applications
based on service requirements and 5G features. 5G smart campus provides a
fully-connected private network that connects various smart devices and
educational equipment. It also delivers the education edge cloud that integrates
computing, storage, AI, and security capabilities to provide an application-
enabling platform with management and security capabilities.
• Network slicing is the basis of 5G-based private education networks. It constructs
multiple dedicated, virtual, isolated, and customized logical networks on a
physical network to carry education services, while meeting different
requirements of services on network capabilities (such as latency, bandwidth, and
connections). Data is shared across 4G, 5G, NB-IoT, and private networks,
preventing data silos between different networks. In addition, local transmission
and storage of private data of teachers, students, and parents ensures user data
security.
• 5G MEC provides basic capabilities such as massive device management, highly
reliable and low-latency networking, hierarchical QoS, real-time data computing
and cache acceleration, application container service, and network capability
openness. It also enables a multi-level edge computing system to provide real-
time, reliable, intelligent, and ubiquitous E2E services for smart education. 5G
MEC solutions can be applied in universities and other education scenarios. MEC
nodes are deployed on the base station side, base station aggregation side, or
core network edge side to provide multiple intelligent network connections and
high-bandwidth, low-latency network bearing. Open and reliable network
connections as well as computing and storage resources allow for flexible bearing
of multiple ecosystem services at the access edge.
• Classroom scenario (University X): The teacher wears AR glasses for onsite
teaching. AR courseware is displayed in the glasses and projected to the large
screen. The 360° panoramic camera collects images in real time.
• Communication network (E2E 5G network): Real-time images are transmitted to
the server equipment room through the 5G network for VR image rendering and
pushed to the remote access side.

• Remote learning (School X): Students wear 5G-enabled VR glasses for an

immersive experience while watching teaching content provided by the teacher
and AR images.
• One-click query

▫ Real-time communication between RFID cards and tags to monitor the

locations of assets

▫ Real-time query of asset trajectory by asset number and time using the 5G
network
• One-click counting

▫ Automatic asset registration and counting (hundreds of thousands of assets

counted within 30s)

▫ Improved employee efficiency and CIAG management

• Security incident alarm generation

▫ Valuable asset monitoring with 5G and IoT APs

▫ Automatic video recording and storage in the system

 Industrial Revolution 1.0 Industrial Revolution 2.0 Industrial Revolution 3.0 Industrial Revolution 4.0
Steam  Mechanization Electric Power  Scale-up Informatization  Automation IoT  Intelligence

Electronic and IT systems in

Electric power application, the automatic production for Cyber-physical
Steam engines in the labor division, and batch the 3rd industrial revolution systems for the
mechanical production for the production for the 2nd 4th industrial
1st industrial revolution industrial revolution revolution

Time

• We are embracing the fourth industrial revolution featuring intelligence. ICTs,

such as big data analysis, cloud computing, mobility, and IoT, have become the
foundation of the intelligent industrial revolution.

• While focusing on the enterprise infrastructure, intelligent products, and

customer platforms, ICTs deeply converge with vertical industries.

Government Finance Transportation Power Education Healthcare Manufacturing Agriculture

• The industrial Internet is mainly based on wired networks, but faces challenges in
expansion and management, such as long cabling duration, poor corrosion
resistance, difficult maintenance, high costs, and high monopoly.
• Wireless network advantages: wide coverage, high scalability, flexible networking,
and easy maintenance
• The main problems of Wi-Fi: poor coverage, instability, poor mobility, and limited
bandwidth and latency

▫ Poor coverage: multipath interference due to reflection and diffraction of

Wi-Fi signals

▫ Instability: interference due to the use of the public frequency band,

affecting the demodulation capability

▫ Poor mobility: long inter-AP handover latency (> 100 ms), causing
interruption when AGVs are moving
• Compared with other wireless communication systems, 5G provides lower
latency, higher rates, and better service experience, featuring ubiquitous
perception, connection, and intelligence. It is expected to become the cornerstone
of the future industrial Internet.

• There are more than 100,000 large- and medium-sized factories in China, leading
to a strong demand for cellular connections to replace Wi-Fi connections. 5G has
become a springboard for operators to develop smart factories.

45 Number of Existing Connections 43

40 Unit: 100 million Proportion of wireless connections:
2017: 6%
35
2022: 27%, breaking point
30 2026: 58%, surpassing point
25
25
18 Wired
20
15
10 5 Wireless
5
0
2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026
• 5G network with slicing enables applications that support massive connections
and high-precision positioning in logistics and warehousing areas.
• AGVs are the core equipment of the factory's smart logistics system. Wireless and
intelligent AGVs become their development trend.

▫ Dispatching quantity: The increasing number of AGVs creates a dispatching

bottleneck, requiring breakthroughs in application scenarios such as
logistics.

▫ Flexibility: Except for visual navigation, the working environment of AGVs

needs to be reconstructed in all navigation modes, leading to poor flexibility
and difficult deployment and reconstruction.

▫ Reliability: Wi-Fi signals are susceptible to interference, and the bandwidth

is insufficient for visual navigation. AGVs adopting laser navigation interfere
with each other during high-density operations.

▫ High costs: Each locally intelligent AGV that adopts visual navigation has a
high cost, which is not conducive to scaled application.

• 5G cloud-based AGVs improve the AGV dispatching capability.

▫ 5G with high bandwidth and low latency supports AGV control on the
cloud.

▫ 5G provides networking capabilities for scaled dispatching of AGVs.

▫ Centralized control considerably reduces the per-unit cost of AGVs.

• The visualization and interactive functions in AR devices ensure optimal efficiency
and functionality for models.
• Application of high bandwidth and AR in the manufacturing field: digital design
collaboration, assembly operation assistance, visualized display for sales, and
O&M guidance
▫ Digital design collaboration: For example, Realibox can convert the 3D CAD
design model into a real-time immersive experience. With the scenario
template and intuitive drag-and-drop tools provided by Realibox Studio,
users can define the real material, light, and scenario effects, as well as add
more complex interactions such as multi-design solution management and
switching for models, enabling designers to build interactive scenarios in
minutes.
▫ Assembly assistance: Assembly accounts for 40% to 50% of total workload
involved with developing high-end complex devices. AR provides the 1:1 3D
virtual model of real parts, enabling personnel to improve assembly
efficiency by dynamically displaying the annotations of parts.
▫ Visualization for sales: AR for car sales and Evergrande online house sales
▫ AR O&M guidance: Frontline operation personnel use AR devices to make
real-time bidirectional audio and video calls with experts in the remote
command center through 5G. The command center can provide technical
guidance for frontline O&M personnel while they collect on-site image
information.
 • Applications of 5G basic service capabilities in smart factories:

Quick Service
Basic Service Capability
Capability
Scenario Content
Large Low High Precise
Slicing MEC
Bandwidth Latency Reliability Positioning

Mobile panel (with ☆ ☆ ☆ ☆

safety control)
Machine
control Inter-machine control ☆ ☆ ☆ ☆ ☆

Motion control ☆ ☆ ☆ ☆ ☆ ☆

AGV dispatching and ☆ ☆ ☆ ☆ ☆

Mobile robot
robot control

Video inspection ☆ ☆ ☆

Remote on-site – AR ☆ ☆ ☆ ☆ ☆
assistance
Industrial AR
Remote on-site – ☆ ☆ ☆ ☆
and
complex VR assembly
surveillance
Predictive maintenance ☆ ☆ ☆ ☆

Production security ☆ ☆ ☆
behavior analysis

5G + large-scale data ☆ ☆ ☆
collection

Massive Energy consumption ☆ ☆

connections monitoring

Personnel and asset ☆ ☆

positioning
• Answer: ABCD