The block diagram of a digital optical receiver has three parts:
1. Front end
2. Linear channel
3. Data recovery section
Front End:
• Consists of a reverse-biased photodiode.
• The photodiode converts optical (light) data into electrical data.
• The output goes through a preamplifier.
Linear Channel:
• Includes a high-gain amplifier (Amp) and a low-pass filter (LPF).
• The LPF reduces noise but also cuts part of the signal.
• The LPF's bandwidth should be optimized for the best signal-to-noise ratio (SNR).
Data Recovery Section:
• Consists of a decision circuit and a clock-recovery circuit.
• Example: If the input bit pattern is '1011', some bits might be corrupted by noise.
• If the received current is above a threshold (e.g., 30 mA), the decision circuit interprets it
as '1'. Otherwise, it's '0'.
• Noise can cause errors if the current is below the threshold for a '1' bit.
• A clock-recovery circuit extracts timing information to sample the signal correctly.
Front End:
• The front end of a receiver includes a photodiode followed by a preamplifier.
• The optical signal is coupled to the photodiode using a method similar to that used for
optical transmitters.
• The photodiode converts the optical data into an electrical signal, which is then amplified
by the preamplifier for further processing.
Design Considerations:
• There's a balance between speed and sensitivity in the design.
• Using a large load resistor (RL) can increase the input voltage to the preamplifier, often
leading to a high-impedance front end.
• A large RL reduces thermal noise and improves receiver sensitivity but also limits
bandwidth.
Bandwidth Limitations:
• The receiver's bandwidth is limited by the slowest component.
• If the bandwidth (Δf) is much less than the bit rate, a high-impedance front end cannot be
used.
• An equalizer might be used to increase bandwidth by attenuating low-frequency
components more than high-frequency ones.
Transimpedance Front Ends:
• These provide high sensitivity and large bandwidth.
• They improve the dynamic range compared to high-impedance front ends.
• The load resistor is connected as a feedback resistor around an inverting amplifier,
reducing the effective input impedance and increasing bandwidth.
• These are often used in optical receivers due to their improved characteristics, but their
design requires addressing feedback loop stability.
Linear Channel:
• The linear channel in optical receivers includes a high-gain amplifier and a low-pass
filter.
• An equalizer may be added before the amplifier to correct for the limited bandwidth of
the front end.
• The amplifier's gain is automatically controlled to keep the output voltage level fixed,
regardless of the incoming optical power.
• The low-pass filter shapes the voltage pulse and reduces noise without causing much
intersymbol interference (ISI).
• Receiver noise is proportional to the receiver bandwidth and can be reduced by using a
low-pass filter with a bandwidth smaller than the bit rate.
• Other components in the receiver have a larger bandwidth than the low-pass filter, so the
overall receiver bandwidth is determined by the low-pass filter.
• If the filter's bandwidth is smaller than the bit rate, the electrical pulse spreads beyond its
allocated slot, potentially interfering with the detection of neighboring bits, causing ISI.
Data Recovery Section:
• The decision circuit and clock-recovery circuit are part of the data-recovery section in
optical receivers.
• The clock-recovery circuit isolates a specific frequency component from the received
signal to synchronize the decision process.
• For RZ (return-to-zero) format, this component is easily isolated. For NRZ (non-return-
to-zero) format, it's more challenging because the signal lacks this component.
Function:
• The decision circuit compares the output signal to a threshold level at specific sampling
times determined by the clock-recovery circuit.
• It decides if the signal is a '1' or '0' based on the threshold.
• The best sampling time is when the difference between '1' and '0' signals is maximum,
identified using an eye diagram. This diagram visually shows the optimal sampling time.
Error Probability:
• Due to noise, there's always a small chance of errors in identifying bits.
• Digital receivers are designed to keep this error probability very low.
• The eye diagram helps monitor receiver performance—if the "eye" closes, it indicates
poor performance.
Coherent Detection:
• A "coherent" optical transmission system can do "coherent detection," meaning it can
track the phase of an optical transmitter to extract phase and frequency information.
• A narrow-linewidth tunable laser acts as a local oscillator (LO), tuning its frequency
close to the received signal's frequency through an optical coherent mixer. This recovers
both amplitude and phase information from the optical carrier.
• "Intradyne" means the frequency difference between the LO and the received carrier is
small and within the receiver's bandwidth, but doesn't have to be zero, avoiding the need
for a complicated optical phase-locked loop.
Direct Detection:
• Used by 10Gb/s or lower-speed systems, direct detection receivers only respond to
changes in optical power and cannot extract phase or frequency information.
Advantages of Coherent Detection Over Direct Detection:
1. Improved Receiver Sensitivity: Much better at detecting weak signals.
2. Extracts More Information: Can extract amplitude, frequency, and phase information,
leading to higher capacity within the same bandwidth.
3. Compensates for Dispersion: Digital Signal Processing (DSP) can correct chromatic
and polarization mode dispersion, removing the need for optical dispersion compensators
and simplifying network design.
4. Better Signal-to-Noise Ratio (SNR): Using balanced detectors with high noise rejection
improves SNR.
Receiver Noise:
• Noise disrupts the transmitted signal in a fiber optic system, setting a lower limit on the
optical power needed for proper receiver operation.
Sources of Noise:
1. Light source noise.
2. Noise from the light interacting with the optical fiber.
3. Noise from the receiver.
Types of Receiver Noise:
1. Thermal Noise: Caused by random electron motion in a conductor. It comes from the
photodetector, load resistor, and amplifiers. Increasing the load resistor can reduce
thermal noise, but it also reduces the receiver's bandwidth.
2. Dark Current Noise: Results from dark current in the photodiode when there's no light.
It doesn't depend on the optical signal.
3. Quantum Noise: Caused by the random generation of electrons due to incident light.
This is a type of shot noise, which arises from current fluctuations because of discrete
charge carriers.
Key Points:
• Receiver noise can be signal-dependent (affected by optical power) or signal-independent
(not affected by optical power).
• Shot Noise: Fluctuations in current due to the discrete nature of charge carriers. Includes
dark current noise and quantum noise.
• Excess Noise: In Avalanche Photodiodes (APDs), an extra shot noise caused by the
random nature of the avalanche process.
Self-Evaluation Questions:
1. What are the main types of receiver noise?
2. What determines receiver sensitivity?
3. To reduce thermal noise, should the load resistor value be increased or decreased?
4. What are the two types of noise that show up as shot noise?
Answers to the Self-Evaluation Questions:
1. Main Types of Receiver Noise:
• Thermal Noise
• Dark Current Noise
• Quantum Noise
2. Main Factor That Determines Receiver Sensitivity:
• Noise is the main factor that limits receiver sensitivity.
3. For a Reduction in Thermal Noise:
• The value of the detector's load resistor should be increased.
4. Two Types of Noise That Manifest as Shot Noise:
• Dark Current Noise
• Quantum Noise
