A modular DTN simulator in the style of the ONE.

Features:

PONS is a modular, Python-native Delay-Tolerant Network (DTN) simulator designed for surrogate-assisted multi-objective optimization of routing strategies. It was developed as part of a Master's thesis to explore and optimize probabilistic forwarding parameters in DTNs with four simultaneous objectives:

• F1: Maximize delivery probability
• F2: Minimize energy consumption
• F3: Minimize delivery latency
• F4: Maximize energy fairness

Unlike traditional Java-based simulators such as The ONE, PONS integrates seamlessly with Python-based machine learning workflows, enabling direct surrogate modeling (Polynomial Regression, Gaussian Processes) and large-scale Bayesian optimization.

• Full Python Integration

  • Built entirely in Python using SimPy, NetworkX, NumPy, and Pandas.
  • Direct compatibility with Python ML/optimization libraries such as scikit-learn, BoTorch, and GPyTorch.

• DTN Routing Protocols

  • Epidemic (probabilistic variant with p1, p2, p3)
  • Spray & Wait
  • First Contact
  • Direct Delivery
  • PRoPHET
  • Static

• Mobility Models

  • Random Waypoint (built-in)
  • External movement from ONE simulator
  • External movement from ns2

• Contact Plan Models

  • Static networkx topologies (GraphML import)
  • Dynamic contact plans (ION, Core Contact Plan)

• Custom Metrics & Logging

  • Delivery probability
  • Energy consumption
  • Average/percentile latency
  • Energy fairness (Jain’s fairness index variant)

• Automation Tools

  • sweep.py – High-throughput parameter sweeps with parallel execution
  • find_pareto_front.py – Identify Pareto-optimal routing strategies
  • train_gp_models.py – Train Gaussian Process surrogates
  • Validate_Polynomials_Models.py – Validate and select polynomial surrogate models
  • Visualize_paretoFront.py – Generate trade-off visualizations (pairplots, parallel coordinate plots)
  • netedit – Generate/edit network topologies
  • ponsanim – Create animated visualizations from contact/event logs

└── pons/ ├── README.md ├── 90simtiming_test14W.csv ├── conv_12h.csv ├── conv_24h.csv ├── conv_6h.csv ├── EDA.py ├── find_pareto_front.py ├── LICENSE ├── plot_converg.py ├── pyproject.toml ├── requirements.txt ├── sim.py ├── sweep.py ├── sweep_sparse.csv ├── sweepSparse_10x50.csv ├── test_results.csv ├── timing_test.csv ├── timing_test12W.csv ├── timing_test8W.csv ├── train_gp_models.py ├── Validate_Polynomials_Models.py ├── Visualize_paretoFront.py ├── examples/ │ ├── corecontactplan-asym.py │ ├── corecontactplan.py │ ├── ext_ns2.py │ ├── ext_one.py │ ├── ioncontactplan.py │ ├── netedit_and_ccp.py │ ├── netplan.py │ ├── netplan2.py │ ├── ping.py │ ├── plan_reader.py │ ├── run_tests │ ├── static_routing.py │ ├── static_routing2.py │ ├── static_routing3.py │ ├── tests.json │ ├── two_net.py │ ├── data/ │ │ ├── 3n-asym.ccm │ │ ├── 3n-dyn-link.graphml │ │ ├── 3n-exported.ccm │ │ ├── 3n-exported.graphml │ │ ├── 3n-netedit.ccm │ │ ├── 3n-netedit.graphml │ │ ├── 3n.ccm │ │ ├── contactPlan_complex.txt │ │ ├── contactPlan_simple.txt │ │ ├── scenario1.ns_movements │ │ ├── simple.ccm │ │ └── topo.graphml │ ├── dtnmail/ │ │ ├── 3n.ccp │ │ ├── bpimapd.py │ │ ├── bpsmtpd.py │ │ ├── dtnmail.py │ │ ├── run.sh │ │ ├── test_recv.py │ │ └── test_send.py │ └── scenario/ │ └── simple_test/ │ ├── contacts.ccp │ ├── contacts.csv │ ├── contacts.json │ ├── flows.json │ └── nodes.json ├── outputs/ │ ├── gp_cv_scores.csv │ └── saved_models/ │ └── scaler.joblib ├── outputs_ValidatePolyModels/ │ ├── model_coeffs_F1.txt │ ├── model_coeffs_F2.txt │ ├── model_coeffs_F3.txt │ ├── model_coeffs_F4.txt │ └── polynomial_cv_scores.csv ├── pons/ │ ├── init.py │ ├── event_log.py │ ├── message.py │ ├── node.py │ ├── simulation.py │ ├── apps/ │ │ ├── init.py │ │ ├── app.py │ │ ├── ping.py │ │ └── udpgw.py │ ├── mobility/ │ │ ├── init.py │ │ ├── movement.py │ │ └── ns2_parser.py │ ├── net/ │ │ ├── init.py │ │ ├── common.py │ │ ├── netplan.py │ │ └── plans/ │ │ ├── init.py │ │ ├── core.py │ │ ├── ion.py │ │ └── parser.py │ ├── routing/ │ │ ├── init.py │ │ ├── directdelivery.py │ │ ├── epidemic.py │ │ ├── firstcontact.py │ │ ├── prophet.py │ │ ├── router.py │ │ ├── sprayandwait.py │ │ └── static.py │ └── utils/ │ ├── init.py │ ├── list_utils.py │ ├── misc.py │ └── vector.py ├── tests/ │ ├── init.py │ └── mobility/ │ ├── README.md │ ├── ns2_example_-500_3600_-482_3508.txt │ ├── ns2_example_0_3600_18_3035.txt │ └── ns2_tests.py └── tools/ ├── netedit/ │ ├── README.md │ ├── Dockerfile │ ├── netedit.py │ ├── requirements.txt │ ├── run-in-docker.sh │ ├── docker/ │ │ ├── entrypoint.sh │ │ └── xstartup │ └── gui/ │ ├── init.py │ └── dialogs.py ├── plot_contacts/ │ ├── plot_contacts.py │ └── requirements.txt ├── ponsanim/ │ ├── ponsanim.py │ └── requirements.txt └── scenariorunner/ ├── scenariohelper.py └── scenariorunner.py

• simpy >= 4.0
• networkx >= 3.2
• plotting:
  • seaborn
  • pandas
  • matplotlib
  • numpy
• tools:
  • pillow
  • opencv-python
  • tkinter

import random
import json

import pons
import pons.routing

RANDOM_SEED = 42
SIM_TIME = 3600*24
NET_RANGE = 50
NUM_NODES = 10
WORLD_SIZE = (3000, 3000)

# Setup and start the simulation
random.seed(RANDOM_SEED)

moves = pons.generate_randomwaypoint_movement(
    SIM_TIME, NUM_NODES, WORLD_SIZE[0], WORLD_SIZE[1], max_pause=60.0)

net = pons.NetworkSettings("NET1", range=NET_RANGE)
epidemic = pons.routing.EpidemicRouter()

nodes = pons.generate_nodes(NUM_NODES, net=[net], router=epidemic)
config = {"movement_logger": False, "peers_logger": False, "event_logger": True}

msggenconfig = {"type": "single", "interval": 30, 
  "src": (0, NUM_NODES), "dst": (0, NUM_NODES), 
  "size": 100, "id": "M"}

netsim = pons.NetSim(SIM_TIME, WORLD_SIZE, nodes, moves,
                     config=config, msggens=[msggenconfig])

netsim.setup()

netsim.run()

# print results

print(json.dumps(netsim.net_stats, indent=4))
print(json.dumps(netsim.routing_stats, indent=4))

Run using python3 or for improved performance use pypy3.

Some of the simulation core functions can be set during runtime without having to change your simulation code.

• LOG_FILE can be set to change the default event log file from /tmp/events.log to something else
• SIM_DURATION can be used to override the calculated simulation duration

For netedit there are also ways to influence its behavior:

• BG_IMG can be set to any image and it while be rendered as a background behind the network topology