pydelay 0.1.2

Network time-delay simulation with statistical analysis and plotting

pydelay

pydelay is a Python package for simulating and analysing network time delays. It supports two statistical delay models, a Gaussian long-distance model and an exponential short-distance model, and produces publication-ready plots of the delay distribution and signal pulse train.

Installation

pip install pydelay

Quick Start

As a library

import pydelay

# Run a long-distance (Gaussian) delay simulation
result = pydelay.delay("l", n_packets=5000, simulation_time=120.0, seed=42)

print(f"Mean delay : {result.mean_delay:.3f} s")
print(f"Std  delay : {result.std_delay:.3f} s")
print(f"Min  delay : {result.min_delay:.3f} s")

# Plot the results
fig, axes = pydelay.plot(result, show=True)

# Save the figure
pydelay.plot(result, save="delay_plot.png")

As a command-line tool

# Long-distance model, default settings
pydelay l

# Short-distance model, custom packet count and seed
pydelay s --npack 10000 --stime 60 --seed 0

# Skip writing the CSV output file
pydelay l --no-csv

Delay Models

Flag	Model	Distribution
l	Long-distance	Gaussian μ = 1.50 s, σ = 0.50 s
s	Short-distance	Exponential min ≈ 1 ms, mean ≈ 2 ms

API Reference

pydelay.delay(model, *, n_packets, simulation_time, seed, output_csv)

Run the delay simulation and return a SimulationResult.

Parameter	Type	Default	Description
model	"l" or "s"	—	Delay model
n_packets	int	5000	Number of packets to simulate
simulation_time	float	120.0	Time horizon in seconds
seed	int | None	None	Random seed for reproducibility
output_csv	str | Path | None	"delay_data.csv"	Path to save CSV; None to skip

pydelay.plot(result, *, zoom_start, zoom_stop, show, save)

Plot the pulse train and delay distributions.

Returns (fig, (ax_signal, ax_dist, ax_dist2)).

SimulationResult

Attribute	Description
.send_time	Packet send timestamps (ndarray)
.simulated_delay	Sampled delays (ndarray)
.pulse_time / .pulse_values	Pulse train arrays
.min_delay	Minimum delay
.mean_delay	Mean delay
.std_delay	Standard deviation of delay
.simdelay	Dict of {time, signals} for external use

Requirements

• Python ≥ 3.10
• numpy ≥ 1.24
• matplotlib ≥ 3.7
• scipy ≥ 1.10

License

MIT © Parham Kebria