A Python-based software and mathematical simulator for Wireless Mesh Networks (WMN). This project demonstrates the critical difference in routing efficiency when utilizing the classic Hop Count metric versus the radio-oriented Airtime Link Metric (ALM), as standardized in IEEE 802.11s. It uses mathematical modeling of the transmission medium to assess throughput degradation without requiring physical hardware deployment.
In multi-node wireless environments, directly using the Hop Count metric leads to severely sub-optimal network performance. Algorithms that minimize the number of hops naturally tend to select the longest physical links to cover maximum distance per step.
However, at the edge of a radio transmitter's physical range, the Signal-to-Noise Ratio (SNR) drops critically. This leads to a decreased modulation rate, high Bit Error Rates (BER), and constant frame retransmissions. As a result, the route with the fewest nodes (hops) often yields the worst throughput and unpredictable latency.
To solve this, IEEE 802.11s introduced the Airtime Link Metric (ALM), which evaluates the amount of radio "airtime" required to transmit a test frame.
The simulator algorithmically calculates the weight of each graph edge using the formula:
Where:
-
$O$ : Overhead constants of PHY and MAC layers. -
$B_t$ : Test frame size (base: 8192 bits). -
$r$ : Physical bitrate in Mbps (modeled based on the distance between nodes). -
$e_f$ : Frame Error Rate (FER), factoring in distance-based signal loss and normally distributed noise.
The simulator is highly modular and structured into 4 core components:
- Topology Engine (
core/topology_engine.py): Generates 2D coordinates for a specified number of nodes and computes Euclidean distances. - RF Simulator (
core/rf_simulator.py): Assigns physical bitrates and calculates FER for potential links based on physical distance, injecting Gaussian noise to simulate real-world interference. - Routing Engine (
core/routing_engine.py): Calculates shortest paths using Dijkstra's algorithm twice: once using Hop Count and once using the computed ALM weights. - Visualizer (
core/visualizer.py): Renders the network graph, highlighting the distinct paths chosen by the two routing strategies.
- Clone the repository and navigate to the project directory.
- Install the required dependencies:
pip install -r requirements.txt
- Run the simulation:
# Run with default parameters python main.py # Run with custom parameters python main.py --nodes 50 --width 200.0 --height 200.0 --seed 42 # View all available options python main.py --help
Upon execution, the script outputs the calculated hop counts and total ALM costs for both routing strategies in the console. It also generates an image wmn_simulation_result.png at the root of the project.
This visualization provides clear proof of how Hop Count often selects long, low-quality links (represented as a direct dashed red line), while ALM successfully navigates through shorter, higher-quality, and more reliable links (solid green line), improving overall network stability.