Learning Wireless Interference Patterns: Decoupled GNN for Throughput Prediction in Heterogeneous Multi-Hop p-CSMA Networks
This work addresses a computational bottleneck in wireless network analysis for researchers and engineers, offering a scalable machine learning solution that is incremental over prior GNN approaches.
The paper tackles the problem of predicting saturation throughput in heterogeneous multi-hop wireless networks using the p-persistent CSMA protocol, where existing models underestimate throughput by 48-62% and exact methods are computationally infeasible. The proposed Decoupled Graph Convolutional Network (D-GCN) achieves 3.3% normalized mean absolute error, outperforming baselines and enabling gradient-based optimization within 1% of theoretical optima.
The p-persistent CSMA protocol is central to random-access MAC analysis, but predicting saturation throughput in heterogeneous multi-hop wireless networks remains a hard problem. Simplified models that assume a single, shared interference domain can underestimate throughput by 48-62% in sparse topologies. Exact Markov-chain analyses are accurate but scale exponentially in computation time, making them impractical for large networks. These computational barriers motivate structural machine learning approaches like GNNs for scalable throughput prediction in general network topologies. Yet off-the-shelf GNNs struggle here: a standard GCN yields 63.94% normalized mean absolute error (NMAE) on heterogeneous networks because symmetric normalization conflates a node's direct interference with higher-order, cascading effects that pertain to how interference propagates over the network graph. Building on these insights, we propose the Decoupled Graph Convolutional Network (D-GCN), a novel architecture that explicitly separates processing of a node's own transmission probability from neighbor interference effects. D-GCN replaces mean aggregation with learnable attention, yielding interpretable, per-neighbor contribution weights while capturing complex multihop interference patterns. D-GCN attains 3.3% NMAE, outperforms strong baselines, remains tractable even when exact analytical methods become computationally infeasible, and enables gradient-based network optimization that achieves within 1% of theoretical optima.