Federated reinforcement learning for cooperative vehicular routing with privacy and communication constraints
Federated reinforcement learning supports cooperative vehicular routing without centralizing trajectory data, but model updates may reveal private information and incur communication overhead over intermittent vehicle-to-infrastructure links, particularly when post-quantum authentication is required.
Correlated private quantization with complementary shared randomness
We study correlated private quantization, using complementary shared randomness between paired vehicles to reduce aggregation mean squared error while preserving the marginal parameter-level local differential privacy and one-bit communication cost of independent quantization.
Tunable extension and privacy accounting
A tunable extension mixes shared and independent randomness to bound privacy leakage given paired outputs and shared-randomness side information. For this extension, Rényi differential privacy accounting establishes vector-level local differential privacy for one model upload even when pairings are revealed.
Analytical aggregation performance
Under matched vector-level privacy and identical local and global parameters, the extension yields 16.5% lower analytical aggregation mean squared error than independent quantization.
Evaluation with post-quantum-authenticated V2I communication
We evaluate cooperative routing by coupling traffic simulation with packet-level simulation of post-quantum-authenticated vehicle-to-infrastructure communication. Compared to full-precision updates, one-bit uploads reduce authenticated uplink size by 92.4% and increase mean verified updates aggregated per round from 19.6 to 26.5.
Mobility-constrained results and attack outcomes
Under mobility-constrained communication at a parameter-level privacy budget of 1, correlated private quantization reduces mean travel time by 14.3% relative to independent quantization at the same budget and communication cost. At a parameter-level privacy budget of 0.3, a vehicle-origin inference attack trained on independent simulations reaches 30% accuracy with correlated private quantization and 22% with independent quantization after 100 rounds, versus 91% without privacy and 10% for random guessing.
- Authenticated uplink size reduction (one-bit uploads vs full-precision): 92.4%
- Mean verified updates aggregated per round: 19.6 → 26.5
- Mean travel time reduction (budget = 1, correlated vs independent): 14.3%
- Vehicle-origin inference attack accuracy after 10