A self-adaptive optimal-threshold energy detection algorithm based on cognitive vehicular networks

In cognitive vehicular networks, dynamic channel conditions due to vehicle movement can cause sensing difficulties. To address this, we propose a self-adaptive optimal-threshold energy detection algorithm. The algorithm incorporates a self-feedback mechanism that minimizes the total error probability by adjusting the detection threshold based on the detector’s judgment results at each time slot. A closed-form expression for the optimal threshold is derived, and mathematical proof confirms the algorithm’s convergence. Simulation results show that the proposed algorithm outperforms fixed-threshold methods, achieving up to a 52% reduction in total error probability across varying Signal-to-Noise Ratio (SNR) and primary user occupancy probabilities. The algorithm shows robust convergence in dynamic SNR conditions, effectively mitigating errors from SNR fluctuations without tracking environmental SNR, thereby reducing system overhead and complexity.