Novel Cooperative Automatic Modulation Classification Based on Received-Signal Quality

A novel cooperative automatic modulation classification (CAMC) scheme based on the new fusion rule related to the received-signal quality is proposed in this paper. A twostage CAMC technique is proposed for the wireless sensor network containing a fusion center. In the first stage, each individual sensor in the network undertakes a graph-based automatic modulation classification (AMC) mechanism to identify the modulation type of an unknown target signal and also estimates the quality of the received signal. In the second stage, the fixed fusion center combines (accumulates) the local decisions from all sensors’ individual decision-weights dependent on blind estimation of the received-signal quality. Through sensors’ cooperation and the aforementioned decision fusion based on the individual received-signal qualities, our new CAMC scheme could mitigate the negative effects of channel distortion and noise often encountered in the existing single-node AMC methods. Furthermore, by allocating the majority of the computational load to the local sensors instead of the fusion center, our new approach can reduce the overall computational complexity while maintaining a low network-communication overhead. Extensive Monte Carlo simulations demonstrate the superior performance and robustness of our new CAMC scheme in comparison with the existing single-node AMC and CAMC methods.