Distributed Cooperative Automatic Modulation Classification Using DWA-ADMM in Wireless Communication Networks

Automatic modulation classification (AMC) is an important component in non-cooperative wireless communication networks to identify the schemes of received signals. In this paper, considering multipath effect practical propagation environments, a distributed cooperative AMC (Co-AMC) network based on machine learning proposed scheme networks. Specifically, feature vectors are first obtained by applying cyclic spectrum facilitate extraction signal. Then, classifier K-nearest neighbor (KNN) method designed obtain local decision for at each node. Meanwhile, reliability estimated two loss functions assess quality decision. Finally, unified result fuse decisions according their reliabilities fusion algorithm weighted average alternating direction multipliers (DWA-ADMM). The simulation results demonstrate that Co-AMC achieves superior accuracy compared existing methods across range and SNRs. More importantly, exhibits great flexibility practicability since it adaptive with various scales topologies.