Based on a comparative analysis of the Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks, we optimize the structure of the GRU network and propose a new modulation recognition method based on feature extraction and a deep learning algorithm. High-order cumulant, Signal-to-Noise Ratio (SNR), instantaneous feature, and the cyclic spectrum of signals are extracted firstly, and then input into the Convolutional Neural Network (CNN) and the parallel network of GRU for recognition. Eight modulation modes of communication signals are recognized automatically. Simulation results show that the proposed method can achieve high recognition rate at low SNR.

Public communication infrastructures are susceptible to disasters. Thus, the Emergency Communication Networks (ECNs) of small groups are necessary to maintain real-time communication during disasters. Given that ECNs are self-built by users, the unavailability of infrastructures and the openness of wireless channels render them insecure. ECN security, however, is a rarely studied issue despite of its importance. Here, we propose a security scheme for the ECNs of small groups. Our scheme is based on the optimized Byzantine Generals’ Problem combined with the analysis of trusted security problems in ECNs. Applying the Byzantine Generals’ Problem to ECNs is a novel approach to realize two new functions, debugging and error correction, for ensuring system consistency and accuracy. Given the limitation of terminal devices, the lightweight fast ECDSA algorithm is introduced to guarantee the integrity and security of communication and the efficiency of the network. We implement a simulation to verify the feasibility of the algorithm after theoretical optimization.