An LPI design for secure burst communication systems

Abstract

An LPI burst communication model based on conditional maximum entropy is presented in this paper. In this model, the conditional entropy of transmitting moments is the largest, and the prior data are used as the sample space, while Lagrange multipliers are selected as optimization variables. Hybrid Chaotic Particle Swarm Optimization (HCPSO) that is used in the model takes the dual programming of the conditional maximum entropy as objective function, and the conditional maximum entropy model is ultimately determined through this optimization algorithm. Compared with the usual method of fixed threshold, the simulation results show that the conditional maximum entropy method not only has longer effective communication time, but also can effectively increase the uncertainty of transmitting moments. The more the uncertainty of transmitting moments, the better the low probability of intercept performance is. So the burst communication has better performance of low probability of intercept using conditional maximum entropy model.