An optimal estimation algorithm for multiuser chaotic communications systems

In a multiuser chaotic communications scheme, each user´s information is modulated with a different chaotic system and then transmitted independently. However, the receiver sees the superposition of these signals with additional noise. In this paper, a novel optimal estimation algorithm for such multiuser chaotic communications systems is presented. The goal is to estimate the chaotic signal being sent by each individual user so that its message information can be estimated. First, it is shown that it is possible, except in certain pathological cases, to separate chaotic signals. Then the symbolic dynamics representation of chaotic systems is exploited to enable the representation of the multiuser chaotic system with an equivalent trellis diagram. The Viterbi algorithm is then used to estimate each chaotic signal by utilizing the trellis diagram representation. This algorithm yields the minimum mean square error estimate of the chaotic signals. The performance of this algorithm in terms of bit-error rate (BER) is simulated for several different scenarios.