Spatial filtering of superimposed convolutional coded signals

In this paper, a method for simultaneously demodulating and estimating the parameters of a number of convolutional coded communication signals incident on an antenna array is presented. The method has the potential to increase the throughput of current multiple-access channel systems, e.g., satellite communications and digital mobile cellular phones, by using an antenna array. The contribution of this paper is the use of sequence estimation combined jointly with parameter estimation in array processing problems. A hidden Markov-model-based technique, the segmental k-means algorithm, is applied to the problem. This algorithm is an iterative procedure with two steps per iteration. The first step involves computing the most likely state sequence for each of the signals (demodulating the signals) given estimates of the signals' parameters. The second step refines the parameter estimates using the signals' mostly likely state sequence estimates. In the simulations presented, it is shown that a significant improvement in the accuracy of the demodulated signals and in the estimation of the signals' angle of arrivals is obtained when compared to a deterministic maximum likelihood estimation method.