Distributed linear multiuser detection in cellular networks based on Kalman smoothing

We consider the problem of multiuser detection in cellular networks. In particular, we present a distributed forward-backward algorithm with local message passing for efficient implementation of the linear minimum mean square error (LMMSE) receiver, for a simple model of a 1D cellular system. The distributed algorithm is based on the well-known interpretation of Kalman smoothing as a linear combination of the forward and backward filtered estimates. We also show that near-optimal performance can be achieved by only relying on information from a local linear segment of the entire array. This results in a limited extent distributed algorithm that greatly reduces processing delay, especially for large networks, yet with little loss in performance.