EM algorithms for sequence estimation over random ISI channels.
This paper presents a new approach using EM (Expectation- Maximization) algorithms for ML (maximum likelihood) sequence estimation over unknown IS1 (inter-symbol interference) channels with random channel coefficients. By using the EM formulation to marginalize over the channel coefficient distribution, maximum-Iikeljhood estimates of the transmitted sequenceare obtained. The EM algorithms are shown to perform better, in terms of BER, than existing algorithms which perform jointly-optimal sequence and channel estimation.
|Main Author:||Perry, Richard.|
|Other Authors:||Berger, W., Buckley, Kevin.|