This paper develops an adaptive maximum likelihood sequence detection (MLSD) algorithm for the Raleigh flat fading environment in association with channel coefficient estimation and channel identification. The design of the MLSD receiver depends on a knowledge of the channel. Along with different channel knowledge assumptions we consider the general case when the channel coefficient is time-variant and the channel statistical characteristics are unknown. The proposed adaptive algorithm has three recursive steps. The channel coefficient is estimated for each path in the trellis diagram by using Kalman filtering; then, based on a dynamic programming algorithm, the transmitted data is detected for each survivor path and, at the final step, the channel is identified by estimating the channel parameters associated with the best previous survivor path. The algorithm is able to track the changes in the channel parameters when the fading rate is changing due to the varying vehicle speed. Performance evaluation and comparisons are considered for different levels of channel knowledge by computer simulation.
[1]
Edward Shwedyk,et al.
Detection of bandlimited signals over frequency selective Rayleigh fading channels
,
1994,
IEEE Trans. Commun..
[2]
Subbarayan Pasupathy,et al.
Innovations-based MLSE for Rayleigh fading channels
,
1995,
IEEE Trans. Commun..
[3]
John H. Lodge,et al.
Maximum likelihood sequence estimation of CPM signals transmitted over Rayleigh flat-fading channels
,
1990,
IEEE Trans. Commun..
[4]
G. David Forney,et al.
Maximum-likelihood sequence estimation of digital sequences in the presence of intersymbol interference
,
1972,
IEEE Trans. Inf. Theory.
[5]
S. Haykin,et al.
Adaptive Filter Theory
,
1986
.
[6]
S. Pasupathy,et al.
Adaptive state allocation in MLSE for Rayleigh fading channels
,
1996,
Proceedings of Vehicular Technology Conference - VTC.
[7]
G. A. Arredondo,et al.
A multipath fading simulator for mobile radio
,
1973
.