The Study For Tracking Estimation Of Time-Varying Mobile Radio Channels

In the telecommunication area, the tracking of time-varying mobile radio channels is a basilic technology. The development of the wireless communication will be a little without the study of time-varying channels. And the design of channel parameters affects the quality of communication directly. In this dissertation, the tracking problem is well studied based on IS-136 TDMA system, which has many applications such as multi-carrier systems, multi-antenna systems and multi-detectors. But it still is a difficult problem to tracking rapidly time-varying channels exactly. The actual models with time variability and fading can not be known explicitly. Furthermore, estimated transmitted symbols may include decision errors. Three algorithms for tracking time-varying mobile radio channels based-on IS-136 TDMA systems are proposed in this thesis:1. WLMS algorithm. The proposed algorithm is based on one key concept: the design is transformed into wiener filtering problem. A time-varying channel model is derived with real background. Then a time-invariant model is obtained, which is the access of using a polynomial approach, according to information reorganization. We will measure tracking performance by the covariance of tracking error. In order to minimizing the tracking error, an optimal filter is designed and can be performed by a polynomial spectral factorization.2. MMSE-based algorithm. In practical communication system, there inevitably exist uncertainties in the model coefficients due to channel variations. One contribution of the proposed algorithm is the treatment of covariance-bounded uncertainties in the analysis and design of the optimal filter for tracking. A robust optimal filter is designed based on MMSE criterion.3. H_∞algorithm. One H∞algorithm is developed by using a polynomial approach...