In all-digital receiver implementations, an oscillator with good frequency stability but not synchronized with the carrier, heterodynes the incoming signal to baseband (direct conversion) where its quadrature components are sampled. In this type of receivers carrier synchnronization is replaced by channel estimation, which consists of the following tasks:
(i) estimation of the phase drift due to Doppler effect and/or frequency poor alignment between oscillators;
(ii) estimation of the channel time-varying gain provoked by fading and/or shadowing.
The problem of channel estimation presents new challenges with the emergence of mobile personal communications based on, for example, non-geostationary satellites, which are affected by significant Doppler frequency shifts. In these cases, the channel is strongly non-stationary (even with very short time intervals), leading to the need to update phase and gain estimates within the symbol interval.
Channel estimation in mobile digital communications is a relevant research subject not only from the applications point of view but also by its theoretical and conceptual aspects. This project proposes to resume previous work in this area by focusing namely on the following aspects:
a) synthesis of new algorithms for the simultaneous joint channel and clock estimation, and symbol detection based on stochastic nonlinear filtering;
b) development of new open-loop frequency and frequency-rate estimators;
c) application of the above-mentioned algorithms in scenarios of practical interest, such as the mobile digital communications with non-geostationary satellites;
d) application to spread spectrum and multicarrier systems where significant improvements are foreseen.
|Start Date: 01-01-2001|
|End Date: 31-12-2004|
|Team: Fernando Duarte Nunes, José Manuel Nunes Leitão|
|Groups: Pattern and Image Analysis – Lx, Pattern and Image Analysis – Lx|
|Local Coordinator: Fernando Duarte Nunes|