Abstract
Error control coding is a scheme that involves the additional redundant bits in the stream of information bits to allow the detection and correction of symbol errors during transmission. In the field of channel coding, the discovery of turbo codes in 1993 was a significant breakthrough, since turbo codes can offer energy efficiencies close to the limits predicted by information theory. However, the reliable knowledge of channel gains and noise variance termed channel side information (CSI) condition the optimum performance of turbo codes. Because turbo codes are usually applied at low SNR, the process of phase estimation and tracking becomes difficult to perform well. Therefore, some forms of channel estimation are necessary at the receiver to estimate the CSI. M. C. Valenti has investigated the iterative channel estimation scheme based on the PSAM for turbo codes over fading channel, which incorporated the Wiener estimator in the turbo decoding process. In order to determine the coefficients of estimator, he assumes that the fading rate and noise variance are available at the receiver. However, in practice these parameters are hard to find. In this thesis, we utilize the LMS and MGOBA adaptive interpolation to estimate the complex channel gains. Firstly, the CSI are estimated from the known pilot symbols and the adaptive interpolations. After each iteration of turbo decoding, the CSI are re-estimated using the information fed back from the decoder. As compared to the system proposed by Valenti, our system needn’t to know the parameters of channel and has lower complexity. Computer simulation results have indicated that the MGOBA channel estimator can perform as well as the Wiener estimator used by Valenti.