Abstract:
Multipath fading channels introduces intersymbol interference (ISI), which is not known a priori, as such we face problem of data transmission over such channels. To protect the integrity of the data a controlled amount of redundancy is added (encoding) using an error correction code (ECC). For coded data transmission over such channels Douillard et al. proposed the "turbo equalization" approach, which is an iterative equalization and, decoding algorithm for receiver. The decoding strategy used to decode convolutional codes is the basic of 'turbo principle', which suggests the iterated exchange of soft information between different blocks of a communication receiver. The potential importance and applicability of this principle has been found to extend to a wide range of problems in communication. In iterative adaptive equalization and decoding, channel equalization and MAP decoding are jointly optimized in an iterative process. One drawback of this process is the exponentially increasing complexity of the equalization step. To overcome this, several equalizers applicable to "turbo equalization" requiring reduced complexity have been proposed. Furthermore, we present an alternate approach to rederive some of these approaches and equalize the received data in frequency domain.
In this dissertation, simulation of iterative frequency domain equalization and Max-Log-MAP decoding has been carried out in MATLAB environment. The performance has been evaluated using approximate MMSE Equalizer, approximate MMSE Equalizer with average variance and Max-Log-MAP algorithm for decoding
