Iterative Receiver Techniques In Wireless Communication

In recent years, the market’s demand for high-speed wireless data communication is showing a rapid growth. However, in high-speed wireless data communication systems, the problem of channel divergence is prominent and makes the system subject to severe inter-symbol interference (ISI) which can deteriorate the quality of communication. Channel equalization technique is one of the important measures to suppress ISI and it can further improve the performance of the system to unite the equalizer and the decoder, which is the so-called turbo equalization technique. In accordance with the classification criteria of the equalizer, Turbo equalization can be divided into the time-domain equalization and frequency-domain equalization. The time-domain equalizer can provide optimal performance, but its implementation complexity depends on the channel memory length, which makes it unsuitable for serious divergence channels. Performance of the frequency-domain equalizer is slightly worse than the time-domain equalizer, but its implementation complexity has nothing to do with the channel length of memory, and can use the Fast Fourier Transform (FFT) technique to reduce the implementation complexity, and therefore it is more suitable for high-speed data communication.This paper mainly focuses on the Turbo frequency domain equalization technique and tries to reduce the implementation complexity and improve the performance. The contents of the study and innovation are summarized as follows.It proves that the maximum a posteriori (MAP) equalization algorithm and the time-domain linear minimum mean square error (MMSE) algorithm both provide superior performance by simulation, but they have a high complexity. It studies the frequency-domain linear MMSE (FDE-MMSE) equalization algorithmand develops the improved frequency-domain linear MMSE algorithm. The latter algorithm keeps the same complexity with the former one, but its performance is significantly better. Therefore the new algorithm has a great application value. It then studies the soft-interference-cancellation frequency-domain equalization (FDE-SIC) algorithm and suggests using full information instead of extrinstic information as the input of the equalizer. The simulation results show that this method can significantly improve the bit error performance. It describes three equalization algorithms based on decision feedback (DFE) and analyses their respective strengths and weaknesses.Finally, itputs forward a scheme whichcombines the improved frequency-domain linear MMSE equalization algorithm with the Space Time Block Code (STBC).This scheme can provide the communication system with lots of diversity gain, and reduce the implementation complexity as well. So the scheme has the characteristics of both high performance and practicality.In addition to the above work, the paper also gives the stability analysis of the above Turbo frequency-domain equalization algorithms, which shows the impact of deviation of channel estimation and signal-to-noise ratio (SNR) estimation on the performance of the algorithms on basis of simulation. The simulation results can gives a meaningful guidance to engineering practice.