Code-aided iterative techniques in OFDM systems

Inspired by the 'turbo principle', this thesis deals with two iterative technologies in orthogonal
frequency division multiplexing (OFDM) systems: iterative interference cancelation
in space-frequency block coded OFDM (SFBC-OFDM) and iterative channel estimation/
tracking in OFDM Access (OFDMA) with particular application to Worldwide
Inter-operability for Microwave Access (WiMAX) systems.
The linear matched filter (MF) decoding in SFBC-OFDM is simple yet obtains maximumlikelihood
(ML) performance based on the assumption that the channel frequency response
remains constant within a block. However, frequency response variations gives
rise to inter-channel interference (lCI). In this thesis, a parallel interference cancelation
(PIC) approach with soft iterations will be proposed to iteratively eliminate ICI in
G4 SFBC-OFDM. Furthermore, the information from outer convolutional decoder is exploited
and fed back to aid the inner PIC process to generate more accurate coded bits for
the convolutional decoder. Therefore, inner and outer iterations work in a collaborative
way to enhance the performance of interference cancelation.
Code-aided iterative channel estimation/tracking has the ability of efficiently improving
the quality of estimation/tracking without using additional pilots/training symbols. This
technique is particularly applied to OFDMA physical layer ofWiMAX systems according
to the Institute of Electrical and Electronics Engineers (IEEE) 802.16 standard. It will be
demonstrated that the performance of the pilot-based channel estimation in uplink (UL)
transmission and the channel tracking based on the preamble symbol in downlink (DL)
transmission can be improved by iterating between the estimator and the detector the
useful information from the outer convolutional codes.
The above two issues will be discussed in Chapter 5 and Chapter 6, and before this, Chapter
2 to Chapter 4 will introduce some background techniques that are used throughout
the thesis.