Interleave division multiple access

Abstract

Wireless communications is the fastest growing sector of the telecommunications industry. The market of wireless communications is undergoing explosive growth. Wireless cellular communication systems evolved from the first-generation of analogue stage to the second-generation (2G) of digital stage, and now are stepping into the third-generation (3G) and fourth-generation (4G) of broadband stage. The services have changed from pure phone services to a variety of ones, such as video conferences, music or movie appreciation, games, Internet access, etc. Typical features for future wireless communication systems include high-capacity (high-spectral efficiency), high-rate, high-speed and broadband data transmission, low-cost, etc. Multiple access interference (MAI) and intersymbol interference (ISI) are two major factors that affect the performance of wireless communication systems. This work mainly focuses on the interleave-division multiple-access (IDMA) scheme to mitigate both MA1 and ISI. We have separated the research into several parts: basic IDMA transmitter and receiver principles in synchronous additive white Gaussian noise (AWGN) channels, receiver design in asynchronous multipath fading channels, performance analysis and optimization, and low-complexity measures. In addition, comprehensive and extensive comparison to CDMA systems is made throughout our study. We propose the IDMA scheme by introducing chip-level interleavers to conventional CDMA. First, following the turbo principle, we present an iterative chip-by-chip (CBC) detection algorithm for AWGN channels. This simple iterative CBC algorithm is very effective to mitigate MAI. It can support a large number of users while achieving near single-user performance. The CBC algorithm can be directly applied to CDMA systems and achieve superior performance. The related complexity is as low as several multiplications and additions, and is independent of the user number. This complexity is much lower than that of conventional CDMA multiuser detectors. Then we develop two low-cost CBC detection algorithms for multipath channels: namely, the log-likelihood ratio combining (LLRC) and joint Gaussian combining (JGC) approaches. The normalized complexities of these two receivers are respectively linear and quadratic to the path number, and both are also independent of the user number. These low-cost receivers can support high rate up to 6 bitslchip with two receive antennas. For multirate IDMA systems, we suggest the hybrid JGC and LLRC detection approach as the receiver compromising performance and complexity. For TDMA or single-user systems, the approximate JGC approach is equivalent to the extending window MMSE equalizer, for which we develop a low-cost recursive implementation technique. We present SNR analysis and evolution to assess the performance of IDMA systems. This approach can accurately characterize the system, and is also much simpler and more concise than the analyses of conventional CDMA systems that are based on the large system assumption. To support high rates, we optimize the received and transmitted power with the aid of SNR evolution. Linear programming and power matching techniques are proposed to solve the optimization problem. Multiple low-cost measures are presented to reduce memory demand and implementation complexity. We perform a comprehensive and comparative study on IDMA and CDMA. The proposed IDMA system provides an efficient and effective solution to high-rate multiuser wireless communications. The low-complexity and high-performance properties make the IDMA scheme a competitive candidate for next generation wireless systems.

Date of Award	15 Feb 2006
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Ping LI (Supervisor)