Asynchronous multi-user OFDM with antenna diversity

Hyejung Jung, Purdue University

Abstract

In this dissertation, adaptive antenna based asynchronous multi-user orthogonal frequency division multiplexing (OFDM) communication systems are developed over fading channels. Although multi-user multiplexing is essential for spectrally efficient communications, synchronous operation may not be practically realizable given the demand for high-speed services with portability. In addition, multiple access interference deteriorates the quality of signal detection. Thus, an asynchronous multiuser OFDM signal model is developed, and a space-time minimum mean squared error (MMSE) equalization scheme is proposed to suppress the asynchronous interference. Furthermore, in order to amplify the performance gain from channel coding, a low-complexity iterative soft interference cancellation scheme is developed. The channel information required for effective equalization can be obtained via a novel subspace based semi-blind channel identification scheme. An inherent property of asynchronous OFDM signals is exploited so that the subspace decomposition method can separate the desired user's channel vector from the interfering user's channel vector, determining it uniquely up to a multiplicative scalar ambiguity. The proposed subspace based semi-blind algorithm improves the estimation performance under practical conditions and guarantees estimation of each user's channel vector. Symbol-timing and carrier-frequency offset estimation in asynchronous multiple access OFDM channels is also studied. The proposed joint method simplifies a complex multiple parameter estimation problem by assigning a unique set of pilot symbols to each user. Finally, the performance of asynchronous multi-user multiple input multiple output (MIMO) OFDM systems is investigated along with various transmit diversity techniques.

Degree

Ph.D.

Advisors

Zoltowski, Purdue University.

Subject Area

Electrical engineering

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS