Alireza Tarighat 2005

Abstract of PhD Dissertation
Electrical Engineering Department, UCLA, July 2005
Advisor: Prof. Ali H. Sayed

Multi-Input Multi-Output (MIMO) OFDM Systems with Implementation Impairments

Alireza Tarighat, UCLA

High speed communications over broadband wireless channels has emerged as a key feature of future communications systems due in part to the explosive interest in information technology applications, including wireless sensor networks, mobile computing, high-speed mobile internet, and real time wireless multimedia applications. The demand for higher information capacity in these and other similar applications has motivated the utilization of broadband wireless channels in order to provide higher data rates. Orthogonal frequency division multiplexing (OFDM) is emerging as the modulation scheme for broadband communications. This is mainly motivated by the simple receiver structures for OFDM systems over broadband channels.

OFDM-based physical layers have already been chosen for several wireless systems such as the IEEE 802.11a wireless local area network (WLAN) in the 5GHz band, the recently adopted IEEE 802.11g wireless local area network (WLAN) in the 2.4GHz band, and the European digital video broadcasting sys- tem (DVB-T). It is also under consideration as the high rate alternate physical layer to the IEEE P802.15.3 wireless personal area network (WPAN), the IEEE 802.20 mobile broadband wireless access (MBWA) and the IEEE 802.16 wireless metropolitan area networks (WMAN).

This dissertation addresses some challenging issues in MIMO OFDM implementations. One limiting issue in implementing low-cost, low-power, and fully integrated wireless systems is the impairments associated with the analog processing due to component imperfections. The impairments in the analog domain are mainly due to fabrication process variations which are neither predictable nor controllable and tend to increase with fabrication technologies scaling down.

Chapters 1-3 of this dissertation address one of the dominating sources of such impairments (i.e., IQ imbalances) and develop novel efficient compensation algorithms in the digital baseband domain. Chapters 4-5 show how the structure of space time codes can be exploited to design e±cient MIMO OFDM receiver structures, both least-squares and adaptive. Chapters 6-7 of the dissertation present two structural enhancements to OFDM receivers in terms of data recovery and channel tracking.

Specifically, in Chapter 1, we shall model some sources of impairments in wireless transceivers, namely the effect of IQ imbalances on OFDM systems. In Chapter 2, we shall develop compensation algorithms for IQ imbalances for single-input single-output (SISO) OFDM systems and derive analytical performance bounds for the proposed compensation algorithms. In Chapter 3, we shall study the effect of IQ imbalances on multiple-input multiple-output (MIMO) OFDM systems and develop a framework to design receiver architectures that are robust to IQ imbalances. The chapter covers MIMO systems in general and MIMO systems with orthogonal space-time block codes (OSTBC). In Chapter 4, we shall show how the structure of space-time codes can be exploited to design e±cient least-squares and adaptive (RLS) receiver algorithms for MIMO OFDM systems. In Chapter 5, we shall show how the receiver algorithms developed in the previous chapter can be extended to OFDM systems that suffer from impairments at both the transmitter and receiver. In Chapter 6, we shall focus on another important issue in OFDM design, namely, the overhead due to cyclic prefix. In this chapter, we shall introduce a procedure at the receiver to exploit the cyclic prefix to enhance the data estimation process. The algorithms are derived for both SISO and MIMO cases. Finally, in Chapter 7, we shall develop an adaptive ¯lter with enhanced performance for channel estimation and tracking. In this chapter, we introduce a criterion that involves both the estimated phase error and the magnitude of the error.

Acknowledgment This work was supported in part by the National Science Foundation under grants CCF-0208573 and ECS-0401188. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.