Diversity techniques for broadband wireless communications: performance enhancement and analysis

摘要

The diversity techniques have been proven to be effective for next generation broadband wireless communications, and are the focus of this thesis. The diversity techniques can be broadly categorized into three types: Space, Time, and Frequency. In this thesis, we are mainly concerned with frequency and space diversity techniques. Orthogonal Frequency Division Multiplexing (OFDM) is a frequency diversity technique which offers several benefits such as easier digital implementation, immunity to multipath channels, low complexity channel equalization, etc. Despite these desirable features, there are few inherent problems in OFDM such as high peak-to-average power ratio (PAPR). High PAPR demands large dynamic range in the transmitted chain such as digital to analog converter (DAC) and power amplifier (PA). Unless pre-processed, the transmitted signal gets distorted due to quantization errors and inter-modulation. In the initial stage of PhD candidature, the author focused on PAPR reduction techniques. A simple modification on conventional iterative clipping and filtering (ICF) technique was proposed which has less computational complexity. The power savings achievable from clipping and filtering method was considered next. Furthermore the ICF is compared with another distortion-less PAPR reduction technique called Selective Mapping (SLM) based on power savings. Finally, impact of clipping and filtering on the channel estimation was analyzed. Space diversity seeks to exploit the multi-path characteristics of wireless channels to improve the performance. The simplest form of the space diversity is the receive diversity where two or more antennas with sufficient spacing collect independent copies of the same transmitted signal, which contributes to better signal reception. In this thesis new analytical expressions for spectral efficiency, capacity, and error rates were presented for adaptive systems with channel estimation error. Beamforming (steering signal towards desired receiver) is another useful technique in multiple-antenna systems to further improve the system performance. MRT (Maximal Ratio Transmission) or MIMO-MRC is such system where the transmitter, based on channel feedback from the receiver, uses weighting factors to steer the transmitted signal. Closed form expressions for symbol error rates were derived for MRT system with channel estimation error. The results were extended to evaluate closed form expressions of error rates for Rectangular QAM. Antenna correlation was considered in another contribution on MRC systems. Relay and Cooperative networks represent another form of spatial diversity and have recently attracted significant research attention. These networks rely on intermediate nodes called "relays" to establish communication between the source and the destination. In addition to coverage extension, the relay networks have shown to offer cooperative diversity when there is a direct link or multiple relays. The first contribution is to analyze a dual-hop amplify-forward relay networks with dissimilar fading scenarios. Next error rates of Rectangular QAM for decode-forward selection relay system are derived. Multiple antenna at relay is included to analyze the benefits of dual spatial diversity over Rayleigh and Nakagami fading channels. Antenna selection is a cost-effective way to exploit the antenna diversity. General Order Antenna Selection (GOAS), based on Ordered Statistics, is used to evaluate signal statistics for a MIMO relay network.