On using diversity to improve the performance of wireless communication systems

Abstract

Diversity is an effective way to combat channel fading and can be exploited at multiple protocol layers to achieve reliable transmissions over wireless communication networks. This dissertation develops resource-efficient algorithms that explore diversity at the physical layer and between physical and medium access control layers. At the physical layer, we propose a novel bandwidth efficient scheme for single transmit/receiver antenna Orthogonal Frequency Division Multiplexing (OFDM) systems. As opposed to the existing methods, the proposed approach is characterized by both spectral diversity and blind channel estimation capabilities through non-redundant linear precoding and low computational complexity algorithms. An optimal precoding scheme is proposed that, for quadrature phase shift keying signals results in minimum channel error and asymptotically minimum bit-error rate. The performance is evaluated analytically and the advantages over other methods are shown with simulations.At the intersection between the physical and medium access control layers, we propose a novel random access algorithm for collision resolution. Collided data is forwarded by various relays in different time interval to achieve temporal and spatial diversity. Upon receiving all the transmitted and forwarded signals, the destination node can formulate and solve a multiple-input multiple-output problem to recover the original packets. A comprehensive analysis of the diversity gain and capacity of the scheme is provided. The proposed scheme is capable of dealing with the wireless fading channel without bandwidth expansion nor additional antenna hardware. It also requires minimal scheduling overhead and enables efficient use of network power.%%%%Ph.D., Electrical Engineering – Drexel University, 2004