Low-rate feedback and low-complexity schemes in wireless communications

Wireless communication is ubiquitous. Ever increasing demand for higher rates and greater reliability is a fact of wireless life. Thus, it is essential to enable more efficient use of wireless resources and smarter scheduling of wireless traffic. In this thesis, we will study low-complexity encoding and decoding schemes for wireless signals. We will investigate how to improve performance using low-rate feedback in wireless communication systems. Implementation will also be addressed.;We begin by focusing on complexity in wireless multi-input multi-output systems. Conditional optimization will be introduced as a decoding primitive for high-rate space-time codes that are obtained by multiplexing in the spatial and code domains. The conditions under which conditional optimization leads to reduced complexity Maximum Likelihood (ML) decoding are captured in terms of the induced channel at the receiver. These conditions are then translated back to the transmission domain, leading to codes that are constructed by multiplexing orthogonal designs. Other algebraic principles of high-rate spacetime coding design with intrinsic low-complexity decoders will be elaborated. Highly parallel implementation of conditional optimization will be realized on an emerging computing platform, Graphics Processing Units that have great potential to improve the cost / performance tradeoff of 4th generation wireless systems.;Next we reduce the complexity of channel decoding by using feedback to improve the quality of the channel. The method is Code Diversity which integrates space-time coding with beamforming. A small number of feedback bits are used to select from a family of space-time codes. Different codes lead to different induced channels at the receiver, where channel state information is used to instruct the transmitter to choose the code. This method of feedback provides gains associated with beamforming while minimizing the number of feedback bits. It complements the standard approach to code design by taking advantage of different (possibly equivalent) realizations of a particular code design. Feedback can be combined with sub-optimal low complexity decoding of the component codes to match ML decoding performance of any individual code in the family. It can also be combined with ML decoding of the component codes to improve performance beyond ML decoding performance of any individual code. Code diversity can also be made resilient to feedback error. We will demonstrate the practicality of code diversity in space-time coded systems by showing that predicted performance gains based on instantaneous feedback are largely preserved when the feedback is based on long-range prediction of rapidly time-varying correlated fading channels. This is verified by simulations with the Jakes channel model and with a highly fidelity channel associated with a more realistic physical scattering environment. The analysis of point to point communications will be extended to multi-user systems.;Given that a channel is usually partially known at the receiver, we will then consider a Bayesian detection framework that interpolates coherent decoding and non-coherent decoding. The channel state information known at the receiver is quantified by channel distribution information that is estimated from any available training symbols. In the particular case of orthogonal signaling, sphere decoding will be adopted to reduce decoding complexity. In practical implementation, channel is quantized at the receiver that typically leads to loss of channel state information. The effects of quantization on the discrete-time channel capacity will also be explored. Based on the analysis of quantization error, a general finite precision implementation methodology will be presented that including two fixed-point conversion criteria for space-time coded systems within an integer optimization framework.