Low Density Parity Check Codes and Turbo Codes for Wireless Communication

Texas A&M Engineering Experiment Station, College Station TX

Investigators

Abstract

As the twenty-first century unfolds, new wireless services are being deployed and several more are envisioned. In order to provide these services, future wireless systems should be capable of handling much higher data rates than today's systems. The limited spectrum available for these services implies that the spectral efficiency of future systems will have to be significantly higher. In addition to being spectrally efficient, next-generation wireless systems will have to be power efficient. Lower transmit powers means lesser interference to other users, better reuse of the spectrum and, hence, higher efficiency for the overall network. Wireless channels offer several impairments to the transmitted signals, deteriorating the quality of the communication link. Increasing the power and spectral efficiency in the presence of such an unreliable link is a challenging task which can be greatly facilitated by the use of efficient coding and signal processing techniques. This research investigates the design of powerful coding techniques such as turbo codes and low density parity check codes for increasing the power and spectral efficiency of wireless systems. This research involves a detailed study of the design and performance analysis of turbo-like codes with low decoding complexity, and low density parity check codes (LDPC) for use in a wireless environment. Specific areas covered in detail are the design of coding schemes for use with multipath channels and turbo equalization, the design of LDPC codes for spectrally efficient modulation, and the design of incremental redundancy schemes using LDPC codes for use in a non-stationary environment. This work differs from the existing work in that it studies several novel coding techniques and concentrates on the design of codes matched to the iterative decoding algorithm being used, rather than the design of codes based on hypothetical maximum-likelihood decoding.