Long-Range Fading Prediction and Realistic Physical Modeling to Enable Adaptive Transmission for Mobile Radio Networks

摘要

The feasibility of adaptive transmission enabled by Long Range Fading Prediction (LRP) was investigated for rapidly varying fading channels encountered in peer-to-peer mobile communication systems. The LRP algorithms were tested using the standard Jakes model and the realistic physical model developed by the Principal Investigators. This research was an interdisciplinary effort in communication theory, physics, and signal processing. Correlated fading was exploited in the development of the LRP algorithm to enable adaptive modulation for a Frequency Hopping Spread Spectrum (FH/SS) mobile radio channel. Adaptive frequency diversity transmission for FH systems with partial band interference also was explored. Significant performance gains were demonstrated relative to non-adaptive methods. Efficient channel loading for wireless Adaptive Orthogonal Frequency Division Multiplexing (AOFDM) systems aided by robust LRP was investigated and reduced feedback techniques were developed. The bit rates approaching those of AOFDM with perfect knowledge of fading conditions were achieved. Several linear and nonlinear Multi-Access Interference (MAI) mitigation and frequency diversity combining transmitter precoding methods aided by the LRP were developed for the downlink of the Direct Sequence Spread Spectrum (DS/SS) systems. The proposed techniques were shown to improve upon previously proposed precoding and detection methods while providing the desired performance/complexity trade-off.