Rateless Rate Adaption In Relay Networks

Improving the spectrum efficiency is very important in wireless communicationsystems. Wireless channels are always affected by time-varying factors, such as fading,interference, noise etc., so how to maximize spectrum efficiency with time-varyingchannel conditions has become a very important field of research. Rate adaption enables asystem to adjust its information transmission rate according to the real-time channelcondition, in order to improve the spectrum efficiency. In this thesis, we study the rateadaption technology in relay networks.In this thesis, we first study the Rateless rate adaption in relay networks, and theimplementation details of rate adaption are given. We then realize the rate adaptionscheme using Raptor Codes and SRA, and analysis the implementation features of thistwo Rateless codes in this scheme. With the adaption ability of Rateless codes, Ratelessrate adaption technology can decompose the rate adaption problem in multi-hop relaynetworks into rate adaption problems of multiple single-hop transmission, reducing thecomplexity of the systems analysis and design, while Rateless rate adaption does notrequire support feedback link and channel estimation, reducing the controlling complexity.CUDA, which is a platform for parallel program development, is applied toaccomplish the simulation. We introduce the idea and method of CUDA parallelprogramming, using coding and decoding of Raptor codes as an example. It shows that,the operating speed of CUDA parallel program is40~100times than that of C-serialprogram. It can be seen from the simulation results, that Rateless rate adaption schemeallows the information rate to meet the multi-hop transmission channel conditions. In caseof a low SNR, Raptor adaption scheme can achieve a higher information transmission rate;while in the case of a high SNR, the SRA adaption scheme can achieve a higherinformation transmission rate. In Raptor adaption scheme, the modulation mode isadjudged according to the channel information, while it’s not necessary for SRA adaptionscheme, reducing the controlling complexity and higher throughput can be achieve withSRA adaption scheme.