可任意阶扩展的认知无线Ad Hoc网络频谱共享模型

Arbitrary Order Extendable Markov Model for Spectrum Sharing in Cognitive Radio Ad Hoc Networks

目前认知无线Ad Hoc网络(CRAHNs)频谱共享模型的研究大多忽视了未被授权频谱被占用过程对授权频谱被一、二级用户占用过程的影响.也有研究同时考虑了一、二级信道被占用的情况,提出了简化的排队模型,但是在进行性能分析时存在较大误差.主要挑战来自于高维马尔可夫状态转移图归纳的复杂性和计算的复杂性.本文研究提出一种新的分析方法,将原始复杂模型分解为易于分析的二维模型,然后将这些二维模型合成为三维模型,所建立的模型可任意阶扩展,可编程归纳实现.并可以用于分析任意一、二级信道数情形下CRAHNs的所有性能指标.大量的仿真和数值结果分析证实了模型的有效性.

Most of works in modeling the spectrum sharing of Cognitive Radio Ad Hoc Networks（CRAHNs） ignored the relationships between processes of using primary channels and that of using secondary channels.Some works considered the relationship and set up simplified queue-based model.However,they are not accurate in analysing the performance of CRAHNs.The main challenge may come from the complexity of setting up states transfer graph and that of computation in the multiple-dimensions Markov model.This paper copes with the challenge by a novel method.We decompose the complex original model into multiple 2-D models which are much easier to analyse,and then compose them all into a structured 3-D Markov model,which is arbitrary order extendable and programmable computation of the stable probability.By the new model,the performance of CRAHNs with an arbitrary number of primary channels and secondary channels can be evaluated.Extensive experimental results testify the validity of the proposed model.