一种新的基于认知用户通信效用的频谱共享算法

摘要：本文提出了一种新的基于认知用户通信效用的频谱共享算法，研究传输功率和比特误码率受限情况下的频谱分配问题。详细分析了瑞利衰落模型下的认知用户效用函数及其优化问题，在此基础上给出了一种基于注水线不断调整的功率分配机制，进而获得了认知用户的最优通信效用，并将该机制应用到频谱分配中，提出了一种基于“二分法”的频谱分配算法。研究中分别考虑了三种不同的优化模型，理论和仿真分析表明，“最大化比例公平”模型在保证用户之间频谱分配公平性的同时能够获得接近“最大化系统总效用”模型的通信总效用。

A Novel Spectrum Sharing Algorithm Based on Communication Utility in Cognitive Radio Networks

A novel algorithm based on users' communication utility is proposed for spectrum allocation problems. The problem is under the constraints of transmit power and bit-error-rate of the cognitive users. After analyzing the optimization process of users' utility in Rayleigh fading model, a mechanism based is proposed on water-filling to obtain the near-optimal communication utility. Then this mechanism is applied to spectrum allocation, and proposed a new algorithm based on bisection method. The approach considers three models with different optimization goals, theory analysis and simulation results show that the Max-Proportional-Fair model not only provides fair spectrum allocation among cognitive users, but also has a comparable system total utility with the Max-Sum-Utility model.