由于受认知无线电与中继通信技术的启发,提出了一种认知中继网络模型.该模型由源节点、目的节点、认知中继节点及主用户(primary user,PU)构成.认知中继节点以与PU共存的方式为源节点辅助传输信息到目的节点,只要保证其对PU通信造成的干扰在PU干扰门限值以下.假设源节点、目的节点和认知中继节点之间的瞬时信道边信息(channel side information,CSI)和认知中继节点到主用户之间的均值信道增益已知的前提下,研究该模型中的认知中继节点分别采用放大转发(amplify-and-forward,AF)和基于AF的中继选择(selection AF,S-AF)下的功率分配策略,该策略以最小化系统中断概率为目标,同时也满足认知中继节点的发射功率约束(包括总发射功率和个体发射功率约束)和对主用户的干扰功率约束.最后,通过数值仿真来验证推导出的功率分配策略.仿真结果表明:本文提出的最优功率分配策略,无论在AF,还是S-AF下,均能明显的改善系统的中断性能和平均吞吐量;同时在S-AF下最优分配策略可以得到更高的平均吞吐量,因此中断概率更小.
A novel cooperative sensing method is proposed in this paper. The proposed scheme adopts sensing creditability degree to characterize the impact of the distance and the channel parameters on the sensing result,and considers that each user has different average SNR and different decision threshold,by using General Nash Bargaining Solution (GNBS) strategy in Cooperative Game Theory (CGT),the detection performance for two-user case are derived. For multi-user case,the sensing performance is obtained with Hungarian method. Compared with the traditional schemes such as Nash Bargaining Solution (NBS) and AND,the proposed scheme covers all the factors mentioned above,and enhances the sensing rationality and reliability. Simulation results show that the proposed scheme can further improve the sensing performance and creditability.
