针对认知无线电系统中认知用户在利用频谱聚合技术时出现的调度不公平问题,提出了一种基于频谱聚合的全局比例公平调度算法.在比例公平调度的基础上,结合频谱聚合技术的特点,引入了频谱聚合的跨度与认知用户剩余数据队列长度这两个参量,使认知用户在聚合可用频谱的范围内,最大限度的保证吞吐量公平性.仿真结果表明,相比于局部比例公平调度算法和最大载干比调度算法,该算法在公平性指数和系统服务时间上更占优势,同时具有较低的吞吐量抖动,从而有效地提高系统容量和系统效率,确保认知用户之间对系统资源的公平利用. Aiming at the problem of unfair scheduling when using cognitive spectrum in cognitive radio system, this paper proposes a global proportional fair scheduling algorithm based on spectral aggregation. On the basis of proportional fair scheduling, combining with the characteristics of spectrum aggregation , The two parameters of the span of spectrum aggregation and the length of remaining data queue of cognitive user are introduced so that cognitive users can maximize the throughput fairness within the scope of aggregating the available spectrum.The simulation results show that compared with the local ratio Fair scheduling algorithm and maximum load-to-interference ratio scheduling algorithm, the algorithm is more dominant in the fairness index and system service time, while having a lower throughput jitter, thereby effectively increasing system capacity and system efficiency to ensure that cognitive users Between the fair use of system resources.