中文摘要：

提出了一种被分成延时和丢弃两部分的特殊队列,建立马尔科夫模型分析了基于此队列的用户个数受限的认知无线电网络性能,并推导其丢包率、分组时延以及频谱利用率表达式,同时,对认知无线电网络进行仿真。分析和仿真结果表明：（1）分析模型能准确预测认知无线电网络性能;（2）次用户的丢包率和分组时延随着延时队列长度的增大分别减小和增大,而频谱利用率不受延时队列长度的影响;（3）采用认知无线电技术,能大幅提高频谱利用率。

英文摘要：

A queue is introduced which is separated into two parts, delay queue and discard queue, and proposed a Markov model to derive the formulas, i. e. , packet loss ratio, packet delay and spectrum utilization, for cognitive radio networks with such a special queue and limited numbers of users. Experiment results have shown that （ 1 ） the proposed analytical models can be used to accurately evaluate the performance of the cognitive radio networks; （2） with the decrease of delay queue length, packet loss ratio decreases and packet delay increases while spectrum utilization is not changed; （3） the spectrum utilization is improved greatly by adopting cognitive radio technology.