中文摘要：

针对传统频谱感知方法易受噪声波动影响的缺点，提出了一种认知无线电自适应双门限能量检测算法．该算法基于双门限能量检测，通过比较认知用户接收的能量值和预定义的门限值，判断当前信道状态并自适应地选择一轮感知或两轮感知．推导了所提算法的检测概率、虚检概率和感知时间的性能表达式，并采用蒙特卡洛仿真得到信噪比与检测概率、感知时间的关系，最后利用GUNRadio和USRP搭建软件无线电系统平台，在实际的无线电环境中对所提算法进行验证．仿真结果与实际验证结果均表明，与传统频谱感知方法相比，所提算法在合理的感知时间范围内，能达到更高的检测概率．

英文摘要：

Due to the fact that the conventional spectrum sensing algorithm is susceptible to noise, an adaptive double-threshold energy detection algorithm for a cognitive radio is proposed. Based on double-threshold energy detection, the algorithm can adaptively switch between one-round sensing and two-round sensing by comparing the observations with the pre-fixed thresholds. Mathematical expressions for the probability of detection, the probability of false alarm, and the sensing time are derived. The relationships including signal to noise ratio （SNR） vs. the probability of detection and SNR vs. the sensing time are plotted using Monte Carlo simulation and the algorithm is verified in a real cognitive system based on GNU Radio and universal software radio peripheral （USRP）. Simulation and experimental results show that, compared with the existing spectrum sensing method, the proposed algorithm can achieve a higher probability of detection within a reasonable sensing time.