Abstract: The single-node energy detection has the drawbacks of low detection accuracy and “hidden terminal” while cooperative sensing algorithms usually use equivalent weights for data fusion instead of considering the influence to the detection performance resulting from communication environment of different nodes. In order to solve the above problems, this paper proposes an adaptive weighted cooperative spectrum sensing algorithm based on improved energy detection which improves single node energy detection and derives the relationship between Signal-to-Noise Ratio（SNR） and decision threshold under the condition of minimal error probability. This paper adopts dichotomy to get dynamic threshold values and corresponding false probability and detection probability at different SNR. It regards the function of false probability and detection probability as the weighted factor to do data fusion. The simulation results show that the proposed algorithm can achieve reliable detection performance of cooperative sensing system with low SNR.

Key words: cognitive radio, cooperative spectrum sensing, self-adaption weight, dichotomy

摘要： 针对单节点能量检测法存在的“隐藏终端”和检测准确性低以及协作频谱感知算法大多采用等权重进行数据融合，未考虑不同节点所处的通信环境对检测性能的影响等问题，提出一种基于改进型能量检测的自适应加权协作频谱感知算法。该算法通过对单节点能量检测方法的改进，在单节点检测错误概率最小的条件下，导出了信噪比与判决门限的关系式，利用二分法求得不同信噪比下的动态门限值，得到相应的虚警概率和检测概率，以虚警概率和检测概率的函数作为加权因子进行数据融合。仿真结果表明，所提算法使协作感知系统在低信噪比条件下也能获得可靠的检测性能。

关键词: 认知无线电, 协作频谱感知, 自适应加权, 二分法