认知网络中的多用户MIMO线性协作频谱感知问题研究

摘要

研究了多用户MIMO的线性协作频谱感知问题以提高频谱检测的可靠性,推导了多用户MIMO线性协作感知系统的局部检测和全局检测策略,建立了在给定误警概率的情况下控制中心通过优化给各个用户的信号所分配的权值来最大化全局检测概率的优化模型.进一步,引入了遗传算法来求解上述全局检测优化问题的最优权值,期望能够有效降低算法在频谱感知过程中检测时间.仿真结果表明,多入多出能够明显提高协作频谱感知系统频谱检测可靠性,而遗传算法高效且稳定,相比较其他算法能够有效改善系统的检测性能.%The multiuser MEMO based linear cooperative spectrum sensing problem was investigated in cognitive radio system to improve the reliability of spectrum sensing. Both the local spectrum detecting strategy and global spectrum detecting strategy for multiuser MIMO based linear cooperative spectrum sensing system were derived. Then the optimization model that the different weights assigned on the receive signals of each user at the fusion center for global decision were optimized to maximize the detection probability given a targeted probability of false alarm is established. Furthermore, genetic algorithm (GA) was introduced to find the optimal weight vector of the above-mentioned cooperative spectrum sensing problem, with the purpose of reducing the sensing time in the spectrum sensing process. The simulation results show that, the reliability of spectrum sensing in cooperative spectrum sensing system can be efficiently enhanced with multiple antennas. Besides, the proposed GA method is efficient and stable, and achieves better detection performance when compared with the existing methods.