复杂环境下的宽频段自适应协作频谱监测技术研究

From the demand of guaranting the national production safety and normalized spectrum access, with regard to the poor spectrum monitoring performance and low spectrum monitoring efficiency under complex electromagnetic and geographic environment in wideband spectrum monitoring senor networks, we investigate the signal detection algorithm based on the power spectral density segment cancellation, adaptive cooperative sensing based on decision feedback, and incremental cooperative sensing transmission protocol.The fusion node adaptively adjusts the cooperative sensing parameters and selects the best cooperative sensing nodes according to the cooperative sensing result and channel state informatoin. As a result, the sensor networks's ability to capture the illegal signal is increased, and the energy and bandwidth consumption for transmitting the sensing result to the fusion node are decresed. Moreover, we focuse on the parallel cooperative spectrum sensing problem in the case of wideband multichannels.The objective function of parallel cooperative sensing is firstly constructered for the cooperative sensing efficiency.Then, according to the objective function, a coalition formatoin game for parallel cooperative spectrum sensing is deployed for the allocation of bandwidth and sensing nodes under the constraint of spectrum sensing performance requirement.

本课题从保证国民生产安全和规范化频谱使用的需求出发，针对复杂电磁环境和地理环境条件下，宽带频谱监测传感器网络中监测性能差和监测效率低的问题，研究基于功率谱密度分段对消的信号检测技术、判决反馈的自适应单目标协作频谱监测技术和递增协作监测传输协议，根据判决结果和信道信息调整协作检测参数，自适应选择协作节点，提升频谱监测网络对系统中非法信号的挖掘与捕获能力，减小协作监测的能量和带宽消耗。研究宽频段多目标条件下的并行协作检测技术，构造并行协作优化目标函数，基于联盟博弈的方法最优化分配协作节点集合和监测频段，在满足频谱检测性能的条件下，提升频谱监测效率。

本课题针对复杂电磁环境和地理环境条件下，宽带频谱监测传感器网络中监测性能差和监测效率低的问题，提出了基于功率谱密度分段对消的信号检测技术，该技术利用被监测信号周期图估计功率谱谱线互不相关的性质，以频带内一些谱线强度和与剩余谱线强度和的比值作为检验统计量，可实时鲁棒地感知监测频段的占用情况。研究了宽频段多目标条件下的并行协作检测技术，构造并行协作优化目标函数，首先提出了具有高复杂度和部分信道状态信息（CSI）的问题的穷举搜索算法，选择分配协作监测频段和监测节点。随后，为了系统降低复杂度，我们提出了一种检测门限最大化算法（DTMA），在保证监测性能的条件下，通过维持尽可能大的检测门限来最大化总SU的吞吐量，提升监测效率。同时，提出了一种混合交织/底层认知网络方法，自适应地切换交织和底层频谱接入模式，分析了协作监测对协作传输性能的影响。