基于非统计量的无线传感网深盲度信号检测算法研究

项目名称： 基于非统计量的无线传感网深盲度信号检测算法研究

项目编号： No.61302155

项目类型： 青年科学基金项目

立项/批准年度： 2014

项目学科： 无线电电子学、电信技术

项目作者： 张昀

作者单位： 南京邮电大学

项目金额： 24万元

中文摘要： 本申请以无线传感网为背景，以提高无线资源利用率为目标，致力于研究无线传感网的深盲度信号检测问题。与已有的信号盲检测算法不同之处在于：本申请所实现的盲 检测算法不依赖信号统计量信息，特别适于未知发送信号星座字符集、含公零点信道或短少数据的深盲度信号检测。 本申请的特点和主要研究内容如下： （1）根据无线传感网络特征和设计目标，构造不同环境下适当的无线传感网模型，并建立基于字符集约束的二次规划性能函数。 （2）研究针对深盲度信号的盲检测算法。所谓深盲度信号即发送信号所属星座字符集未知，统计量未知，或者星座信号缺失的极端信号。这项研究对于无线传感网技术的实际应用具有重要的理论意义和参考价值。 (3)着重研究基于Hopfield神经网络的无线传感网深盲度信号检测算法。

中文关键词： 无线传感网；盲检测；分簇；Hopfield神经网络；

英文摘要： The proposal focuses on increasing resource utilization rate of wireless frequency spectrum on the base of wireless sensor networks. The new method for detecting deeply blind signals has several differences with the existing signals blind detection methods as follows: the methods of this proposal need not relying on the statistics information of the source signal, which are suitable to be used in the cases that the transmitted signal constellation characters are unknown and the channel is with common zeros and the data block is very short. The main features and contents of the proposal are as follows : (1) The channel model based on the characteristic and design object of the wireless sensor networks will be established. A new Quadratic programming performance functions based on alphabet control will be constructed. (2) The blind detection methods for the deeply blind signals will be discussed. And the deeply blind signals mean that the constellation and the statistics of the transmitted signals are unknown, or extreme signal missing alphabet in the constellation. The study has important theoretical meaning and reference value for the actual application of the wireless sensor network. (3)The proposal is to study Hopfield Neural Networks which can be used to detect deeply blind signals in wireless sensor netwo

英文关键词： Blind Signals Detection；Wireless Sensor Networks；Clustering；Hopfield Neural Network；