Abstract: Electric power wireless communication network supports enormous quantity wide users, high concurrent services, and complex operating environment, which is represented as heterogeneous networks with different complementary cover scope coexistences. In order to support intelligent power terminals to select access network, network discovery and identification must be performed firstly. Based on TD-LTE, WiMAX, Smart Grid NAN and 230 MHz heterogeneous networks coexistence scenario, a network identification algorithm is proposed by integrating physical signal time-frequency characteristics with MAC layer protocol characteristics to discover and identify coexisting wireless networks. This algorithm combines the modified window-sliding energy detection with cyclostationary feature detection to execute network discovery and identification, and multicycle characteristics are used in cyclostationary feature detection to improve the signal identification precision. The simulation results show that the proposed algorithm can effectively identify various heterogeneous cognitive wireless networks.

Key words: heterogeneous multi-networks coexistence, network discovery and identification, energy detection, cyclostationary feature detection

摘要： 电力无线通信网支撑用户量大面广、业务高并发、运行环境复杂，表现为异构多网混合共存。为了支持智能电力终端动态选择网络接入，必须首先执行网络发现与识别。针对TD-LTE无线通信专网、WiMAX无线通信专网、智能电网邻域网和230 MHz电力无线专网异构多网共存场景，提出一种融合物理层信号时频特征和MAC层协议特征的网络识别算法。该算法结合改进的窗口滑动能量检测和多周期特性加权循环平稳特征检测执行网络发现与识别。仿真结果表明，该算法能有效识别异构的多种电力无线通信网络。

关键词: 异构多网共存, 网络发现与识别, 能量检测, 循环平稳特征检测