摘要
噪声作为电力线载波通信的主要影响因素,特别是脉冲噪声对电力线信道特性极大破坏,严重阻碍了电力线载波通信的广泛应用。该文提出基于高斯混合模型(Gaussian Mixture Model,GMM)的马尔可夫(Markov)脉冲噪声建模,研究脉冲噪声的特性,以便噪声的处理。同时,结合K-means初始化对参数进行簇类划分,确定初始的GMM参数,包括均值、协方差矩阵、混合权重,等。再利用EM算法不断迭代,确定最终的参数,通过映射关系,形成高斯成分状态空间,构建出Markov链的状态空间。多个状态需要马尔可夫链决定转移与否,也即由状态转移矩阵决定,模拟出符合原始电力线噪声特性的生成噪声,实现脉冲建模的有效性。
Noise,particularly impulsive noise,is a major influencing factor that greatly disrupts the channel characteristics of power lines,significantly hindering the widespread adoption of power line communication.In response to this,a Markov pulse noise modeling approach based on the Gaussian Mixture Model(GMM)is proposed to study the characteristics of impulsive noise for better noise handling.Using K-means clustering for parameter initialization,the initial GMM parameters,such as mean,covariance matrix,and mixture weights,are determined.The Expectation-Maximization(EM)algorithm is then applied for iterative optimization of these parameters.Through mapping relationships,the Gaussian component state space is formed,and a Markov chain state space is constructed.The transition between multiple states is determined by the Markov chain,with the state transition matrix governing the process.This approach simulates generated noise that accurately reflects the characteristics of original power line noise,thereby ensuring the effectiveness of pulse noise modeling.
作者
刘康
滕清松
董武
杨成江
林靖
王平
LIU Kang;TENG Qingsong;DONG Wu;YANG Chengjiang;LIN Jing;WANG Ping(Guizhou Power Grid Co.,Ltd.Power Dispatching and Control Center,Guizhou,550002,China;School of Communication and Information Engineering,Chongqing University of Posts and Telecommunications,Chongqing,400065,China)
出处
《长江信息通信》
2025年第7期29-32,共4页
Changjiang Information & Communications
基金
受贵州电网有限责任公司电力调度控制中心2024年“低压电力线终端通信可靠性提升及环境重构技术研究”(0665002024030103TX00028/060000KC23100026)资助。
