基于图像K-means聚类分析的频谱感知算法

近年来,基于能量检测的协作频谱感知算法被广泛应用于频谱感知领域。由于该方法在计算能量检测的判决门限受噪声影响较大以及受限于认知用户的数量等问题,导致其检测性能受到影响。为了解决这一问题,本文提出一种基于图像K-means聚类分析的频谱感知算法。这种方法利用主用户信号存在与否的两种认知信号状态映射成图像,经过调整图像强度和高斯滤波预处理之后利用提取图像像素分布直方图的方法提取出特征向量,然后利用改进的K均值聚类算法对这些特征向量进行训练得到分类模型。最后利用训练好的分类模型对未知信号进行检测,从而实现频谱感知。仿真结果表明,本文所提出的频谱感知算法,在检测性能上优于传统能量检测以及协作频谱感知算法,尤其在低虚警概率、低信噪比的环境下效果更加突出。 

Spectrum Sensing Algorithm Based on Image K-means Clustering Analysis

In recent years,cooperative spectrum sensing algorithms based on energy detection are widely used in the field of spectrum sensing.Because the method is computationally inspected,the decision threshold of energy detection is greatly affected by noise and limited by the number of cognitive users.In order to solve this problem,this paper proposes a spectrum sensing method based on image K-means clustering.In this method,two cognitive signal states of the presence or absence of the main user signal are mapped into images,and feature vectors are extracted by image processing,and then the K-means clustering algorithm is used to train the feature vectors to obtain a classification model.Finally,the trained classification model is used to detect the unknown signal to achieve spectrum sensing.The simulation results show that the spectrum sensing algorithm based on image classification proposed in this paper is superior to the energy detection sensing algorithm and cooperative spectrum sensing algorithm in detection performance,and the effect is more obvious at low SNR and low false alarm probability.