摘要
为了提高OFDM信道估计的性能,提出了小波与孪生支持向量机相结合的基于导频的信道估计算法.由于TSVR算法对不同位置的训练样本赋予同样的权重,导致了性能降低.本文利用小波变换对训练数据进行预处理得到权值矩阵和权值向量,以此建立孪生支持向量机回归预测模型WTWTSVR,利用WTWTSVR算法来预测无导频处的子载波的信道频率响应.以误码率和均方误差为评价标准的仿真结果表明,在Jakes快衰落信道模型条件下,改进后的孪生支持向量机导频信道估计方法与传统的孪生支持向量机回归方法和传统的信道插值方法比较,具有更好的预测性能.
To improve the performance of OFDM channel estimation,a channel estimation algorithm based on pilot is proposed,which combines wavelet and twin support vector machine.Because the TSVR algorithm gives the same weight to training samples in different locations,it results in reduced performance.In this paper,the weight matrix and weight vector are obtained by preprocessing the training data by wavelet transform,and the regression prediction model of twin support vector machine WTWTSVR is established.The WTWTSVR algorithm is used to predict the channel frequency response of subcarrier at the non-pilot.The simulation results with bit error rate and mean square error as evaluation criteria show that under the condition of Jakes fast fading channel model,the improved pilot channel estimation method of twin support vector machine is compared with the traditional twin support vector machine regression method and the traditional channel interpolation method,which has better predictive performance.
作者
马艺梅
王立东
陈雪波
吴文良
MA Yi-mei;WANG Li-dong;CHEN Xue-bo;WU Wen-liang(School of Electronic and Information Engineering,Liaoning University of Science and Technology,Anshan 114051,China;Julong Company Limited Development Department,Anshan 114051,China)
出处
《小型微型计算机系统》
CSCD
北大核心
2020年第7期1433-1437,共5页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(71571091,71771112)资助
辽宁省教育厅科研项目(2019LNJC02)资助
辽宁科技大学领军人才项目(601011507-03)资助。
