摘要
为了提高正交频分复用(OFDM)系统的传输质量和有效性,提出了一种基于最小二乘支持向量机的OFDM非线性信道估计算法.通过在OFDM符号中插入导频而获得训练数据,利用最小二乘支持向量机将训练数据映射到高维空间,并在此空间采用结构风险最小化准则对时变信道频率响应函数进行回归估计,把低维空间的非线性估计转化为高维空间的线性估计,提高了估计的精度.仿真结果表明,该算法能够有效地减小由多径引起的频率选择性衰落的影响,与传统算法相比,在同一误码率条件下的信噪比提高了3～7dB.
In order to improve the communication efficiency and quality of orthogonal frequency division multiplexing (OFDM) systems, a pilot-aided OFDM channel estimation algorithm based on the method of the least square support vector machine (LS-SVM) was presented. Using pilots inserted of OFDM symbol, training data are gained. Depending on LS-SVM, the algorithm maps trained data into a high dimensional space and employs the principle of structure risk minimization in the space to carry out the regression estimation for the frequency response function of the time-varying channel. This algorithm transforms the nonlinear estimation in low dimensional space into the linear estimation in high dimensional space, so it improves the estimated precision. The simulation result indicates that this channel estimation algorithm effectively decreases the attenuation of frequency selection caused by multi-path channel. Compared with other traditional algorithms the signal to noise ratio is improved by 3 to 7 dB under the same bit error rate.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2005年第6期637-640,共4页
Journal of Xi'an Jiaotong University
基金
国家自然科学基金资助项目(90207012).
关键词
正交频分复用
信道估计
最小二乘支持向量机
时变信道
Algorithms
Communication channels (information theory)
Estimation
Signal to noise ratio
