Author(s)

Fang Weiwei, Siping Hu

Corresponding Author

Siping Hu

Abstract

In this paper, by improving the traditional methods in the generation of training sequences and local matrices, a channel estimation method based on superposition training sequences is proposed. The method firstly generates a unipolar non-negative periodic real sequence by performing conjugate symmetric transformation, parallel string transformation, limiting and split combination on the complex sequence; then, according to the time domain statistical characteristics of the received signal and the value of the power distribution factor, the local matrix is reasonably designed; the channel estimation is completed by performing first-order statistical averaging processing on the received signal. Computer simulations show that if the power allocation factor remains the same, as the number of subcarriers in the system increases, the MSE decreases gradually, and the estimated performance is better. As the power allocation factor increases, the energy of the training sequence increases, the better the channel estimation performance and the higher the accuracy, and the effectiveness of the algorithm is verified.

Keywords

Orthogonal Frequency Division Multiplexing, power distribution factor, channel estimation, channel estimation