Detection of OFDM Signal in Fading Channel

概要

論文の詳細を見る
This paper proposes a method for reconstructing the QPSK signal in the orthogonal frequency division multiplexing (OFDM) system. The OFDM has several advantages, i.e., the possibility ofa simple equalization of the received signal. We consider the problem of reconstructing OFDM signals which are passed through fading channel and then corrupted with additive white Gaussian noise. This problem is known as the channel equalization which can be viewed as a received phasor classification problem. The proposed method is a radial basis function (RBF) neural network which learns the characteristic of signal constellations in the channel state and reconstructing the correct signal constellation under noisy circumstances. The hybrid learning process is used to train the RBF network. The hidden layer is trained by the hard c means clustering. The supervised learning with given input-output pairs are used to train the output layer.
社団法人電子情報通信学会の論文
2004-09-09