Machine?learning?based pilot symbol assisted channel prediction

In this paper, machine learning (ML) algorithms are used for channel prediction in wireless communications. The performances of five ML compared terms the accuracy and symbol error rate (SER) different modulation schemes based on prediction. result shows that, prediction, support vector (SVM) has best performance stability. For signal detection, SVM linear regression (LR) have their own advantages ranges to noise ratio (SNR). At high constellation size, methods give similar existing scheme. From numerical examples, SERs LR can both reach lower than 10?3 binary phase shift keying 16-ary quadrature amplitude signalling, 1.13 × 10 ? 2 $\times 10^{-2}$ 4.28 3 10^{-3}$ signalling respectively. time, is more efficient.