Support Vector Machine Regression and Artificial Neural Network for Channel Estimation of LTE Downlink in High-Mobility Environments

Abstract

In this paper we apply and assess the performance of support vector machine regression (SVR) and artificial neural network (ANN) channel estimation algorithms to the reference signal structure standardized for LTE Downlink system. SVR and ANN where applied to estimate real channel environment such as vehicular A channel defined by the International Telecommunications Union (ITU) in the presence of nonlinear impulsive noise. The proposed algorithms use the information provided by the received reference symbols to estimate the total frequency response of the time variant multipath fading channel in two phases. In the first phase, each method learns to adapt to the channel variations, and in the second phase it predicts all the channel frequency responses. Finally, in order to evaluate the capabilities of the designed channel estimators, we provide performance of SVR and ANN, which is compared with traditional Least Squares (LS) and Decision Feedback (DF). The simulation results show that SVR has a better accuracy than other estimation techniques.

Optimized Downlink Scheduling over LTE Network Based on Artificial Neural Network

“…Charrada [6] proposed an accurate channel environment estimation technique using the ANN and support vector machine regression (SVR) models for the standardized signal structure of the LTE downlink system [6]. This technique was used under the impulsive, non-linear noise that interfered with reference codes after considering the high mobility conditions.…”

“…Min-max scaling: This method scales the data to a fixed range, typically between 0 and 1. The formula used for this scaling is Equation (6).…”