Robust real-time channel prediction based on inaccurate instantaneous measurements: an approach

Abstract

The goal of this work is to show the feasibility of real-time signal strength prediction on the base of instantaneous channel measurements. In real-world systems, like currently available commercial WLAN equipment, the instantaneous signal levels can only be measured with little accuracy. They are affected by a lot of noise from different sources, the measurement hardware is often not calibrated and the sampling interval is coarse and varying. Channel prediction should therefore be able to deal with noisy and non-equidistant data. Furthermore, the algorithms should be relatively simple, since no complex hardware can be afforded for simple WLAN, Bluetooth or sensor systems. In this paper we introduce a prediction algorithm based on a simple artificial neural network. The algorithm is tested on signal strength values measured with a standard 802.11b WLAN card in various environment conditions. The prediction results are compared with a state-of-the-art channel prediction algorithm. We show that a simple artificial neural network can be used as a robust and comparatively accurate predictor.