Two-Dimensional Channel Estimation and Prediction for Scheduling in Cellular Networks

In this paper, we derive an accurate and simple statistical model for other-cell interference and received signal-to-interference-and-noise ratio (SINR) in single-tier cellular networks and propose an effective radio channel-estimation scheme from the current and previous feedback information to extract 2-D information on geometry and mobility about the radio channel status of mobile stations. We validate the effectiveness of the proposed scheme by evaluating the performance through mathematical analysis and simulations. Furthermore, we propose a linear SINR prediction scheme to alleviate the performance degradation due to delayed channel state information (CSI), and simulation results demonstrate that the predicted CSI provides a significant increase in throughput gain for the normalized SINR scheduling, providing a 78.5% increase for a single-tap finite-impulse response (FIR) filter and a 105% increase for a three-tap FIR filter.