Predictive Techniques for Enabling Fast and Accurate Medium Access Control in Distributed Power-Controlled Networks

The goal of this study is to define a scheme for fast and accurate medium access control (MAC) in distributed power-controlled wireless networks. The system model assumes (i) arbitrary topologies, (ii) arbitrary call arrival rates, (iii) multiple links transmitting simultaneously over shared interference-limited channels, and (iv) transmitters updating their powers to maintain a predefined signal-to-interference and noise ratio (SINR) at the receiver. Departing from our own theoretical framework on the computation of the dominant eigenvalue of the network information matrix, we discuss an accurate MAC scheme that uses interference measurements as its only decision-making input. By accuracy is meant that the MAC scheme grants channel access to all links with achievable target SINRs and rejects others. In contrary to other schemes, our approach allows once-admitted links to continuously monitor the achievability of their SINR targets without any overhead. However, the initial admission decision of passive links relies on energy-consuming channel probing, which can be protracted in networks with high channel reuse. While maintaining the accuracy and reliability, we reduce the overall call admission delay and energy usage of the MAC scheme by employing simple techniques for future prediction by data estimation/extrapolation. Power control based on Kalman filtration is suggested for noise suppression. Numerical simulations demonstrate the potential of the proposed scheme.