A power allocation algorithm for throughput maximization in mobile networks

The paper focuses on competitively optimal power-control and signal-shaping for "ad-hoc" networks composed of multiple-antenna noncooperative transmit/receive terminals affected by spatially colored multi-access interference (MAI). The target is the competitive maximization of the information throughput (measured in bits/slot) sustained by each link active over the network. For this purpose, the MAI-impaired network is modeled as a noncooperative strategic game, and sufficient conditions for the existence and uniqueness of the Nash equilibrium are provided. We develop fully distributed and scalable power-control and signal-shaping algorithms allowing the implementation of asynchronous space-division multiple access strategies (SDMACSs) able to guarantee the competitive maximization of the users\´ throughput under both best effort and contracted QoS access policies. We give evidence that the developed SDMACSs outperform (in terms of aggregate throughput) the conventional centralized ones (such as TDMA/FDMA/CDMA), especially in operating scenarios affected by strong MAI. We study the convergence property of the presented SDMACSs and show that, when the throughput set requested by the users is not achievable by the network, then the developed SDMACSs are able to move the working point of the system to the nearest one sustainable by the network. By exploiting the distributed feature of the presented SDMACSs, we propose two connection admission procedures (CAPs) able to optimize (in a competitive sense) the tradeoff between aggregate networking throughput and connection requirements advanced by the users.