Power control in WCDMA networks: a queue-based approach

In modern cellular communication systems, power control plays a fundamental role for efficient resource utilization. In such systems, in fact, many users transmit over the same radio channel using the same frequency band and time slots so that the signal of an individual user becomes interference for the other users. Hence the transmission power levels need to be smartly manipulated so as to achieve an adequate quality of service for as many users as possible and, thus, an efficient network utilization. Conventional power control algorithms adopt the Signal-to-Interference-plus-Noise Ratio (SINR) as controlled variable and neglect the important effects of the manipulated control variables (transmission powers) and of the retransmission mechanism on the queueing dynamics. In this paper, we pursue a different queue-based approach which takes into account the queuing dynamics and adopts the queue size as controlled variable. In this framework, the power control problem is formulated as an optimal control problem with a criterion trading off high performance (low queue sizes) vs. low costs (low transmision powers). In particular, a novel queue-based power control algorithm with low on-line computational burden is proposed and its performance is evaluated via simulation experiments. The simulation results exhibit an improved behavior of the proposed queue-based algorithm compared to conventional SINR-based algorithms.