Resource management in CDMA networks based on approximate dynamic programming

In this paper a power and rate control scheme for downlink packet transmission in CDMA networks is proposed. Under the assumption of stochastic packet arrivals and channel states the base station transmits to multiple mobile user at any time instant within rate and power capacity constraints. The objective is to maximize system throughput, while taking into account the queue length distribution over a time horizon. We are interested in optimal rate allocation policies over time and thus we formulate the problem as a discrete stochastic dynamic program. This dynamic program (DP) is exponentially complex in the number of users which renders it impractical and therefore we use an approximate dynamic programming algorithm to obtain in real time sub-optimal rate allocation policies. Numerical results reveal that the proposed algorithm increased the performance (in terms of a number of different measured parameters such as average queue size) of at least 3.5 times compared to a number of different baseline greedy heuristics