Effect of Power Control on Performance of Users in an Interference-Limited Network with Unsaturated Traffic
Min Sheng,Wanguo Jiao,Xijun Wang,Guoqing Liu
DOI: https://doi.org/10.1109/tvt.2016.2584085
IF: 6.8
2017-01-01
IEEE Transactions on Vehicular Technology
Abstract:As the base of quality-of-service (QoS) guarantee and system optimizing, performance analysis is very important for interference-limited networks. The performance of a user is closely related to its service rate. However, in interference-limited networks, the transmissions of users interfere with each other, and the power control and unsaturated traffic make the interference relationship among users more complicated. Hence, the service rates of users are correlated with each other and are difficult to derive. According to the relation between the queue length and the service rate, we propose an N-dimensional Markov chain, which models queue states of users, to study the performance of an interference-limited system. By calculating the conditional service rate, which is only related to the channel quality given the power policy, the transition matrix is derived. By using the transition matrix, we solve the Markov chain and obtain the stationary probability of the queue length of each user, which enables the calculations of other performance metrics, i.e., delay and packet overflow probability, and so on. Furthermore, to reduce the computational complexity, we decompose the proposed N-dimensional Markov chain. Then, based on the decomposition, we propose an iterative algorithm. The comparisons between simulation results and analytical results under different scenarios verify that our method can provide good estimations of each user's performance. The results derived by our method can determine the condition under which the power control algorithm for the saturated system is suitable in unsaturated networks and in which the new power policy is needed.