User mobility profile prediction: An adaptive fuzzy inference approach

Predicting the probabilities that a mobile user will be active in other cells at future moments poses a significant technical challenge to network resource management in multimedia wireless communications. The probability information can be used to assist base stations to maintain a balance between guaranteeing quality of service (QoS) to mobile users and achieving maximum resource utilization. This paper proposes a novel adaptive fuzzy logic inference system to estimate and predict the probability information for direct sequence code division multiple access (DS/CDMA) wireless communications networks. The estimation is based on measured pilot signal strengths at the mobile user from a number of nearby base stations, and the prediction is obtained using the recursive least square (RLS) algorithm. Numerical results are presented to demonstrate the performance of the proposed technique under various path loss and channel shadowing conditions.