Abstract
In code division multiple access (CDMA) system, power control technique is very important to solve the near-far problems, to maintain the acceptable signal to interference ratios (SIRs) for all users, and to extend the system lifetime. The carrier-to-interference ratio (CIR) denotes the ratio of the desired signal power to the total co-channel interference (CCI) power. According to SIR requirement and data rate, system can know the user’s CIR requirement. Thus, CIR is generally used to represent the link quality in mobile cellular systems. Unfortunately, as the mobile moves, the propagation environment changes accordingly and so does the link performance. Shadowing effect causes the large variations in receiving signal strength and total co-channel interference ratio (CCI). Therefore, the CIR is also changed randomly as the mobile moves, and using power control technique to guarantee the mobile’s CIR preceding the requirement in all the way is very difficult. Accurate prediction of the CIR variation profits analyzing the behavior of link quality. In [2], regardless of multiple access interference (MAI) from the serving cell, the spatial correlation of the CIR received by the moving mobile in the downlink environment can be modeled as a simple Gaussian-Markov stochastic model. Furthermore, [3] proposed a methodology to characterize the spatial correlation and link variation of the CIR involved with MAI. In this research, by using the methodology of [3] to predict the next time CIR variation, we propose the CIR-prediction-based power control scheme to reserve enough power against the probable channel variation in advance, and accomplish the desired purpose that guaranteeing the link quality of a traveling mobile station with a required reliability in all the way. Moreover, we simplify our proposed scheme, and compare them with the general non-prediction-based power control schemes in the simulation. This research can be extensively applied in adaptive power and rate allocation, downlink capacity analysis, performance simulations, and so on.