Optimizing statistical uplink admission control for W-CDMA

This paper presents an optimized uplink admission control for W-CDMA. In order to ensure that all the accepted requests achieve their QoS requirements in the planned coverage area and taking into account the power constraints, the algorithm controls the uplink load factor, which is statistically estimated. The key parameters that have influence over this estimation are presented as well as their adequate setting in order to achieve a high admission probability while keeping QoS guarantees. Specifically, some innovative modifications with respect to state of the art admission control approaches are proposed and it is shown that significant performance improvements may follow. In particular, it is shown that a traffic averaging method is adequate to smooth traffic fluctuations in the estimation that would lead to unnecessary rejections. Similarly, it is shown that users in soft handover should be accounted for in the estimation with a reduction factor. Finally, it is also shown that the proper setting of the f factor that accounts for the intercell to intracell interference ratio should not take into account the average factor but another more representative statistic like the 50% percentile.