Scheduling bursty data at WCDMA downlink using fuzzy inference

One of the major concerns for wideband code division multiple access (WCDMA) wireless network dimensioning is to improve the downlink capacity to accommodate more data traffic. Due to the "soft capacity" of CDMA system, there exists a conflict between the maximum throughput and the fairness for data users. To deal with the tradeoff, we propose a scheduling scheme using fuzzy inference involving factors of dynamics of traffic and wireless channels, and preferences of the decision maker. An exemplar design is given to improve the data throughput. Simulation results show that the fuzzy scheduler has superior performance to regular rate allocation schemes, and it increases the capacity of WCDMA that is elastic in nature. Moreover, the, fuzzy approach demonstrates its robustness, flexibility, and simplicity in the decision-making process when human preference is considered.