Abstract: To resolve the performance decrease of iSLIP under burst traffic, a traffic adaptive algorithm named TA-iSLIP, which iterates the scheduling decision between slots, is proposed. This algorithm intellectually estimates the traffic type by using queue length, and then adopts different processing methods according to the different traffic types to make good use of the matched resource and decrease the system matching overhead. TA-iSLIP is described and its performance is compared with those of iSLIP and FIRM. Simulation results show that TA-iSLIP achieves high performance under both of the uniform and non-uniform traffic,