Estimating the number of competing terminals by cost-reference particle filtering in non-saturated wireless-LAN

In IEEE 802.11 systems, the system throughput performance can be enhanced by optimizing the back-off window size based on the number of competing terminals n. Therefore, the information of n is crucial factor for optimal throughput performance in this system. However, estimating n is not an easy task because n varies with unknown statistical patterns. Most of previously proposed approaches additionally adopt a state variation detector in the algorithm, which incurs latency in tracking n. Therefore, we propose cost-reference particle filtering (CRPF) approach which is applied without an additional state variation detector. Consequently, a critical flaw of the detector, i.e. latency which causes highly degraded throughput performance, can be avoided. The proposed method promptly tracks varying n without any latency; therefore, it is highly efficient in non-saturated network condition. By computer simulations, we justify that CRPF is a new state of the art-approach for the investigated problem in terms of the tracking performance.