A Fuzzy Logic-Based Adaptive Backoff Algorithm for OFDMA-Based Next-Gen WLANs

The Medium Access Control (MAC) layer of Wireless Local Area Networks (WLANs) undergoes several amendments since the birth of the legacy standard IEEE 802.11. The legacy standard IEEE 802.11 uses Distributed Coordination Function (DCF) for the MAC layer protocol. The DCF algorithm performs better in low dense WLAN networks where only one station (STA) can send data at one time. The selection of Contention Window (CW) plays a vital role in getting optimum WLANs performance. The legacy IEEE 802.11 and most of its amendments use Binary Exponential Backoff (BEB) algorithm to resolve collision in WLANs. In a conventional BEB algorithm, the size of CW is adjusted forward and backward i.e., the CW is initialized at minimum CW (CWmin) and for each collision the CW is exponentially incriminated to an optimal CW size where the packets can be transferred successfully. Hence the BEB mechanism is not suitable for dense environment as after every successful transmission the CW size is resumed to CWmin value where collision is obvious due to many contending stations. Finding the optimum CW is the key factor to maximize throughput and minimize delay. Moreover, Orthogonal Frequency Division Multiple Access (OFDMA) technology is used in next-generation IEEE 802.11ax WLANs to accommodate more active stations in a dense environment. To solve these issues, we proposed adaptive algorithm. In the proposed algorithm, First, we calculate the number of estimated active stations using a probabilistic approach. Second, we use Markov chain model to find the throughput ratio of the previous Contention Window (CW). We will consider the aforementioned parameters as input to the Fuzzy system, which adjust the optimum CW size to enhance throughput and minimize delay. The proposed approach enhances the throughput performances of IEEE 802.11ax and IEEE 802.11 medium access protocols up to 10% and 40% respectively.