Performance improvement of WLAN, LR-WPAN, WBAN using genetic fuzzy mechanism

IEEE standards play a crucial role in wireless technology due to their low cost, power, complexity, and high throughput. However, challenges such as collision avoidance and node energy consumption remain. Nodes communicate with coordinators on a first-come, first-served basis, which increases the chances of collisions and high energy consumption. Wireless technology advancements emphasize the need to overcome these obstacles. We developed a Genetic Fuzzy mechanism by combining Genetic Algorithms and Fuzzy Logic Controllers (FLC) to improve the performance of three IEEE standards: WLAN, LR-WPAN, and WBAN. The GA was used to optimize FLC configurations for enhanced system efficiency. Three approaches were proposed: GF-CWO for WLAN: This approach employs three algorithms to optimize the Binary Exponential Backoff (BEB) and Channel Status-based Sliding Contention Window (CS-SCW) mechanisms. GFCO for LR-WPAN: This method comprises five algorithms and uses Random Exponential Backoff (REB) and Survivability Aware Channel Allocation (SACA) algorithms integrated with fuzzylite in NS-3.20. Simulations showed GFCO outperformed SACA, boosting throughput, SR, PLR, and PD by 15.11%, 3.11%, 3.11%, and 5.52% in scenario-I, and by 12.06%, 9.0%, 9.0%, and 2.23% in scenario-II. GFuCWO for WBAN: This technique involves three algorithms to optimize the contention window using Alternate Binary Exponential Backoff (ABEB). Implemented in Castalia OMNeT++, it demonstrated improved results for the enhanced CSMA/CA method. The GFuCWO technique demonstrated superior performance in PDR, PLR, and E2D, with average enhancements of 4%–11% and 3%–13%, respectively. A comparative analysis of these methods highlighted their effectiveness in addressing network challenges and improving system performance.