Enhancing WSN Performance with DRL-CaCC: An Adaptive Cache-Aware Congestion Control Mechanism

Traditionally, rule-based is the predominant approach in Wireless Sensor Networks (WSN) communication protocols for managing caching and congestion control that relies on explicitly defined, unchanging models. A growing trend has recently been toward incorporating adaptive methods that leverage machine learning (ML) and deep learning (DL) during network congestion conditions. However, the design of an adaptive cache-aware congestion control mechanism in WSN using Deep Reinforcement Learning (DRL) is yet to be explored. Hence, this work developed a DRL-based adaptive cache-aware congestion control mechanism called DRL-CaCC to alleviate WSN during congestion scenarios. DRL-CaCC employs interme-diate caching parameters as its state space and adaptively moves the congestion window as its action space. It aims to find the optimal congestion window movement to avoid further network congestion while ensuring maximum cache utilization. Results show that DRL-CaCC achieved an average improvement gain from 10% to 40% compared to two other cache-based congestion control mechanisms in terms of cache utilization, throughput, and end-to-end delay metrics.