Adaptive contextual memory network for enhanced communication and efficiency in the internet of underwater things

The Internet of Underwater Things (IoUT) is becoming increasingly critical in navigation and the investigation of aquatic environments, providing solutions for real-time data acquisition and interaction within underwater environments. Despite that, implementing IoUT systems faces challenges such as signal attenuation, high latency, low bandwidth, and high energy consumption. To this end, this paper presents the Adaptive Contextual Memory Network (ACMN), a novel and advanced framework for the above challenges. Tactile interface ACMN responds to changes in the underwater conditions, which increases reliability and conserves energy. In addition to ACMN, an Adaptive Modulation Optimization (AMO) algorithm is introduced to continuously adjust the modulation to minimize signal degradation and prevent distortion of the transmitted data. Moreover, to enhance the adaptability of the routing decision based on feedback from the real network environment, Energy-Aware Reinforcement Learning (EARL) is proposed in this study. Together, these provide a strong and autonomous architecture of IoUT, paving the way for molecular improvements in marine biology, ocean surveillance, and underwater infrastructure.