Abstract

Integrated communication networks (CN) have proven successful in tracking environmental activities, wherein several sensors are installed throughout diverse surroundings to gather data or observe certain events. CNs, comprising several interacting detectors, have proven effective in various applications by transmitting data via diverse transmission methods inside the communication system. The erratic and constantly changing surroundings necessitate conventional CNs to engage in frequent conversations to disseminate the latest data, potentially incurring substantial connection expenses through joint data gathering and dissemination. High-frequency communications are prone to failure due to the extensive distance of data transfer. This research presents a unique methodology for multi-sensor environmental monitoring networks utilizing autonomous systems. The transmission system can mitigate elevated communication costs and Single Point of Failing (SPOF) challenges by employing a decentralized method that facilitates in-network processing. The methodology employs Boolean systems, enabling a straightforward verification process while preserving essential details about the dynamics of the communication system. The methodology further simplifies the data collection process and employs a Reinforcement Learning (RL) technique to forecast future events inside the surroundings by recognizing patterns.