AI Assisted Routing Optimization in Opportunistic IoT Networks using Machine Learning: A Comprehensive review on Protocols & Simulators

The concept of Opportunistic IoT Networks (Opp -IoT) has become a center of recent studies in wireless communication due to the ability to act successfully in the conditions that are unpredictable and lack traditional infrastructure. This paradigm also builds on a store-carry-forward framework, where a message is momentarily held by a node on a store, and it is sent on a trip in the ambulation with the node before it is sent out again to a node that is met. Although conceptually beautiful, this has significant challenges to energy usage, dependability, and latency, which the current protocol designs seek to address at least partially. This manuscript will offer a comprehensive overview of Opp -IoT routing methods and a machine-learning-based optimization framework to maximize the intelligence, speed and energy efficiency of routing decisions. As an extension of the ONE Simulator as a test platform, this investigation brings about a few important findings: supervised, reinforcement, and unsupervised learning techniques provide a productive dimension, and an adaptive feedback pipeline that incorporates these methodologies is the most promising direction of future Opp -IoT research.