This is the authors' abstract. We don't add key points for this paper.
Wireless sensor networks (WSNs) play a backbone of most modern Internet‐of‐things (IoT) applications such as smart city, smart surveillance, and smart healthcare; in such applications, the sensor usually collects information either through time or event‐driven in energy efficient and reliable manner. In this work, heterogeneous sensors are used where two sensors are time driven such as temperature and humidity and one sensor is event driven such as motion sensor. Owing to node mobility and along with time‐driven sensory information, routing becomes extremely challenging; the current method uses the clustering technique to improve energy efficiency; further, existing methods have done multihop multipath‐based routing optimization; however, these methods suffer from hotspot problems; thus, unequal clustering have been explored with good effect; however, the current method failed to balance load among best cluster heads. In addressing the research issues, this paper introduces a Multiagent Heuristic and Knowledge‐Driven Routing (MAHKDR) algorithm. The MAHKDR algorithm ensures that there is better load balancing with minimal energy consumption. The MAHKDR first introduces a novel unequal clustering assuring network coverage; second, multiobjective‐based group head selection metrics are introduced. Third, multihop multipath‐based routing and load balancing algorithm is designed. The multiagent deep reinforcement learning (MA‐DRL) is introduced to optimize heuristic function to attain optimal routing performance. The simulation result shows that MAHKDRS improves network lifetime by 84.26%, 62.28%, and 11.89%; improves throughput by 48.42%, 57.25%, and 14.12%; reduces routing overhead by 21.04%, 32.05%, and 42.65%; reduces communication overhead by 31.86%, 30.17%, and 16.12%; and reduces delay by 58.20%, 52.65%, and 9.65% in comparison with EEACR (energy‐efficient adaptive clustering routing), HTOM (hierarchical traffic offloading mechanism), and MADRL (multiagent deep reinforcement learning)–based routing algorithm, respectively.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Padmini et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: