Machine Learning and Bio-Inspired Deep Learning Approaches for Energy-Efficient and Secure Clustering and Routing in Wireless Sensor Networks:A Survey
Keywords:
wireless sensor networks; cluster-head election; energy-hole problem; hotspot mitigation; ant colony optimization; grey wolf optimizer; reinforcement learning; deep Q-network; LSTM; autoencoder; trust management; intrusion detection.Abstract
Wireless Sensor Networks (WSNs) remain constrained by the finite battery energy of their constituent nodes, which makes energy-aware clustering and routing the central design problem of the field. Since the introduction of the Low-Energy Adaptive Clustering Hierarchy (LEACH) protocol, a large body of research has sought to improve cluster-head (CH) election and multi-hop routing along four broad lines: (1) geometry- and threshold based unequal clustering to counter the "hotspot" or energy-hole problem near the base station; (2) machine learning and metaheuristic CH election using k-means, fuzzy logic, particle swarm optimization (PSO), genetic algorithms (GA), and artificial neural networks; (3) bio-inspired swarm-intelligence routing using Ant Colony Optimization (ACO), the Grey Wolf Optimizer (GWO), and hybrid combinations of these and other nature inspired metaheuristics; and (4) reinforcement learning (RL) and deep learning approaches, including tabular Q learning, Deep Q-Networks (DQN), Long Short-Term Memory (LSTM) energy prediction, and autoencoder based anomaly detection for security. This survey reviews and synthesizes 71 peer-reviewed and preprint sources across these four lines, organizes them into a coherent taxonomy, and builds comparative tables of their clustering/routing mechanisms, evaluation scale, and reported performance gains. The review is deliberately structured to support a three-phase enhancement of an existing Energy-Efficient Machine-Learning-based Clustering and Routing (EEMLCR) protocol: a hotspot-aware adaptive CH scheduling phase, a bio-inspired ACO-GWO multi-objective routing phase, and a unified deep-learning phase combining LSTM-based proactive energy prediction, DQN-based routing, and autoencoder/trust-based security. The survey concludes by identifying three concrete gaps in the existing literature — the absence of a threshold-driven, TDMA-integrated hotspot recovery mechanism; the lack of a systematic ACO-explores/GWO-exploits routing pipeline validated at scale; and the absence of a single framework unifying proactive LSTM prediction, DQN routing, and trust cross-validated anomaly detection — that motivate the proposed enhancements.
Downloads
Downloads
Published
Issue
Section
License
You are free to:
- Share — copy and redistribute the material in any medium or format for any purpose, even commercially.
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
- The licensor cannot revoke these freedoms as long as you follow the license terms.
Under the following terms:
- Attribution — You must give appropriate credit , provide a link to the license, and indicate if changes were made . You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
Notices:
You do not have to comply with the license for elements of the material in the public domain or where your use is permitted by an applicable exception or limitation .
No warranties are given. The license may not give you all of the permissions necessary for your intended use. For example, other rights such as publicity, privacy, or moral rights may limit how you use the material.

