Energy-efficient threshold-based reinforcement learning for WSN routing

Chaudhari, Archana and Abdullah, Masuk and Deshpande, Vivek and Midhunchakkaravarthy, Divya (2026) Energy-efficient threshold-based reinforcement learning for WSN routing. POLLACK PERIODICA: AN INTERNATIONAL JOURNAL FOR ENGINEERING AND INFORMATION SCIENCES, 21 (2). pp. 64-70. ISSN 1788-1994

Preview

Text 606-article-p64.pdf - Published Version Available under License Creative Commons Attribution. Download (1MB) | Preview

Abstract

The proposed method uses a threshold-based reinforcement learning algorithm, Q-learning, for efficient routing in wireless sensor networks. It penalizes node energy, hop count, and distance to sink using three key thresholds to select the next best forwarder. This prevents self-loops and ensures packet delivery by avoiding low energy or high hop count nodes. The combination of reinforcement learning and key threshold values allows intelligent data packet routing, handling congestion near sink nodes, and reliable transmission. Experimental results show a 35% enhanced network lifetime compared to the state-of-the-art algorithm.

Actions (login required)

Edit Item