Wireless sensor networks have become the nervous system of modern precision agriculture, streaming real-time measurements of soil moisture, temperature, humidity, nutrient levels and crop health from fields to cloud platforms. But the sensor nodes that gather this data are battery-powered devices scattered across vast, weather-beaten terrain, where recharging or replacing batteries is rarely practical. A new systematic review published in Discover Networks examines how researchers have tackled this energy bottleneck, systematically compiling and comparing energy-efficient routing protocols designed specifically for wireless sensor networks in precision and smart agriculture.
The review, led by Tumpa Pal, Sohail Saif, Ramesh Saha and Suparna Biswas, follows the PRISMA framework for systematic reviews. The authors initially identified 250 research studies from major databases including IEEE, Springer, ScienceDirect, Wiley Online Library and MDPI. After removing duplicates and screening titles, 150 articles remained. A further 50 papers were excluded as irrelevant to the research questions, and after a structured quality assessment, 43 high-quality research articles published between 2016 and 2025 were selected for full-text review and detailed analysis.
The selected routing frameworks were classified into six categories: clustering-based, optimization-based, fuzzy-based, machine learning-based, security-aware and hybrid approaches. The count of studies in each category reveals where research attention has concentrated. Clustering algorithms dominate the field with 15 papers, reflecting their widespread use and well-established methodologies. Optimization-based approaches follow with 8 papers, while fuzzy-based, machine learning-based, security-aware and hybrid methods each account for 5 papers. The authors also observed that research activity increased significantly after 2020, with the highest number of articles published in 2020, 2021 and 2022, whereas before 2020 publication frequency ranged from only one to three papers annually.
Clustering-based protocols rely on dividing the network into groups and periodically selecting cluster heads that aggregate and relay data, reducing communication distances and balancing energy consumption across nodes. Among the reviewed examples, a multi-tier dynamic energy-based clustering protocol for smart agriculture reported simulation results showing energy consumption reductions of 30 to 40 percent, network lifetime increases of 50 percent, and packet delivery ratios of 90 to 98 percent compared with traditional routing protocols. Other clustering designs described in the review include multi-threshold segmentation-based clustering, grid-based data aggregation for IoT routing, gateway-based centroid selection, and neighborhood-aware clustering that avoids isolated nodes. The review concludes that clustering and optimization-based frameworks provide the greatest improvements in network lifetime and data delivery, though their performance can degrade in large-scale or highly dynamic agricultural conditions due to uneven energy usage and hotspot problems.
Optimization-based protocols take a different route, using metaheuristic algorithms to search for optimal cluster heads and routing paths. Reviewed examples include a lightweight clustering protocol that applies a bacterial foraging algorithm to cluster head selection using residual energy, received signal strength, node degree and bandwidth rate; a hybrid of ant colony optimization and cuckoo search for cluster head choice; and a protocol combining particle swarm optimization with the wild horse optimizer. These approaches improved routing efficiency, load balancing and network lifetime in the reported simulations, but the review notes that their added computational complexity makes them less suitable for severely resource-constrained nodes.
Fuzzy and machine learning-based protocols introduce more adaptive decision-making. Fuzzy systems evaluate multiple parameters simultaneously, such as residual energy, distance, node density and centrality, through linguistic rules and membership functions, and the reviewed fuzzy protocols showed strong performance in coverage, stability and throughput, although effectiveness depends heavily on how membership functions and rule sets are designed. Machine learning approaches, including deep learning-based predictive analytics, Q-learning and multi-agent reinforcement learning, allow nodes to select routing paths dynamically based on remaining energy, distance to the sink, hop count and queue length. The review found these methods offer better scalability, adaptability and intelligent routing under dynamic agricultural scenarios, but their training requirements and computational and memory demands can exceed the capabilities of low-power sensor hardware.
Security-aware protocols address a growing concern as agricultural systems connect to cloud and IoT platforms vulnerable to data tampering, spoofing, denial-of-service attacks and false data injection. The five reviewed security-focused studies include lightweight cryptographic techniques against denial-of-service attacks, anomaly-based intrusion detection, a cluster head management scheme called AgroKy, a fuzzy and reputation-based trust management protocol named AgriTrust, and a blockchain-and-edge-computing framework called AgriChainSync. These frameworks improve reliability and data integrity under malicious attacks, but the review emphasizes that security mechanisms increase computational overhead, routing overhead, energy consumption and processing delay, and that most security solutions in the literature have been implemented independently of routing performance.
A key finding of the review is that no single protocol family consistently outperforms across all quality of service metrics, which include energy consumption, network lifetime, packet delivery ratio, throughput, latency, routing overhead, coverage ratio and cluster stability. Because the reviewed studies used different simulation tools, network topologies, node densities, traffic patterns and energy models, the authors adopted a relative rather than quantitative cross-study comparison, rating performance as high above 70 percent, medium between 40 and 70 percent, and low below 40 percent. Hybrid protocols, which combine clustering, optimization, fuzzy logic and machine learning techniques, demonstrated the most balanced performance across quality of service dimensions, though they carry higher implementation and computational costs. From an application perspective, the authors suggest clustering and optimization-based protocols suit large-scale environmental monitoring, machine learning approaches fit predictive irrigation and decision support, and security-aware protocols are essential for cloud-connected smart farming.
The review also confronts the gap between simulation and real deployment. Most protocols in the literature were evaluated under controlled simulation assumptions, while practical deployments face hardware costs, battery depletion, hardware failures, calibration needs, device and network heterogeneity, vegetation and terrain-induced packet loss, and limited maintenance access across large fields. The authors call for field deployments, testbed experiments and long-term validation studies to bridge the distance between theoretical performance and practical implementation, and they identify future directions including federated learning for privacy-preserving distributed training, unmanned aerial vehicles as relay nodes or mobile sinks, tiny machine learning for lightweight adaptive clustering, digital twins, renewable-energy-powered nodes, 6G-enabled agricultural IoT, explainable AI and zero-trust security frameworks, arguing that integrating these technologies into unified routing strategies will be necessary to meet the energy, security, reliability and real-time demands of large-scale smart agriculture.
Subject of Research: A systematic review of energy-efficient routing protocols for wireless sensor networks in precision and smart agriculture
Article Title: Energy efficient routing protocols in WSN for precision and smart agriculture
Article References: Pal, T., Saif, S., Saha, R., & Biswas, S. (2026). Energy efficient routing protocols in WSN for precision and smart agriculture. Discover Networks, 2(1), Article 31. https://doi.org/10.1007/s44354-026-00045-y
Image Credits: AI Generated
DOI: 10.1007/s44354-026-00045-y
Keywords: wireless sensor networks, precision agriculture, smart farming, energy-efficient routing, PRISMA systematic review, clustering protocols, optimization algorithms, fuzzy logic, machine learning, network lifetime, packet delivery ratio, IoT security