Recent Advances in Optimized Riemannian Residual Neural Networks: An Advanced Energy-Efficient Environmental Monitoring in Precision Agriculture using LoRa-based Wireless Sensor Networks: A Systematic Review

Abstract

Precision agriculture has emerged as a transformative approach to enhance crop productivity, resource utilization, and environmental sustainability. The integration of artificial intelligence, particularly deep learning models such as Riemannian residual neural networks, with Internet of Things (IoT)-enabled wireless sensor networks has significantly improved environmental monitoring systems. This study presents a systematic review of recent advances in optimized Riemannian residual neural networks combined with LoRa-based wireless sensor networks (WSNs) for energy-efficient environmental monitoring in precision agriculture. Recent studies highlight that LoRa-based WSNs provide long-range, low-power communication capabilities, making them ideal for large-scale agricultural monitoring systems. Additionally, the incorporation of optimized neural network architectures enhances predictive accuracy and data processing efficiency. Riemannian residual neural networks, in particular, enable learning over complex manifold-structured data, improving model performance in environmental sensing applications. Furthermore, IoT-based sensor networks facilitate real-time monitoring of parameters such as soil moisture, temperature, and nutrient levels, contributing to improved crop management and yield optimization. Despite these advancements, challenges such as energy consumption, network scalability, data heterogeneity, and model complexity persist. This review analyses current trends, identifies key challenges, and outlines future research directions for developing intelligent, energy-efficient precision agriculture systems.