Internet of Things Implementation Framework for Marginal Land Agriculture in Indonesia: Challenge Analysis and Adaptive Strategies

Abstract

   The deployment of Internet of Things (IoT) in agriculture has enabled data-driven decision-making and resource optimization; however, its application in marginal agricultural environments remains limited by fragmented system design and lack of context-aware integration. This study proposes a taxonomy-driven conceptual framework for IoT implementation in marginal agriculture, addressing multi-dimensional constraints specific to the Indonesian context. A qualitative conceptual methodology is employed using a structured literature review based on a rapid review strategy. Relevant studies (2019–2026) are systematically selected and analyzed using directed thematic analysis across five dimensions: technical, communication, energy, environmental, and social. A taxonomy analysis is subsequently constructed to organize existing research, identify dominant design patterns, and expose structural gaps in current IoT architectures. The results reveal a strong bias toward isolated optimization of sensing and communication layers, with limited consideration of cross- layer integration, energy-aware operation, environmental adaptability, and user-centric design. These findings indicate the absence of a unified architectural model capable of addressing the constraints of marginal agricultural systems. To overcome these limitations, this study introduces a multi-layer IoT framework comprising sensing, communication, energy management, data analytics, and application layers. The framework emphasizes cross-layer interaction, adaptive resource management, and contextual decision support, enabling robust operation under limited infrastructure, variable environmental conditions, and low technology adoption settings. The novelty of this study lies in the integration of taxonomy analysis and multi-layer system design to construct a context- aware IoT framework that captures the complexity of marginal agricultural environments. This framework provides a scalable and adaptive reference model for future empirical validation and practical deployment of IoT-based smart agriculture systems in marginal environments.