Can Cross-Layer Intrusion Detection Secure Agriculture 4.0 Systems?

Abstract

Agriculture 4.0 leverages the Internet of Things (IoT), advanced communication technologies, and artificial intelligence (AI) to improve agriculture efficiency, optimize resource utilization, and promote sustainable crop production. IoT devices continuously monitor environmental conditions and transmit data to enable real-time decision-making in agricultural operations. However, the increased connectivity also expands the attack surface, exposing critical agricultural infrastructures to diverse cyber threats. Current intrusion detection systems (IDSs) primarily focus on detecting external attacks at isolated layers, often neglecting challenges related to data integrity, such as silent data corruption, sensor inconsistencies, and cross-layer anomalies. In this paper, we review existing IDS approaches and emphasize the need for a comprehensive system design methodology that addresses these challenges across all layers of Agriculture 4.0 systems. We propose DIVA-IDS, a cross-layer framework that integrates data integrity validation and anomaly detection to provide robust security for agricultural IoT environments. The framework aims to ensure reliable and secure data transmission, supporting the coexistence of various agricultural applications with differing security priorities.