ASSOCIATE: A simulator for assessing soft security in the Cognitive Internet of Things

Abstract

By augmenting the Internet of Things (IoT) with Artificial Intelligence (AI), the Cognitive Internet of Things (CIoT) has emerged, where intelligent objects autonomously sense, perceive, make decisions, and collaborate with minimal initial knowledge. Ensuring trust and security in such a cognitive society poses significant challenges. Soft security approaches, which leverage behavioral norms, have arisen as a promising solution to address these challenges. In this paper, we introduce ASSOCIATE, a modular and multi-layered tool designed for simulating and evaluating soft security models in the CIoT domain. ASSOCIATE applies a trust model to CIoT simulations and assesses the society’s soft security. The evaluation process encompasses society preparation, generation, trust model preparation, execution, and report generation. Input configuration files in the JSON format define simulation parameters, allowing the simulator to generate the requested society and conduct simulations. Evaluation results are presented through charts and graphs and saved in image and CSV formats. We demonstrate the tool’s functionality by enhancing the Santander dataset using the society generation module and analyzing various trust profiles. The proposed tool provides a valuable means of assessing the efficacy of soft security models in CIoT simulations, facilitating a comprehensive analysis of trust and security in CIoT systems.