Context-Aware Prediction with Secure and Lightweight Cognitive Decision Model in Smart Cities

Cognitive networks with the integration of smart and physical devices are rapidly utilized for the development of smart cities. They are explored by many real-time applications such as smart homes, healthcare, safety systems, and other unpredictable environments to gather data and process network requests. However, due to the external conditions and inherent uncertainty of wireless systems, most of the existing approaches cannot cope with routing disturbances and timely delivery performance. Further, due to limited resources, the demand for a secure communication system raises another potential research challenge to protect sensitive data and maintain the integrity of the urban environment. This paper presents a secured decision-making model using reinforcement learning with the combination of blockchain to enhance the degree of trust and data protection. The proposed model increases the network efficiency for resource utilization and the management of communication devices with the alliance of security. It provides a reliable and more adaptive paradigm by exploring learning techniques for dealing with the intrinsic uncertainty and imprecision of cognitive systems. Also, the incorporation of blockchain technology reduces the risk of a single point of failure, malicious vulnerabilities, and data leakage, ultimately fostering trust for urban sensor applications. It validates the incoming routing links and identifies any communication fault incurred due to malicious interference. The proposed model is rigorously tested and verified using simulations and its significance has been proven for network metrics in comparison to existing solutions.