A Decentralized Approach to Smart Home Security: Blockchain With Red-Tailed Hawk-Enabled Deep Learning

Abstract

The fast development of smart home devices and the Internet of Things (IoTs) presents unprecedented accessibility into our day-to-day lives; however, it has also increased major problems regarding security and privacy. A smart home network is a vital element of modern home automation systems, enabling the interconnectivity and control of different smart devices. These networks allow homeowners to remotely control lighting, security, temperature, and entertainment systems via voice commands or smartphones. These offer energy efficiency, convenience, and improved security by permitting residents to monitor and modify their living surroundings. Safeguarding the flexibility of smart home networks against cyberattacks and unauthorized access is important to comprehending the maximum ability of smart living while retaining data integrity and privacy of connected devices. This research develops the Blockchain with Red-Tailed Hawk Algorithm-Enabled Deep Learning (BC-RTHADL) model, aimed to strengthen the safety of smart home systems. BC-RTHADL integrates the safety features of blockchain with a strong malicious action recognition procedure. The blockchain module certifies immutability, transparency, and decentralization, donating to a safe smart home atmosphere. The malicious action detection influences the Red-Tailed Hawk Algorithm for feature selection and an ensemble of Extreme Learning Machine (ELM), Gated Recurrent Unit (GRU), and Long Short-Term Memory (LSTM) techniques for precise recognition. The Equilibrium Optimizer algorithm enhances parameters for improved effectiveness. Complete tests show the greater performance of BC-RTHADL across numerous metrics, reaffirming its promising potential in safeguarding smart home networks.

Keywords

• Blockchain
• deep learning
• ensemble learning
• red-tailed hawk algorithm
• feature selection