Future Trends in Artificial Intelligence Driven Security
Utpal Ghosh, Uttam Kr. Mondal · 2025
Wireless acoustic sensor networks (WASNs) are essential for applications such as environmental monitoring, surveillance, and healthcare. However, securing audio data transmission remains challenging due to vulnerabilities in wireless communication. This chapter introduces an automated technique for secure audio transmission in WASNs using recursive key rotation (RKR) and a convolutional neural network (CNN) model. The system operates in two phases: first, SpeechRecognition converts audio to text (.txt/.docx) that reduces encoding time through CNN audio compression. Second, the text file is encrypted and decrypted using RKR at the sender and receiver ends, respectively, ensuring secure transmission over WASNs. The CNN decoder then decodes the compressed audio. RKR enhances security by dynamically rotating encryption keys, minimizing the risk of key compromise. Extensive simulations and real-world experiments demonstrate the technique&s;s effectiveness in protecting audio data from eavesdropping and unauthorized access. Results indicate that RKR significantly improves audio transmission security without significantly impacting network performance. Simulation outcomes highlight the technique&s;s efficacy, offering a secure and energy-efficient solution for WASN audio transmission. This research contributes to the advancement of secure communication protocols for WASNs, paving the way for further exploration in audio data security.