Interference Detection in IoT Environment
Neethu Suman, Mariya Sunny, Malavika S Menon, N Sneha Das, Niranjana Biju · 2024
In wireless communication, signal interference poses a significant challenge, particularly within the Internet of Things (IoT) environment where many devices communicate autonomously. This paper explores the application of deep learning algorithms to enhance interference detection, thereby improving the reliability of communication channels. The deep learning Bi-LSTM (Bidirectional Long Short-Term Memory) algorithm is used to analyze interference across four distinct wireless networks: Bluetooth, Wi-Fi, Zigbee, and GPS.Our study conducts a comparative analysis of these networks, focusing on their capacity to transmit signals with minimal interference. The findings reveal that Wi-Fi and Zigbee networks exhibit a remarkable 100% accuracy in signal transmission, indicating their superior performance in interference mitigation. This high level of accuracy is attributed to the utilization of random signals during the evaluation process, which may vary when applied to real-time signals. The research presented herein lays the groundwork for future exploration into more sophisticated neural network topologies and the integration of machine learning in IoT for robust interference detection.