Efficient EEG Data Compression Technique for Internet of Health Things Networks

The number of patients with chronic conditions that require ongoing monitoring has risen dramatically in recent years, prompting scientists to build scalable health applications that can work remotely. Nonetheless, the quantity of real-time data that can be communicated across current dynamic networks with limited bandwidth, end-to-end latency, and transmission power restricts data transmission efficiency. The vital signs reduction at the transmitter side using data reduction techniques offers an effective edge-based strategy that considerably decreases the volume of data transmission, which is motivated by the Internet of Health Things (IoHT) Networks. However, a new issue arises, the ability to receive data from the server with an allowable distortion rate. This paper suggested an Efficient EEG Data Compression Technique (EDaCoT) for IoHT Networks. It compresses the patient’s EEG data at the edge node before transferring it to the cloud data center. At the edge gateway, the suggested EDaCoT combines two powerful techniques: clustering followed by lossless encoding. DBSCAN clustering splits a vast quantity of EEG data into small sets of data that are identical (or near). The RLE is used to encode the EEG data of different constructed sets into a single file. The produced file is then encoded using Huffman encoding and uploaded to the cloud via the edge gateway. Several simulation experiments are applied to evaluate the proposed EDaCoT technique. The results show that the proposed EDaCoT provides better results compared with other approaches in terms of compression ratio, compression time, and decompression time.