High Performance Fractal Compression for EEG Health Network Traffic

Abstract

Intelligent healthcare systems need recording, sending and treating large amounts of medical data produced from various kinds of medical sensors nodes. This is a challenging task and may lead to increase the network traffic, latency, and energy consumption. Therefore, it is important to apply a health data reduction mechanism for eliminating the data redundancy and saving the energy thus enhance the smart healthcare system performance. In this paper, a potential EEG Fractals compression model is proposed for reducing the transmitted EEG traffic from Patient Data Aggregator (PDA) to the destination (doctor, smart hospitals, emergency response, etc.). The proposed model supports EEG patient data communication and improve Wireless Body Sensor Network by reducing network traffic. Main model metrics are inspected as the determined Fractals Block Size found to be the vital playing role on producing higher Compression Ratio (CR) and drive the required Percentage Residual Difference (PRD). Both results and performance are compared with other techniques as the proposed model has outperformed them completely. The resultant CR can reach up to 160 and keep PRD less than 1.