Adaptive Energy-aware Encoding for DWT-Based Wireless EEG Monitoring System

TL;DR

This paper introduces an adaptive encoding framework for wireless EEG systems that dynamically adjusts computational complexity to optimize energy consumption and signal quality under power constraints.

Contribution

It proposes a novel power-distortion-CR framework for DWT-based EEG encoding, enabling energy-aware reconfiguration of encoding parameters.

Findings

Enhanced reconstruction accuracy at fixed power levels

Effective complexity control matching energy constraints

Extended PRD-CR model for wireless EEG encoding

Abstract

Wireless Electroencephalography (EEG) tele-monitoring systems performing encoding and streaming over energy-hungry wireless channels are limited in energy supply. However, excessive power consumption either in encoding or radio channel may render some applications inapplicable. Hence, energy efficient methods are needed to improve such applications. In this work, an embedded EEG encoding system should be able to adjust its computational complexity, hence, energy consumption according to the channel variations. To analyze the distortion-compression ratio (PRD-CR) behavior of the wireless EEG system under energy constraints, both encoding and transmission power should be taken into consideration. In this paper, we propose a power-distortion- compression ratio (P-PRD-CR) framework, which extends the traditional PRD-CR to P-PRD-CR model. We analyze the computational complexity for a typical discrete wavelet transform (DWT)-based encoding system. Using our developed P-PRD-CR framework, the encoder effectively reconfigures the complexity control parameters to match the energy constraints while retaining maximum reconstruction quality. Results show that using the proposed framework, we can obtain higher reconstruction accuracy for the same power constrained-portable device.