Robust multichannel EEG signals compression model based on hybridization technique

Abstract

Tele monitoring of Electroencephalogram (EEG) via wireless is very critical as EEG. EEG medically is a tool test used to estimate the electrical activity of the brain. There are many channels through which EEG signals are recorded consistently and with high accuracy. So the size of these data is constantly increasing, need large storage area and a bandwidth for the transmission of the EEG signal remotely. In last decade, the EEG signal processing grew up, additionally; storing and transmitting EEG signal data requirement is constantly in-creasing. This article includes the analysis method of an EEG compression and de-compression. This method is evaluated on the basis of various compression and parameters quality such as CR (compression ratio), SNR (Signal to noise ratio), PRD (percent-root-mean-square-difference), quality score (QS), etc. The steps of EEG compression are pass through many stages: 1. Preprocessing and after that classification. 2. Linear transformation, and [3]. Entropy coding. The EEG compression is specified during processing and coding algo-rithm for each of the steps. The decompression process is the reverse of the compression process, reconstructs the EEG original signals by using lossy algorithm but with the simple loss of significant information. The proposed compression method is a bright step in the compression field where getting a high compression ratio