A Comparative Study Using DCT, Delta Modulation, and Double Shift Coding for Compressing Electroencephalogram Data

Abstract

      Storing, transferring, and processing high-dimensional electroencephalogram (EGG) signals is a critical challenge. The goal of EEG compression is to remove redundant data in EEG signals. Medical signals like EEG must be of high quality for medical diagnosis. This paper uses a compression system with near-zero Mean Squared Error (MSE) based on Discrete Cosine Transform (DCT) and double shift coding for fast and efficient EEG data compression. This paper investigates and compares the use or non-use of delta modulation, which is applied to the transformed and quantized input signal. Double shift coding is applied after mapping the output to positive as a final step. The system performance is tested using EEG data files from the CHB-MIT Scalp EEG Database. Compression Ratio (CR) is used to evaluate the compression system performance. The results are encouraging when compared with previous works on the same data samples.