Improved Piecewise Constant Approximation Method for Compressing Data Streams

Stream data, obtained from the various sensors and devices with high speed, used daily, are indispensable parts in our day-to-day life. There is an immense necessity to store these data comprising of large volume. Data compression is one of the effective ways of storing data efficiently. In this paper, different data compression techniques such as PCA, APCA have been improved and thus we propose a new hybrid model. In the proposed model, the window size is fixed but data can be compressed efficiently than PCA. Besides, the delay time is minimized compared to the other existing techniques since the model is developed by considering both the delay time and several data points to be compressed. Our experiment shows that the proposed method performs twice as better than PCA in terms of compression and minimizes RMSE by 10%-25% compared to APCA method.