9353204 Park The proposed research will develop a revolutionary and practical data compression technique by using least squares Autoregressive Moving Average (ARMA) modeling of a signal. Because of computational complexity and numerical instability, least squares ARMA modeling has been applied infrequently for data compression. In data compression, a stream of analog or very high rate discrete data is converted into a stream of relatively low rate data for transmission over a digital communication link or for storage in digital memory. After developing the data compression technique, a VLSI encoder will be designed and tested for the proposed data compression. Parallel processing will be used for fast computation. Then, the functional and layout levels of the new VLSI encoder chip will be designed, simulated, and verified by using computer aided design (CAD) tool. ***
