Encoding of sampled -data systems: Applications to finite wordlength controller implementation and adaptive sampling of atomic force microscopy
Abstract
As the state-to-the-art mechatronic technology drives the ongoing revolution in consumer electronics and in many other domains, the issues of efficient data transmission and system representation have become important topics. This research focuses on two issues of encoding systems: the finite wordlength effect in discrete-time controller synthesis and the development of fast imaging algorithm for atomic force microscopy. There are two major considerations when implementing discrete-time controllers: the required sample rate and a suitable wordlength. Higher sample rate is often preferred since it implies a potential higher bandwidth and performance. It is also well known that the upper bound of the achievable sample rate may be significantly lower than that limited by hardware when a discrete-time controller is implemented with finite precision. In this research, we will investigate this apparent coupling between finite precision implementation and the limit on the sample rate. We will propose conservative bounds that relate the controller characteristics and the minimum wordlength to the maximum achievable sample rate based on different discrete-time controller realizations. In typical atomic force microscope experiment, a three dimensional image of a substrate is obtained. With the total number of samples remains constant, there is a trade-off between the size of the scanned image and the resolution. For a given scanning mechanism, the time needed to image an area depends mainly on the number of samples and the size of the image. It is therefore desirable to improve the imaging speed with limited impact to the effective resolution of the portion of the sample that is of interested. By utilizing an adaptive sampling scheme with fractal compression technique, we have demonstrated that the number of the required samples can be significantly reduced with minimal impact to the effective image quality.
Degree
Ph.D.
Advisors
Chiu, Purdue University.
Subject Area
Mechanical engineering|Materials science|Condensation
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.