Design of digital filters for electronic instrumentation: Dissertation

This thesis considers the design of digital filters for electronic instrumentation.The overall design and realization problem addressed in this thesis can be divided into five subproblems: sampling and reconstruction, digital equalization, digital filtering and noise rejection, sampling rate alteration, and digital coding and transmission.The basic discriminating feature compared with other related studies is the requirement of the preservation of the time domain characteristics.Novel design methods are presented both for recursive and nonrecursive digital filters covering a wide range of electronic instrumentation.Medical signal processing has been the major field of application in this thesis requiring most delicate signal processing to preserve the clinically important time domain characteristics. However, the design problems considered and results achieved are general and can be applied to any signal processing task bearing similar principal constraints.The cost effective implementation of the required digital filters and the restrictions of medical routine impose stringent constrains upon the filter characteristics.To satisfy the various requirements, new filter structures and design algorithms had to be developed.Since most of the new filters were designed for real applications, they have been implemented using modern signal processors.In summary, the development of application oriented and effective design algorithms and filter realizations has been of paramount interest in this thesis.Experiences with the designed and implemented systems have confirmed the success of the research efforts.