Start Date
4-1969 8:00 AM
Description
Data Compression using representation of signals 05^ finite series of orthogonal functions has been often discussed in the literature as giving compression ratios of twenty to one or more. Such schemes should be designed, however, based on the accuracy achieved at the receiver, including errors introduced in quantizing and transmitting the coefficient values.
A means for performing such a design is presented, and, assuming particular functions and signal characteristics, design curves are given. This procedure is compared with a sample and hold scheme illustrating significant improvements in data compression ratios with the orthogonal function approach. Experimental results verifying these theoretical relationships are presented and discussed.
Data Compression Using Orthogonal Functions
Data Compression using representation of signals 05^ finite series of orthogonal functions has been often discussed in the literature as giving compression ratios of twenty to one or more. Such schemes should be designed, however, based on the accuracy achieved at the receiver, including errors introduced in quantizing and transmitting the coefficient values.
A means for performing such a design is presented, and, assuming particular functions and signal characteristics, design curves are given. This procedure is compared with a sample and hold scheme illustrating significant improvements in data compression ratios with the orthogonal function approach. Experimental results verifying these theoretical relationships are presented and discussed.
Comments
No other information or file available for this session.