Title :
Symbolic simplification of digital signal processing software
Author :
Heck, L.P. ; Schwartz, D.A. ; Mersereau, R.M. ; McClellan, J.H.
Author_Institution :
Sch. of Electr. Eng., Georgia Inst. of Technol., Atlanta, GA, USA
Abstract :
The problem of automatically designing efficient implementations of digital signal processing (DSP) algorithms based on application-specific information is considered. A rule-based code simplification system is presented that is capable of performing code simplifications on a broad class of programs. Several types of symbolic processing rules and their uses in the problem of code pruning are discussed. The design of efficient radix-2 FFT programs is presented as an example, and it is shown that considerable operational savings can be obtained if certain properties of the input data are known
Keywords :
computerised signal processing; program processors; DSP algorithm implementations; application-specific information; code pruning; computational speed improvement; digital signal processing software; extended flow graph; program processor; radix-2 FFT programs; rule driver; rule-based code simplification system; symbolic processing rules; Application software; Computational complexity; Computer applications; Contracts; Digital signal processing; Fast Fourier transforms; Flexible printed circuits; Hardware; Signal design; Signal processing;
Conference_Titel :
Circuits and Systems, 1989., IEEE International Symposium on
Conference_Location :
Portland, OR
DOI :
10.1109/ISCAS.1989.100346
