A novel approach to code analysis of digital signal processing systems

Authors:
Olaf Lüthje;Martin Coors;Holger Keding;Heinrich Meyr
Affiliations:
Aachen University of Technology, Aachen, Germany;Aachen University of Technology, Aachen, Germany;Aachen University of Technology, Aachen, Germany;Aachen University of Technology, Aachen, Germany
Venue:
CASES '01 Proceedings of the 2001 international conference on Compilers, architecture, and synthesis for embedded systems
Year:
2001

Citing 8
Cited 1

Compilers: principles, techniques, and tools

Compilers: principles, techniques, and tools
Optimally profiling and tracing programs

ACM Transactions on Programming Languages and Systems (TOPLAS)
System level fixed-point design based on an interpolative approach

DAC '97 Proceedings of the 34th annual Design Automation Conference
FRIDGE: a fixed-point design and simulation environment

Proceedings of the conference on Design, automation and test in Europe
Fast bit-true simulation

Proceedings of the 38th annual Design Automation Conference
Principles of Program Analysis

Principles of Program Analysis
High Performance Compilers for Parallel Computing

High Performance Compilers for Parallel Computing
FRIDGE: An Interactive Code Generation Environment for HW/SW CoDesign

ICASSP '97 Proceedings of the 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP '97) -Volume 1 - Volume 1

Automating Customisation of Floating-Point Designs

FPL '02 Proceedings of the Reconfigurable Computing Is Going Mainstream, 12th International Conference on Field-Programmable Logic and Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present a technique for program analysis targeting digital signal processing systems. It is utilized by the FRIDGE design environment to transform floating-point systems into fixed-point systems. While the transforming algorithms have been subject to a number of other publications, we now put the focus on the novel technique of program analysis employed by FRIDGE. The code is analyzed by interpreting it without processing any concrete input data. This enables most accurate analysis even capable of detecting dynamic code properties without relying on input vectors as e. g. profilers do.