Application of joint time-frequency analysis in mixed signal testing

Combining the processing power and flexibility of modern DSP based test-systems with the broad area of joint time-frequency signal analysis yields a potentially promising new line of research in the area of analog and mixed-signal test. This paper discusses some practical examples of applying analyses tools, like the short-time-Fourier and wavelet transform for testing ADCs. The research focuses on gaining insight into the advantages of these analysis techniques, for instance in revealing device performance deviations or faults which are not detectable using conventional analysis tools, and also the ability to cover the same faults in a single test which would have taken several tests using conventional methods.

[1]  J. Flanagan Speech Analysis, Synthesis and Perception , 1971 .

[2]  O. Rioul,et al.  Wavelets and signal processing , 1991, IEEE Signal Processing Magazine.

[3]  F. Hlawatsch,et al.  Linear and quadratic time-frequency signal representations , 1992, IEEE Signal Processing Magazine.

[4]  J. Morlet Sampling Theory and Wave Propagation , 1983 .

[5]  Taco Zwemstra,et al.  MixTest: A mixed-signal extension to a digital test system , 1993, Proceedings of IEEE International Test Conference - (ITC).

[6]  F. Harris On the use of windows for harmonic analysis with the discrete Fourier transform , 1978, Proceedings of the IEEE.

[7]  Keith Baker,et al.  Improving Total IC Design Quality using Application Mode Testing , 1992, Proceedings International Test Conference 1992.

[8]  E E David,et al.  Identification of a speaker by speech spectrograms. , 1969, Science.