Considered in this paper is a framework for addressing sensor issues that are related to nonlinearity. When a signal is picked up by a nonlinear sensor, it is often the case that the high amplitude part of the signal is distorted by nonlinearity and the low amplitude part is indistinguishable from noise. The distorted output or sensed signal may no longer represent the original input signal due to the presence of high and low frequency foreign spikes in its frequency spectrum. This situation poses a challenging problem: Would it be possible to uniquely extract the original information from the distorted output? A treatment of this problem is given and it is shown that unique signal recovery is possible when the nonlinear characteristics of the sensor satisfy certain requirements. Based on the analysis, an algorithm is developed to recover finite length signals and its validity and efficiency are demonstrated by simulation results. When the sensor model is not available, it is shown how a model identification scheme may be incorporated into the developed scheme. Experiments performed on a physical sensor support the proposed recovery scheme.

Keywords:
sensors, nonlinear systems, nonlinear distortion, signal processing, signal detection, noise, algorithm theory, identification
Topics:
Algorithms, Sensors, Signals
1.
Rush, A., 1998, “Nonlinear Sensors Impact Digital Imaging,” Electron. Eng. (U.K.).
2.
Martin
,
W.
,
2001
, “
High Dynamic CMOS Image Sensors
,”
G.I.T. Imaging Microsc.
, pp.
26
–
28
.
3.
Baras
,
J.
, and
Bensoussan
,
A.
,
1989
, “
Optimal Sensor Scheduling in Nonlinear Filtering of Diffusion Processes
,”
SIAM J. Control Optim.
,
27
, pp.
786
–
814
.
4.
Jayasuriya, S., and Langari, R., 1996, “Nonlinear Filtering of Signals Exhibiting Large Fluctuations in Strength,” Proc. ASME International Mechanical Engineering Congress and Exposition, pp. 159–163.
5.
Proakis, J., and Manolakis, D., 1996, Digital Signal Processing, Prentice–Hall, Inc., Upper Saddle River, NJ.
6.
Heuser, H. G., 1982, Functional Analysis, John Wiley & Sons, New York.
7.
Gustafson, D. E., and Kessel, W. C., 1979, “Fuzzy Clustering With a Fuzzy Covariance Matrix,” Proc. IEEE Conference on Decision and Control, pp. 761–766.
8.
Babusha, R., and Vebruggen, H. B., 1995, “Identification of Composite Linear Models Via Fuzzy Clustering,” Proc. of the European Control Conference, pp. 1207–1212.
9.
Frank, W., 1995, “Compensation of Linear and Nonlinear Sensor Distortions by Digital Post Processing,” Proc. Kongressband Sensor ’95, pp. 889–892.
Copyright © 2004
 by ASME
You do not currently have access to this content.