In this paper, a new method is given for estimating strain in extrinsic, Fabry-Perot, interferometric (EFPI) fiber-optic
sensors under sinusoidal excitation at the sensor. The algorithm has a low complexity and is appropriate for low-cost
applications. It is an iterative search algorithm based upon a known, sinusoidal excitation and a mean-square-error
objective function. The algorithm provides an estimate of the maximum time-varying strain due to the excitation. It is
shown that, for a broad range of parameters, the algorithm converges to the global minima with a high degree of
probability. Empirical test results for two fiber-optic sensors with different gauge lengths along with corresponding
measurements from a resistive strain gauge are given and shown to compare very well.
A number of Extrinsic Fabry-Perot Interferometer processing techniques have been demonstrated for use to extract gaugelength
measurements from optical detector output signals. These include: (1) an artificial Neural Network method, (2) a
direct phase synthesid method, and (3) an iterative search method. For applications where the processing is to be
performed with low-power hardware, co-located with the sensor, the hardware implementation architecture and
complexity become critical for a practical solution. In this paper, implementation complexity tradeoffs and comparisons
are given for various implementation architectures for each method with respect to each gauge-length estimate. Our
research considers complexity as measured in terms of the number of hardware-resident arithmetic operators, the total
number of arithmetic operations performed, and the data memory size. It is shown that accurate gauge-length estimates
are achievable with implementation architectures suitable for applications including low-power implementations and
scalable implementations.
Artificial neural networks are studied for use in estimating strain in extrinsic Fabry-Pérot interferometric sensors. These networks can require large memory spaces and a large number of calculations for implementation. We describe a modified neural network solution that is suitable for implementation on relatively low cost, low-power hardware. Moreover, we give strain estimates resulting from an implementation of the artificial neural network algorithm on an 8-bit 8051 processor with 64 kbytes of memory. For example, one of our results shows that for 2048 samples of the transmittance signal, the presented neural network algorithm requires around 24,622 floating point multiplies and 35,835 adds, and where the data and algorithm fit within the 64-kbyte memory.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.