Optical waveguides – Optical waveguide sensor
Reexamination Certificate
1999-10-07
2001-04-03
Sanghavi, Hemang (Department: 2874)
Optical waveguides
Optical waveguide sensor
C359S349000, C250S227190
Reexamination Certificate
active
06212306
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a method and apparatus for optical demultiplexing multiple Bragg grating sensors in a serial array in optical fibers.
BACKGROUND OF THE INVENTION
Fiber optic Bragg gratings may be used as sensors to monitor perturbations in their environment. A Bragg grating is formed in a single mode optical fiber by creating a periodic refractive index perturbation in the fiber core as described by Kawaski, Hill, Johnson and Fuhjii in Optics Letters, Vol. 3, pp. 66-68, 1978. The diffraction grating in the fiber core will reflect optical frequencies within a narrow bandwidth around the Bragg wavelength of the optical grating. The Bragg wavelength of the diffraction grating can be altered by changing the grating pitch. If an external influence alters the grating pitch then the reflection spectrum of the grating can be monitored to determine the magnitude of the external influence. If the grating is subject to varying strain or temperature, the pitch of the grating is altered as described by Morey, Meltz and Glenn in the Proceeding of the IEEE, vol. 1169, pp. 98-107, 1989. By coupling the grating to an appropriate transducer, the grating can be used to monitor a wide variety of parameters including but not limited to strain, temperature, vibration, pressure, and acceleration.
Fiber optic Bragg grating sensors offer many advantages over traditional electrical sensors for monitoring the various parameters. They provide inherent immunity to electromagnetic interference and provide a reliable signal with very little noise. They can also withstand large variations in temperature and pressure and are compact in size allowing them to be used in locations where conventional sensors are impractical. Bragg grating fiber sensors have the additional advantage that the signal is encoded directly into an absolute wavelength shift of the optical signal, so the signal is insensitive to optical power fluctuations and other signal perturbations.
Unfortunately, the design of Bragg grating sensor systems is often more costly than the conventional electrical sensor alternatives and this has prevented their widespread adoption in many applications. To increase the utility of Bragg grating sensors, it would be advantageous to be able to multiplex many grating sensors in the same optical fiber in order share expensive resources such as the optical source and the sensor measurement unit among the many sensors thereby dramatically reducing the cost per sensor. The placement of many sensors in the same fiber often simplifies the installation of the sensors in structures or systems by reducing bulk and complexity. It is also desirable that the functionality and performance of the system not be degraded by the multiplexing technique.
These potential advantages have motivated significant efforts into developing methods of multiplexing Bragg grating sensors. It would be very beneficial to be able to multiplex a hundred sensors or more in a single optical fiber using only one light source and spectral measurement system. Current systems have fallen short of this goal with about ten sensors per fiber in demonstrated systems that do not severely restrict the sensor's application. As the number of sensors grows there is an increased demand on the optical source power and the complexity of the multiplexing and/or demultiplexing. For a very large number of sensors the cross talk between the sensors can become a significant problem.
Many different multiplexing techniques have been developed for Bragg grating sensors. The most successful techniques for use with a large number of sensors have been wavelength division and time division multiplexing. Examples of these systems are described in the paper by Kersey et al. in the Journal of Lightwave Technology vol. 15, pp.1442-1462, 1997.
In wavelength division multiplexing, the Bragg wavelength of each sensor is set at a separate and unique wavelength. The separations of the Bragg wavelengths are made to be far enough apart so that any reasonable external influence to the grating sensors will not be sufficient to cause the Bragg wavelengths of any two sensors to overlap. Thus each sensor is given a unique wavelength band or slot for its Bragg wavelength. In many situations, the size of each wavelength slot may need to be very large. This requirement can result from the necessity to be able to detect a large range of the parameter being sensed or due to the fact there may be uncertainty in the nominal Bragg wavelength of the sensors. Uncertainty may arise from variations in the fabrication process of the gratings, by static strains or uncertain operating temperatures when the sensor is used. The variability can necessitate a wavelength slot for each sensor in excess of 15 nm for Bragg wavelengths near 1550 nm. When the number of multiplexed sensors is large, the bandwidth requirement on the optical source can become intractable thus limiting wavelength division multiplexing to well controlled sensors that are subject to small external influences.
To overcome the aforementioned problems associated with limited optical bandwidth, the Bragg wavelengths of the sensors may be fabricated with nearly identical Bragg wavelengths and multiplexed with time division multiplexing. In this method a short optical pulse is sent along the fiber containing the Bragg sensors. The pulse will partially reflect off of each sensor and return the sensor information from each grating. The signals from each sensor can be distinguished by their time of arrival. Previous demonstrations of time division multiplexing have determined the time of arrival of the signal by converting the optical pulses into an electrical signal and then gating the electrical signal with a known time delay. Only the pulse that is passing through the electronic detector at the time of the gate is measured. By varying the time delay of the gate, the signals from each of the sensors can be read out.
A previous method used in the art to identify the sensor signals is to electrically gate the sensor signals as disclosed in U.S. Pat. No. 5,680,489. Since the sensors are now identified by time discrimination instead of wavelength, bandwidth requirements of the source will not limit the number of sensors. However, different problems can be encountered in time division multiplexing that can limit the performance of the system. Time division multiplexed systems generally experience more noise than wavelength division multiplexed systems. A significant contribution of the noise is from multiple reflection between the different grating sensors that cause a pulse to arrive back from the sensor array at a time later than expected. Noise is also be contributed by the optical source which may not be pulsed in an ideal manner so that there is a finite level of optical power between successive pulses.
Bragg grating sensor systems often require a very high dynamic range of eighty to a hundred and twenty decibels. Therefore any small sources of noise can be significant. To optimize the performance of the system it is necessary to perform the signal gating in as short a time period as possible. This allows the system to reject a large portion of the noise that does not return at the same time as a sensor pulse. With the method of gating used previously in the art, the performance of the system is limited. An electronic circuit performs the gating action after an optical detector has detected the optical signal. Therefore the electronic circuit must be operated at the speed of arrival of the optical pulses. It is difficult to operate electronic circuits at very high speed and still maintain very high signal fidelity due to noise and distortion. Since the gating is done after the optical signal is detected, the wavelength measurement on the signals must be done before the gating. Therefore any noise or distortions in the gating process will create errors in the sensor signal. Furthermore, the limited operation of this gating method will reduce the spatial resolution of the sensor system since the pul
Cooper David J. F.
Smith Peter W. E.
Cooper David J. F.
Hill & Schumacher
Sanghavi Hemang
LandOfFree
Method and device for time domain demultiplexing of serial... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and device for time domain demultiplexing of serial..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and device for time domain demultiplexing of serial... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2515331