In order to improve identification rates (IRs) of signal recognition for optical fiber perimeter defense systems, a novel signal recognition method based on the fast dynamic time warping (FastDTW) algorithm and nearest neighbor criterion is proposed. The distributed optical fiber sensing system based on an in-line Sagnac interferometer is employed as a simulated perimeter defense system to acquire three different kinds of sensing output signals. The signals are divided into several signal segments according to their categories and selected as reference templates and test samples, respectively. The FastDTW can calculate the optimal warping path distance between the test sample and each reference template. The signal recognition results are obtained according to the nearest neighbor criterion. The experimental results show that the average IR of the three kinds of sensing signals is above 99%. The proposed recognition method does not need special training process, hence is simple, and easy to implement. It can achieve a high identification rate under small sample condition which provides a new approach for the signal recognition of optical fiber perimeter defense systems.
In order to reduce the false alarm rates of optical fiber perimeter systems, this paper proposes a signal recognition method based on optoelectronic reservoir computing (RC) to identify pedestrian intrusion signals from various vibration signals acting on the sensing fiber. The optoelectronic reservoir consists of a single nonlinear node and an optoelectronic feedback loop. The nonlinearity of the reservoir is provided by a Mach-Zehnder intensity modulator. The walking signals were acquired in the laboratory through a distributed optical fiber sensing system based on an in-line Sagnac interferometer. The input data of the reservoir is a random combination of walking signals and the output signals of the sensing system under no interferences. The training input data contain three walking signals at different times. The testing input data contain one walking signal according to the most common case. The average identification rate (IR) for the testing data of 10 different walking signals is as high as 97.3%. The highest IR is 99.3%. The simulation results show that the proposed recognition method of intrusion signals of optical fiber perimeter systems is feasible and effective. RC, as an improved training algorithm of recurrent neural networks, has no need of a large number of samples in the training process and seeking the signal features for the signal recognition task. Therefore, the proposed intrusion signal recognition method based on optoelectronic RC is fast in recognition speed and low at cost.
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.