To improve the accuracy of signal recognition in optical fiber perimeter defense systems, a method based on Mel frequency cepstral coefficients (MFCCs) and a fast dynamic time warping (FastDTW) algorithm is proposed. Four kinds of sensing signals are acquired by employing an in-line Sagnac interferometer system. MFCC features of the signals are extracted. The FastDTW algorithm is utilized to calculate distances of the optimal warping paths between a test sample and all reference templates. According to the nearest neighbor criterion, signal recognition results are obtained. The average accuracy is over 96%. This proposed identification method is highly efficient and easy to implement.

A bidynamical all-optical reservoir computing (RC) system for parallel task processing is proposed based on a unidirectional semiconductor optical amplifier optical fiber loop. The polarization dynamics and intensity dynamics are excited by the input signals injected into the reservoir via phase modulation and intensity modulation, respectively. Simultaneous computation of two independent tasks is implemented based on the dynamical responses in polarization and intensity of the optical fiber loop. To our knowledge, this is the first time that two kinds of dynamical responses of an all-optical RC system are used as independent task processing channels to implement parallel task processing. The proposed RC system can achieve good parallel task processing performance with low system cost.

A novel fiber-optic sensor based on the alternating electric field force actions on polyimide tubing with space charge for power-frequency electric field sensing is presented. In structure, the sensor consists of a lightweight fiber cantilever beam covered with a length of electrically charged polyimide tubing as the field sensing element. A twin-FBG based Fabry–Perot interferometer is embedded in this fiber beam to detect the beam vibrations excited by the force of power-frequency electric field to be sensed. Space charge in polyimide tubing is formed through a dielectric charging process. The basic concept, structure, fabrication and operation principle of the sensor are introduced with detailed theoretical analyses. The comprehensive experiments with two sensor prototypes are carried out, in which a sensor exhibits a high sensitivity of 173.65 μV/(V/m) with a minimal detectable field strength of 0.162 V/m, and another has a durability of continuous operation for over a year.

A distributed optical fiber vibration sensing (DOFVS) system for signal recognition and location is proposed and implemented by using a reservoir computing (RC) system based on mutually injected semiconductor lasers (SLs). The sensitivity of the RC to the environment is used to detect an external vibration, that is, the mutually injected channel of the SLs is the sensing fiber of the DOFVS system. The phase changes induced by the vibration is converted into the intensity fluctuations through the mutually injected effects of the SLs. The signal recognition and location are both realized by the training and testing process of the RC classifier model, especially, building multiple linear classifiers of the RC and adopting a winner-takes-all method. It is verified numerically and experimentally that the system can detect well the external vibration and meets the basic properties as a reservoir in the injection locking state. The recognition and location accuracies of two types of vibration signals and 20 sensing positions can reach 99.99% and 100% in an experimental system with a 1.12 km-long sensing fiber. The experimental results show that the proposed system can simultaneously realize the signal identification and location with the simple method, fast processing speed and high accuracy.