Neural processing-type fiber-optic strain sensor

A neural-processing-type strain sensor insensitive to thermal variation is presented and calibration of the device through modulation of the processing system´s internal parameters is described. The sensor exploits the variation of the far-field polarization pattern in a single-mode birefringent fiber under the influence of longitudinal strain. A temperature-compensating fiber element is built in, making the sensor assembly immune to thermal variation. Sampling of the sensor output and parallel distributed processing of the samples are integrated within the sensor. The processor manages both a training function and a generalization function. The training function modulates a small-size linear network built into the system. In the working phase, the generalization function is used to recover measurement information. If the sensor is thermally compensated, the network gives a reading of the measurand with an error not exceeding 0.1%. Applicability of the processing system to bimodal sensor output is also described