Optical Fiber Sensors and Sensing Networks: Overview of the Main Principles and Applications

21 Nov.,2023

 

Optical fiber sensors present several advantages in relation to other types of sensors. These advantages are essentially related to the optical fiber properties, i.e., small, lightweight, resistant to high temperatures and pressure, electromagnetically passive, among others. Sensing is achieved by exploring the properties of light to obtain measurements of parameters, such as temperature, strain, or angular velocity. In addition, optical fiber sensors can be used to form an Optical Fiber Sensing Network (OFSN) allowing manufacturers to create versatile monitoring solutions with several applications, e.g., periodic monitoring along extensive distances (kilometers), in extreme or hazardous environments, inside structures and engines, in clothes, and for health monitoring and assistance. Most of the literature available on this subject focuses on a specific field of optical sensing applications and details their principles of operation. This paper presents a more broad overview, providing the reader with a literature review that describes the main principles of optical sensing and highlights the versatility, advantages, and different real-world applications of optical sensing. Moreover, it includes an overview and discussion of a less common architecture, where optical sensing and Wireless Sensor Networks (WSNs) are integrated to harness the benefits of both worlds.

The document is organized as follows: Section 2 presents the different types of optical fiber sensors, their main operating principles, and applications. Furthermore, the main characteristics of these sensors are summarized and compared; Section 3 describes the fundamentals of optical fiber sensing networks (multi-point and distributed), how they are arranged, their main areas of application and the characteristics of two commercial solutions are compared; Section 4 addresses the relevance of optical sensing integration in WSNs; Finally, the conclusions and future prospects are provided in Section 5 , reflecting the most important aspects of this paper.

This paper provides a review of optical fiber sensors, in addition to optical fiber sensing networks and their real-world applications. Moreover, we analyze the integration of optical fiber sensing with OFSNs, allowing us to combine the benefits of both areas. This integrated approach is relatively recent but can significantly evolve and gain attention in the future. provides an outline of the main topics addressed in this paper.

Optical fiber sensor systems are normally used in pre-defined positions. Therefore, extensive lengths of fiber optic cable are necessary for connecting all the sensors and creating an optical fiber network, which can be expensive and impractical. In recent years, Wireless Sensor Networks (WSNs) have gained considerable attention for their effectiveness in acquiring information on parameters such as temperature, pressure, acceleration, or vibration [ 21 ]. Nevertheless, most WSN systems do not integrate optical fiber sensors and do not benefit from their special characteristics and advantages. Therefore, the integration of optical fiber sensors in WSNs introduces benefits and new capabilities for the design of advanced hybrid-sensing systems [ 22 ].

The fiber has low optical attenuation enabling propagation over long distances (kilometers) between monitoring stations. The low attenuation is also important to perform multiplexed measurements. By using a single optical source and detection unit, it is possible to operate large arrays of distributed sensors without active optoelectronic components in the measurement area. In turn, the electromagnetic passiveness and environmental resistance can be maintained [ 19 , 20 ].

Optical fiber sensors are electromagnetically passive. This characteristic is very important as it allows the use of optical sensors where other types of sensors cannot be employed, for example, in high and variable electric field environments where there are explosion risks. Furthermore, the silica compound, which is the basic transduction material of optical fiber, is resistant to most chemical and biological agents and thus can be used in this type of environment and materials. Another advantage is that optical fiber sensors can be small and lightweight [ 18 ].

Optical fibers provide sensing solutions for many types of applications and environments with high performance. The design of the fiber sensors can take advantage of one or several optical parameters of the guided light, such as intensity, phase, polarization, and wavelength. The optical fiber offers dual functionality: measurement of several parameters through changes in the properties of light propagating through the fiber; and functions as a communications channel, therefore dispensing an additional dedicated communication channel, and thus presenting an advantage concerning all other sensing technologies.

The evolution of optical fiber technology also enabled the development of devices for optical processing entirely in fiber, reducing insertion losses and improving the quality of processing [ 3 ]. One factor that contributed to the full migration of optical fiber technology was the identification of photosensitive optical fibers. This discovery was made in 1978 by Hill et al. [ 4 ] and led to the development of the optical Fiber Bragg Grating (FBG). In parallel with the interest and use in optical communications, the Bragg gratings have gained a prominent position in optical fiber sensors, due to their versatility in different sensing applications [ 5 , 6 ]. Several markets, such as aeronautics [ 7 ], aerospace [ 8 ], civil engineering [ 9 , 10 , 11 , 12 , 13 , 14 , 15 ], and biological [ 16 ] or environmental monitoring [ 17 ], have assimilated the advantages of this technology.

The development of optical fiber technology marked an important step in global communications technology. In the 70s, the emergence of optical fibers with low attenuation [ 1 ] enabled long-distance communications with high bandwidth. Since these advancements, the production volume has continued to grow, and by 2000, optical fibers had already been rapidly installed around the world [ 2 ].

2. Optical Fiber Sensors

An optical fiber is a cylindrical dielectric waveguide, where both the core and the cladding are composed of glass or plastic, and the surrounding coatings used to protect the optical fiber are made of acrylate or polyimide materials. Optical fibers can be multi-mode or single-mode. An optical fiber sensor expands or contracts according to strains or temperature variations. When light is sent down the fiber to the sensor, it is modulated according to the amount of expansion or contraction. Subsequently, the sensor reflects back an optical signal to an analytical device, which translates the reflected light into numerical measurements of the change in the sensor length. These measurements indicate the level of strain or the temperature [11].

As mentioned before, optical fiber sensors have several advantages relative to other sensor technologies for a variety of applications with extensive potential in sensing applications. Some advantages of optical fibers with regard to sensing include their small size, no requirement of electrical power at the remote location, and many sensors can be multiplexed along the length of the fiber by using light wavelength shift for each sensor, or by sensing the time delay as light passes along the fiber through each sensor [20]. Because optical fiber sensors are immune to electromagnetic interference and do not conduct electricity, they can be used in hazardous environments where high-voltage electricity or flammable material such as jet fuel may be present. Optical fiber sensors can also be designed to resist high temperatures [12].

For these reasons, the application environments range from dangerous scenarios where there are radioactive, chemical, and other industrial-based hazards to more common and simple uses. However, the development of certain types of optical fiber sensors, for example, in corrosion detection, remain in its infancy [11,23].

In the literature, optical fiber sensors can be classified or categorized considering different aspects. These sensors are frequently grouped according to the sensor location in the fiber, and the operating principle or the application [24,25,26,27]. When considering matters of application, optical sensors can be categorized considering the type of parameters they are intended to measure, namely: physical (e.g., strain, temperature), chemical (e.g., oil parameters, pH, ammonia, detergents, pesticides and humidity) [28,29] or bio-medical (e.g., oxygen, carbon dioxide, proteins, cells, proteins, and DNA) [29,30,31,32,33]. Concerning the sensor location, the optical sensor can be classified as intrinsic or extrinsic [22] ( ).

Intrinsic sensors (upper part of ) directly use an optical fiber as the sensitive material (sensor head) and also as the medium to transport the optical signal with the information measured. They operate via direct modulation of the light guided into the optical fiber, and the light does not leave the fiber, except at the detection end. In this type of sensor, physical perturbations modify the characteristics of the optical fiber, changing the properties of the light carried by the fiber. Alternatively, the modulated light may be coupled back into the same fiber by reflection or scattering and then guided back to the detection system. The simplest fiber sensors vary the intensity of the light and require only a light source and a detector.

Intrinsic optical fiber sensors can be used in distributed sensing over large distances to measure different parameters, for example: temperature can be measured by analyzing the Raman scattering of the optical fiber or by using a fiber bearing an evanescent loss that varies with temperature; electrical voltage can be measured by analyzing the polarization of light as a function of voltage or electric field considering the nonlinear optical effects in specially doped fibers; angles can be measured through the Sagnac effect; and, direction recognition is possible using special long-period fiber grating [18].

Furthermore, intrinsic optical fiber sensors are used in different fields for different purposes. They are employeed as hydrophones for seismic and sonar applications [34,35,36,37,38,39], building a system with several sensors per fiber cable. An advantage of these sensors is that they can simultaneously measure the temperature and acoustic pressure at the same location [40]. This is particularly useful when acquiring information from small complex structures. In oil wells, intrinsic optical fiber sensors are used for measuring temperature and pressure with precision.Simultaneous temperature and strain sensing over large distances is also possible exploring the Brillouin scattering effects, which enable sensing over larger distances (>30 km) [18]. Another application of these sensors is in healthcare for medical imaging and diagnosis [41,42], due to their higher resistance to chemical agents and immunity to electromagnetic interference. Intrinsic sensors are also used in airplanes [43] and cars [44] as high-accuracy optical fiber gyroscopes for navigation purposes.

An extrinsic or hybrid optical fiber sensor (usually based on a multimode fiber cable) (see ) guides the light to/from a location where the optical sensor head is located. The sensor head is external to the optical fiber and is based on miniature components that are used to modulate the properties of light in response to environmental changes associated with physical perturbations of interest. The optical energy is transmitted to the head of the sensors from one end of the fiber, and the other end of the fiber is modulated and coupled to the optical sensor.

In this type of sensor, the optical fiber is simply used to guide the light to and from a location where an optical sensor head is located. The typical configuration is one fiber to transmit to the sensor head, and a second fiber to guide the modulated light back to the optical detector. Another configuration may use only one fiber, the modulated light may be coupled back by reflection or scattering and then guided back to the detection system. This type of sensor can reach places that other transmission methods cannot, for example, inside an aircraft jet engine for temperature measurement, or in locations with extreme electromagnetic fields. Extrinsic sensors are used to measure vibration, rotation, displacement, velocity, acceleration, torque, and temperature [18]. Extrinsic optical fiber sensors provide resistance to noise and signal corruption, however, integration with other types of sensors can be difficult since other sensors produce an electrical output that then has to be converted into an optical signal. Fabry–Perot interferometers are an example of extrinsic sensors, which have a cavity at the end of the fiber where light comes out from and allows users to perform sensing of numerous parameters, e.g., pressure and temperature in geothermal wells [45], ultrasound, humidity, liquid level sensing [46], and structural health monitoring of bridges [12].

With regard to the principle of operation, the optical fiber sensors can be classified as intensity-modulated, wavelength-modulated, phase-modulated, scattering-based or polarization-based ( ) [24,25,26,27]:

  • Intensity-modulated sensors were among the first optical fiber sensors to be developed [ 26 ]. These sensors can detect physical changes or perturbations in the received light (bend loss, attenuation, evanescent fields). Simplicity and lower costs present the advantages of this optical sensor type; however, these sensors are susceptible to fluctuations in optical power loss leading to false readings, and therefore requiring a reference system to minimize the problem.

  • Wavelength-modulated sensors measure the wavelength change in the fiber. Examples of wavelength-modulated sensors include black body sensors, fluorescence sensors, and the Bragg grating wavelength-modulated sensors. The FBG sensor represents the most popular type of wavelength-modulated sensor and is frequently used in different applications since it is capable of single-point or multi-point sensing.

  • Phase-modulated sensors use the interferometry principle to measure interference of the optical fiber light. These sensors are popular owing to their high sensitivity and accuracy; however, this also translates to a higher cost. The most popular phase-modulated sensors include the Mach–Zehnder, Sagnac, Michelson, and Fabry–Perot interferometers.

  • Scattering-based sensors use an Optical Time Domain Reflectometer (OTDR) to detect changes in the scattered light. These sensors are very popular since they enable distributed sensing along the length of the fiber with interesting applications in structural health monitoring, and measuring changes in strain.

  • Polarization-based sensors detect changes in the light caused by an alteration in the polarization state. These sensors exploit the birefringence phenomenon in the optical fiber, where depending on the polarization the refractive index changes. When applying strain to the optical fiber, the birefringence effect occurs and results in a detectable phase difference.

In the literature, the most popular optical fiber sensors are classified into three main groups: Grating-Based, Interferometric and Distributed [47]. Next, we will provide more details about the sensors in each of these three categories, with more focus on those with higher potential for sensing networks.

2.1. Fiber-Bragg Grating Sensors

FBGs [4] are simple, versatile, and small intrinsic sensing elements that have all the advantages generally attributed to optical fiber sensors. Since the information to be measured is encoded in the wavelength of the structure, which is an absolute parameter, FBG sensors can be easily multiplexed in multi-point sensing networks.

The FBG sensor works by exposing a section of the optical fiber core to a periodic pattern of UV light, which results in a permanent alteration of the refractive index of the core. This process provides spectrally controlled reflective properties to the UV-light-treated portion of the fiber [4]. The reflected wavelength exhibits high sensitivity to extension and temperature variations as shown in the lower part of . These sensors are capable of eliminating the problems of amplitude or intensity variations because they are integrated into the light guiding core of the fiber and are wavelength encoded.

Open in a separate window

The fabrication process of FBGs has undergone continual improvement in recent decades to enable functional FBG operation in harsh environments with very high temperatures, such as the oil and gas industries, aircraft engines, among others. Recently, femtosecond laser technology was explored for fabricating microstructures, including FBGs [48]. This fabrication process enabled FBG sensors to operate at very high temperatures, for example, a type-II FBG fabricated on a conventional single-mode fiber achieved long-term thermal stability at 1000 °C for several hundred hours [49], and an FBG based on pure-silica Photonic Crystal Fiber (PCF) was able to operate at ∼1300 °C [50]. PCF is a type of fiber that contains air holes along the fiber in the cladding and/or core. A detailed description of femtosecond laser technology and several exploratory works can be found in [48].

The Bragg wavelength λB in FBG sensors or the wavelength of the reflected light is given by [4]:

λB=2neffΛ

(1)

where neff is the effective refractive index of the fiber core and Λ is the grating period. The Bragg wavelength varies with changes in the grating period and effective refractive index, as can be seen in (1). The grating period is affected by variations in the strain, and the effective refractive index is affected by variations in the temperature.

In order to measure each physical parameter in FBGs, the temperature and strain effects need to be separated from each other. The use of a reference grating presents a practical and simple approach to separate the effects caused by temperature and strain. FBG sensors require demodulators, also known as interrogators, that are used to extract measurand information from the light signals coming from the sensor heads. Since the information is encoded in the Bragg wavelength, the interrogators are expected to read the shifts in the Bragg wavelength and provide the measurand data. presents the interrogator of an FBG sensor.

Typical strain FBG responses include ∼0.64, ∼1, and ∼1.2 pm/μϵ (μϵ = micro-strain) for the Bragg wavelengths of around 830, 1300, and 1550 nm, respectively, [51]. Despite being dependent on the FBG type, the temperature response is typically 6.8, 10, and 13 pm/°C, respectively, [52]. FBG sensors are used in various areas to measure numerous physical parameters [53], such as: temperature and strain [54,55], salinity of sea water [56], pressure and temperature of geothermal wells [45], and displacement and liquid levels [53].

2.2. Interferometric Sensors

Interferometric optical fiber sensors, also referred to as interferometer sensors, are phase-modulated sensors that measure the interference of the optical fiber light. The most popular types of interferometers include the: Mach–Zehnder, Sagnac, Michelson, Fabry–Perot and ring resonator. illustrates a schematic representation of different fiber optic interferometers.

Open in a separate window

The Mach–Zehnder interferometer [57,58] works by sending a light beam split into two parts so that the signal propagates through a reference arm and a sensing arm. It then measures the phase shift between the two light beams which are recombined at the detector. The phase shift occurs due to a change in sensing arm length following mechanical or thermal strain. This type of interferometer is mostly used to test telecommunication industry networks, for underwater sensing as a hydrophone [59], for sensing temperature and refractive index [60], in the healthcare industry for optoacoustic imaging applications [61], and as a heart rate and respiratory rate sensor [62].

The Michelson interferometer [63] is similar to the Mach–Zehnder, but instead of having a second beamsplitter, it uses mirrors to reflect the light in the reference and sensing arms back to the source. Some of its applications are as a refractive index sensor [64] and as a hydrophone sensor [65].

The Fabry–Perot interferometer [66] is an extrinsic sensor that uses two parallel reflective surfaces separated by a certain distance and measures the interference between the transmitted and received signal. It is widely used to monitor structural components and downhole pressure in the oil and gas industry [45,67], and for sensing temperature, acoustic waves, ultrasound waves, gas, and liquid levels, among others [46].

The ring resonator [68] detects inertial rotation normal to the plane of the ring resonator and may be used for optical switching and photonic biosensors, among other applications.

The Sagnac interferometer [69] is used to measure the phase shift between two different wavelengths propagating in opposite directions. An interesting type of Sagnac interferometer is the fiber optic gyroscope, initially introduced in 1976 [70]. It detects the phase shift in the optical fiber, assuming a rotating fiber optic coil with two light waves traveling in opposite directions in the coil. These light waves will travel different distances, which results in different travel times and different phases between the two waves. The phase difference ΔΦ is given by:

ΔΦ=8πNλcA×Ω

(2)

where N is the number of coil turns, λ is the wavelength in a vacuum, c is the speed of light, A is the area vector of the fiber coil, and Ω is the rotating rate (angular frequency) vector.

The output response of a gyroscope sensor is a raised cosine function with respect to the phase difference. When a point corresponds to ΔΦ=0 in the response curve, it is at maxima (or minima), and therefore the sensitivity is low and the rotational direction of the fiber coil cannot be distinguished due to the symmetric response. Thus, the operating point has to be shifted to a position where the response is not zero. This is achieved using a phase modulator at the end of the fiber coil.

The fiber coil rotates in the clockwise (CW) direction. In order for the rotating light waves (one in the CW direction and another in the counter-clockwise (CCW) direction) to meet and interfere with each other at the same exit point, they should have entered the coil at different time instants. Since the coil is rotating, different entering times mean different entering points.

When compared with other alternatives, the fiber optic gyroscope has an important advantage—ruggedness. Since it does not contain any moving parts, in contrast to mechanical and ring laser gyroscopes, it overcomes cross-axis vibration and acceleration achieving higher resolution than ring laser gyroscopes. The performance of these sensors has improved significantly rendering them suitable for meeting the most demanding of requirements in gyroscope accuracy, finding applications in military and commercial markets [71,72]. With the evolution of technology, the price of these sensors continues to decrease due to the expansion of the optical fiber communications market and reduced costs of the building components [73]. Hence, they are considered the most cost-effective solution for high-accuracy inertial navigation applications [74], being particularly useful in GNSS-denied environments where it is necessary to locate a vehicle recurring to inertial navigation.

Optical fiber acoustic sensors or optical fiber hydrophones use interferometry for measurements in underwater environments [34,38,39]. These sensors are often preferred as substitutes for piezoelectric ceramic sensors owing to the advantages associated with optical fibers, namely, high sensitivity, high dynamic range, and immunity to electromagnetic interference. Fiber optic hydrophones rely on the operating principle of interferometers and may be based on Mach–Zehnder, Sagnac, Michelson, or Fabry–Perot interferometers to detect the phase change of light which is usually introduced by pressure. Propagating acoustic waves cause pressure-induced variation in the refractive index, which in turn produces a phase shift in the light propagating through the fiber [75]. There are several applications of fiber optic hydrophones, for instance, in seismic exploration for oil reserves [76] and geoacoustic seafloor exploration [77].

The work in [78] reviews several optical fiber-based sensors for real-time measuring and monitoring of volatile organic compounds (VOCs), such as alcohols, carbonyls, alkanes, among others. This paper includes several applications of interferometers for sensing volatile organic compounds, for example, a sensor using a polydimethylsiloxane (PDMS) sensing film over a glass substrate. PDMS has unique features that cause it to change its refractive index when it interacts with various VOCs causing a fringe shift in the interferometer.

2.3. Distributed Sensors

Optical Time Domain Reflectometers (OTDRs) are some of the most well-developed in-line sensors based on the scattered light propagating through the fiber, which contains Rayleigh, Brillouin and Raman scatterings as displayed in and . In addition to the original wavelength (called the Rayleigh component), the scattered light contains components at wavelengths that are higher and lower than the original signal (known as the Raman and Brillouin components). These shifted components contain information about the local properties of the fiber, such as strain and temperature.

Open in a separate window

Open in a separate window

The functioning of OTDR [79] details a process whereby light injected into the system is shortly pulsed in order to achieve spatial resolution. As the backscattered light is detected with a certain delay relative to the emitted beam, the region where the scatter originated is identified. Hence, the loss in this region can be measured because the intensity of the scattered light in that region is different. Using the frequency or arrival time of the scattered light allows determination of the measurand amplitude and location.

Rayleigh scattering results from the interaction of light with refractive index fluctuations in the fiber core, that appear in spatial scales much shorter than the light wavelength. Rayleigh scattering has the same light frequency as the incident light and is very weak, especially in single-mode fibers [80], and thus is not frequently used in sensing applications.

The Brillouin scattering appears from the interaction of light with acoustic modes in the medium, which are induced by the light propagation [80]. Determining the Brillouin frequency shift relative to the incident light provides a measure of temperature or strain, and allows distributed sensing in long-range fibers. This frequency shift is an intrinsic property of any silica fiber, and therefore allows the production of low-cost sensing elements. The measurements are stable over time because the optical effect only depends on the fiber material. Therefore, Brillouin scattering has been used in distributed sensing of large structures, mainly sensing strain in oil wells, pipelines, bridges, or power lines [18,81]. Brillouin-OTDR technology can be used to measure strain and temperature continuously at any point distributed along an optical fiber, thus it is used for full-scale monitoring of large structures such as tunnels, dams, pipes, subways, and large bridges in the order of hundreds of kilometers, where point-measurement monitoring techniques cannot be used. In addition, the optical fiber in Brillouin-OTDR functions both as a sensor and as a transmission medium, enabling long-distance and real-time remote monitoring [11].

Distributed strain sensing with Brillouin scattering achieves a range up to 30 km with a 20 μϵ (micro-strain) resolution (more advanced schemes can achieve higher resolutions, reaching up to 0.1 μϵ) [27]. The sensing point associated with a physical perturbation can be resolved up to 1 m over a 10 km length; however, the accuracy decreases as distance increases. Brillouin scattering for temperature sensing has a resolution of 0.5 °C [27].

Raman scattering results from the interaction of the propagating light with molecular vibrations in the medium. The scattering characteristics only depend on temperature, which may present an advantage when temperature is the sensing parameter of interest since there are no cross-sensitivity effects. However, Raman scattering in optical fibers has a much higher power threshold than Brillouin scattering. Therefore, Raman-OTDR is used for temperature distributed sensing [82] because thermally excited acoustic waves generate the spontaneous Raman scattering process rendering the scattering cross-section temperature-dependent. Distributed temperature sensing with Raman scattering has a temperature resolution of 0.5 °C with a measurement range up to 15 km at 1 m resolution (or up to 25 km at 1.5 m resolution) [27].

OTDR technology is used for distributed sensing and has significantly evolved in recent decades to the extent that nowadays there is accessible equipment providing a loss distribution map with a spatial resolution of a centimeter [83]. Examples of OTDR-distributed-sensing applications can be found in Section 3.2.

2.4. Summary

compares the different types of optical fiber sensors according to the measurand employed for sensing, the fields where they are used, the parameters that they measure, and the possible network configurations that are applied in each type of sensor. It also includes the typical performance specifications of some of these sensors.

Table 1

SensorMeasurandField(s)Sensing Application(s)Network ConfigPerformance [27]Fiber Bragg GratingsWavelength shiftEngineering, Physics, CryogenicsTemperature, pressure, strain, liquid level, displacement, salinity.Single/Multi-point sensingStrain res. < 0.5 μϵ
Long-term accuracy < 1%
Temp. res. 1 °CInterferometersPhase-shift in lightNavigation, Engineering, NetworksGyroscope (inertial navigation, surveying, defense), hydrophone (refractive index sensor, pressure monitoring of structural components), and optical switching.Single/Multi-point sensingN.A.Rayleigh-OTDRRayleigh scatteringEngineering, Physics, CryogenicsDistributed acoustic sensing.Distributed sensingFreq. range 1 mHz to 100 kHz
Spatial res. 1 m
Length up to 50 kmRaman-OTDRRaman scatteringEngineering, Physics, CryogenicsLong-distance and real-time monitoring, distributed temperature sensing.Distributed sensingTemp. res. 0.5 °C
Meas. range up to 15 km w/ 1 m spatial resolution
Meas. range up to 25 km w/ 1.5 m resolutionBrillouin-OTDRBrillouin scatteringEngineering, Physics, CryogenicsLong-distance and real-time monitoring, distributed temperature, and strain sensing.Distributed sensingTypical strain res. 20 μϵ
but can achieve up to 0.1 μϵ
Meas. range up to 10 km w/1 m spatial resolution, but supports larger ranges w/reduced accuracy
Temp. res. 0.5 °COpen in a separate window

With high quality products and considerate service, we will work together with you to enhance your business and improve the efficiency. Please don't hesitate to contact us to get more details of Fiber Optic Fence Sensor System.