SMARTPHONE PLASTIC OPTICAL FIBER SENSORS

I declare that this manuscript, entitled "thesis title", is the result of my own work, except for quotations and citations that have been duly acknowledged. It can easily be used to monitor the condition of patients located in remote locations away from hospitals. For example, respiratory rate is one of the important physiological parameters that requires monitoring, as it can be used in the diagnosis of respiratory diseases.

The proposed solution is a fiber sensor, where the flashlight acts as a light source and the camera acts as a photodetector. In the work, the intensity modulation technique in POF is adapted to sense the breathing rate. Furthermore, multiplexing is also a promising direction for sensing in optical fibers, as it can be used to measure multiple parameters.

Introduction

Requirements for the introduction of telehealth and remote monitoring include low costs and the possibility of use in home conditions. Therefore, smartphones are considered a viable tool for applications in this field as they have become widely available. It has been shown that the pulse can be counted with a smartphone camera by observing changes in the skin color of the fingertips [9].

In the current work, the combination of smartphones and plastic optical fiber (POF) will be investigated. Multiplexing is also a promising direction in the field of sensors, as it can be used to collect information on several parameters [11]. In addition, the possibility of multiplying several POFs with smartphones will be investigated.

Literature review

Optical fiber technology and sensing methods based on it
POF-based sensing applications

POF-based sensor for detecting breathing rate
Sensors implemented by multiplexing POF

Sensors based on smartphones
Previous work
Summary of the reviewed information

The chemical reaction between H2S and silver results in a decrease in the spreading qualities of the fiber. As a result, the amount of light coupled into each of the output fibers was changing. It is used to reflect part of the incoming light into the reference fiber.

This light was coupled into a sensing fiber, which was placed in the second part of the setup. When the sensor configuration was bent, the position of the input fiber was changing relative to the output fibers. The water containing the pH was illuminated by the source, resulting in the excitation of the studied sample.

The corresponding changes in the fluorescence intensity of the sample were monitored by the camera. The chemical reaction between H2S and the silver layer caused a deterioration of the transmission properties of the fiber compound.

Figure 2.2: Transmission between two fibers through mirror [14]

Sensing methodology

Working principle

POF parameters
Intensity modulation
Modulating displacement
Collection of information about intensities observed by smartphone

In our case, the surrounding medium of the fiber is air, which means that the refractive index, n, is equal to 1. Optical power is produced by a source such as LED or laser placed at the transmission end of the fiber. The amount of power received is registered on the other hand by direct detection.

Because breathing rate is measured, changes in propagation properties within the link must be caused by breathing movements. The total length of the POF connection used in the measurements is 39 cm, but the system can operate with any POF length up to several meters, limited only by the POF damping. In the initial position, both the transmitting and receiving fibers are aligned horizontally, as shown in Figure 3.4.

Until it reaches the receiving fiber, the light travels a distance equal to 𝑧 = 𝑎 + 𝑑, where a is the distance between the origin of the light ray and the surface of the transmitting fiber. Where Pin is the power of the light source, ɛ is the attenuation within the fiber and r(z) is the radius of the light cone (colored blue in Figure 3.4) at point z. The first is when the radius of the lighting cone, r(z ), is smaller than the radius of the receiving fiber, rf, i.e.

The same conditions are observed in our case, as the critical angle of POF, which is 28°, will result in the illumination cone completely covering the surface of the receiving fiber even when displaced. When implementing the breathing sensor, the coupling area should be placed on the sensitive chest area. As a result of breathing, the angular displacement between the cores of the fibers changes slightly, and less light is coupled into the second part of the link compared to the initial condition shown in Figure 3.4.

The actual implementation of the sensor includes the application specifically developed to perform the collection of such data by the smartphone architecture.

Figure 3.2: Schematic of intensity modulation

Programming setup

Simulation of the setup based on the displacement between fibers
Extraction of breathing rate
Android application

If the angles match the acceptable range, as specified by (3.9), the corresponding photon propagates further along the z direction, as shown in Figure 3.4. The final x and y coordinates of the photon upon reaching the end of the z line are defined as 𝑧 ∗ tan 𝜃𝑥 and 𝑧 ∗ tan 𝜃𝑦. The corresponding photon is received by the receiving fiber if these coordinates are within the area of the receiving surface, which is defined in (3.5).

The graph in Figure 3.11 is a more realistic representation of the relationship previously shown in Figure 3.6. After filtering, FFT is applied to obtain information about the periods of the signals comprising the analyzed sinusoid. The application for Android 7 is developed to set all the required functions and collect the sum of pixels.

The "javaCameraView" class needs to be modified, as it sets camera parameters during application initialization. The "onCameraFrame" function is where the camera starts to receive the image and deliver it to the application. The conversion of the image to grayscale is implemented, so that it has only one parameter to represent brightness.

The sum of the current frame seen on the camera is calculated using a predefined function called . The sum is written as a string value to a text file in an internal directory of the smartphone memory by the "writeToFile" function. The function receives the value of the sum as the input and adds it to the file in the txt format.

As a result, the sum of the images received by the camera in the time between pressing the buttons is stored in the text file.

Figure 3.11: Simulated dependence of the received intensity on the displacement between fibers

Multiplexing

Compared to the lower connectors, the upper ones are supposed to be located at a distance of 15 mm from the smartphone cover, as shown in Figure 3.15b. Otherwise, the fibers would be too close to the camera and it would be impossible to distinguish between the received intensities. The camera lens is not wide enough to cover the images from 3 fibers at a distance shorter than 15 mm.

The current work aims to discover whether each of the multiplexed fibers can act as an independent sensor.

Figure 3.15: a) Front and b) back side of the connector for multiplexing

Testing and Results

Designing connectors
Calibrating camera
Measuring breathing rate
Measurements by the Android application
Multiplexing

The sensory region must be attached to the lower chest to begin detection, as shown in Figure 3.3 from the previous chapter. The link consisting of the 2 fibers, previously discussed in the theoretical background, was created and connected to the smartphone. For the plot, intensities were calculated relative to the first intensity, i.e., the sum of pixels in each image divided by the sum of pixels in the first image.

The results in Figure 4.4 are free of interference from Figure 4.3 and are in line with expectations. This behavior of the curves can be attributed to the previously discussed deviations that occur during recording. Stronger fluctuations are visible in the spectral patterns of the 2nd and 4th experiments, as the breathing was more intense.

Variations in measurement may be explained by uncertainties in equipment acquisition rates; measurements from the POF sensor were taken at 17 fps. The maximum frequency obtained by the developed sensor is the same as that of the accelerometer. For the first experiment, one of the fibers was slightly bent while the other two were at rest.

It was expected that bending one of the fibers would not affect the intensities in the unbent fibers. The bending of the fiber is supposed to cause a change in the respective intensity that can be easily distinguished from the transmissions in the unbent fibers. In these two aforementioned cases, the behavior is closer to the expected one, since the losses of the bent fibers were significantly higher than the intensity changes observed in the unbent fibers.

Similar to the 20° bending case in Table 4.4, the losses in the bent fibers were significantly higher than the losses in the undeformed fiber. The cut areas of the fibers, as seen in Figure 4.14, were not covered by the fiber jacket. However, it was possible to monitor the bending by 20°, since the deformation of each of the two fibers resulted in the clear change in intensity.

Figure 4.1: Connector for measuring breathing rate

Conclusion

Summary
Future work
Dependence of the received intensity on the displacement between fibers
Android application
Representation of the measurements from the displacement modulation
Processing the measurements of breathing
Processing the measurements done by the sensor and the accelerometer
Processing the measurements collected by the Android application

A possible way to further improve the performance of the breathing sensor is to embed it in wearable textiles. Furthermore, due to the textile flexibility, the fiber in it can more accurately follow the changes in the chest shape during breathing, since more points on the surface of POF will be in direct contact with the sensing area. To proceed, equal amount of light must be coupled into each of the fibers when they are not bent.

Živanov, "Low-cost wearable human joint motion monitoring system based on fiber-optic curvature sensor," IEEE Sensors J., vol. 5] Elisa Spanò, Stefano Di Pascoli, and Giuseppe Iannaccone, "Low-Power Wearable ECG Monitoring System for Multiple-Patient Remote Monitoring, " IEEE Sensors J., vol. 10] Daniёl Lakens, "Using a smartphone to measure heart rate changes during re-experienced happiness and anger," IEEE Trans.

Ahmad, “Taper-shaped plastic optical fiber coated with Al-doped ZnO nanostructures for sensing relative humidity,” IEEE Sensors J., vol. Logier, “Medical textiles with built-in fiber optic sensors for monitoring respiratory movements,” IEEE Sensors J., vol. Jamalipour, “Lab-in-a-Phone: Smartphone-based portable fluorometer for ambient water pH measurements,” IEEE Sensors J., vol.

Tosi, “Design of a plastic optical fiber chemical sensor for hydrogen sulfide detection,” IEEE Sensors J., vol. The following code aims to plot the dependence of the intensity values on the respective displacements between the transmitting and receiving fibers, as captured by the camera. Then each frame of the video was saved as an image and imported into Matlab.