Mobiele beeldschermen van de volgende generatie met extra functies in het paneel

Summarise with VIRAT
Authors: Aris Siskos
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Mobile devices are an essential part of everyday life. Various applications on mobile phones allow users to access functionalities like e-banking, messaging and calls, cameras, and others. However, these functionalities are not limited to personal data and use; they can even connect, access, monitor, and control sensors, data, and platforms that are interconnected on a larger scale and interact with other mobile devices. Such a scheme is the Internet of Things. This increase in functionalities in mobile devices requires specific security and safety protocols. Biometric identification is often necessary to identify and authenticate a person using a mobile device. Fingerprint sensing is one of the most common methods used to identify and allow a person to access information and applications on mobile devices. FBI provides strict standards for fingerprint identification for high security, and there are various protocols depending on the size of the fingerprint sensor and the number of fingers it allows to identify simultaneously. A fingerprint sensing method requires multiple sensors consisting of pixels. Sensing a detailed fingerprint for high security means that there must be at least 500 pixels per inch, according to the FBI protocols. Chapter 2 analyses various sensing technologies that are used for fingerprint sensing, and this thesis is focused on optical sensing since it can be integrated with relatively low-cost and compact designs in mobile devices. Each sensor requires a pixel circuit, which is responsible for reading the signal from the sensor and feeding it to a series of circuits for buffering, amplification, filtering, and digitization. Since fingerprint sensing is considered a large-area application, the usual silicon technologies are not cost-friendly anymore. Thin-film technologies are the best candidates for enabling low-cost, large-area applications. The most common ones are the a-IGZO and the LTPS, with the latter being the choice of many mobile phone manufacturers. However, the performance of the TFT technologies is lower than that of their silicon counterparts due to the nature of the semiconductor, resulting in a rather challenging circuit design. Another issue is that these technologies do not offer small channel lengths and critical dimensions on the design, and the pixel size is not as compact as it would be in silicon. Finally, the noise of these devices, due to their lower performance and increased trap density, is significantly higher than that of silicon devices. Therefore, it is necessary to examine how much noise they contribute to the overall system. In chapter 3, we solve the first challenge of designing a compact pixel in TFT technologies to fit the 500 ppi resolution. Using a dual-gate TFT technology, we propose a new 2T APS topology, where the amplifier of the pixel circuit operates both as an SF and as a switching transistor by changing the voltage of the backgate of the amplifier. The effect of the backgate voltage on the DC output, gain, time response, and noise of the pixel circuit is analyzed. The pixel circuit is implemented into a 128 x 128 array in a-IGZO technology, which is fabricated together with the frontplane part using organic photodiodes from TNO/Holst Centre. A measurement setup was implemented for the characterization of the 2T APS topology. The pixel shows a satisfying performance with an output range of 1.25 V with less than 10 % linearity error and a sensitivity of 0.23 μV/photons. The pixel circuit can operate for the FAP 50 protocol and 20 fps. Scaling the image sensor further limits the design in terms of scanning speed, area, and power. Chapter 4 explores solutions to address these challenges. A combination of an assembled system and a pixel solution is proposed, where the scanning happens in two phases. A hybrid pixel in literature and the previously proposed 2T APS pixel are combined for a dual-mode operation that will serve the two scanning phases. In a very large-area image sensor, initially, the fingerprint is detected via coarse/fast scanning, where the pixel circuit operates in PPS mode and multiple pixels are binned. Then, in the second phase, the fingerprint details in a specific area of the display are scanned with the pixel circuit operating in APS mode for higher performance. This method not only addresses the scanning speed, but it reduces significantly the number of readout circuitry needed for thousands of pixels. A noise analysis was conducted for both APS and PPS modes. In PPS mode, the SNR variation as a function of the number of binned rows and columns is analyzed. The system was implemented into a 16 x 16 matrix, including various test structures and dummy pixels in LTPS technology. The frontplane was post-processed with organic photodiodes by the PhD student Abu Bakar Siddik. A measurement setup was designed using an FGPA board to drive the gate signal. The pixel circuit can operate for the FAP 60 protocol and presents an output response of 1.85 V with high linearity. A pixel comparison among different technologies and between a-IGZO and LTPS technology is realized in Chapter 5. Technology limitations such as leakage currents and charge injection are analyzed to determine how they are impacting the system operation for both modes. An improved pixel topology is proposed, reducing the leakage currents of the transistor and optimizing the switching of the topology. Finally, the technology LTPO is introduced as the best candidate for image sensor design, improving both the pixel and the readout circuit design. Finally, Chapter Conclusion summarizes the work of the thesis and presents the next steps for a complete characterization of fingerprint sensing as well as the need to characterize and model the noise in TFT transistors, which plays a significant role in image sensor design.

Fields of Study & Subjects

Biometric Identification and Security Thin-Film Transistor Technologies CCD and CMOS Imaging Sensors

Publication details

Journal / source
Lirias
Publication date
25 March 2026
Language
EN
Document type
Thesis