File Download

There are no files associated with this item.

  • Find it @ UNIST can give you direct access to the published full text of this article. (UNISTARs only)

Views & Downloads

Detailed Information

Cited time in webofscience Cited time in scopus
Metadata Downloads

Indirect time-of-flight sensor demodulation pixel for kT/C noise suppression

Author(s)
Son, Insang
Advisor
Kim, Seong-Jin
Issued Date
2024-02
URI
https://scholarworks.unist.ac.kr/handle/201301/82179 http://unist.dcollection.net/common/orgView/200000743192
Abstract
In recent years, there has been a growing demand for 3-D imaging applications, including those related to the metaverse and gesture recognition for mobile devices. These applications require high spatial resolution while also minimizing area and power consumption for user convenience. Among the various 3-D imaging methods available, Time-of-Flight (ToF) sensors have emerged as promising candidates. They offer a simple pixel structure and less complex data processing characteristics. Leveraging these advantages, ToF sensors have been commercialized and further developed to achieve lower spatial resolution and higher depth resolution. These advancements parallel the historical progression of color image sensors. To achieve further development, two approaches have been pursued at the pixel level and circuit level. Pixel-level approaches focus on enhancing the performance of individual pixels, addressing factors such as Full Well Capacity (FWC), Dark Current (DC), and Quantum Efficiency (QE). Since most ToF sensors utilize Near-Infrared (NIR) light for user convenience and background light mitigation, the photogenerated signal can be affected by its wavelength. Improving QE enhances the photogenerated signal, particularly in the presence of NIR light. While smaller pixel sizes are often targeted for higher spatial resolution, they come with limitations, including reduced signal storage capacity and increased dark current, which can lead to diminished depth resolution. Therefore, achieving further reductions in pixel size necessitates improvements in FWC and DC. With this improved pixel signal, ToF sensors require a readout circuit that converts electrons into voltage and then into digital values. Circuit-level approaches are primarily focused on optimizing these readout circuits to ensure a clear and accurate conversion of the signal. Many ToF sensors utilize 4T structures for the conversion of electrons to voltage due to their simplicity. However, this electron- to-voltage conversion process in 4T structures can introduce thermal noise from reset transistors, known as reset noise. To mitigate this noise, 5T structures have been introduced, which incorporate an additional transistor for sampling the reset level. In this research, we are focusing on pixel structure and the development of pixel performance. The proposed pixel employs a 6T structure designed to eliminate reset noise while providing sufficient charge storage capability. It also utilizes a two-step charge transfer mechanism involving different n-doping inside the pinned photodiode (PPD) to generate lateral electric fields. The unit pixel consists of three PPD areas. The first is the light receiving PPD, which features both lightly doped and highly doped regions to create a potential slope toward the storage node. The lightly doped area is designed to enhance the optical Fill Factor, while the highly doped area is optimized for faster charge transfer to storage. The second and third areas are dedicated to charge storage within the PPD, featuring a poly gate on top to allow for adjustable potential. This adjustable potential in the PPD results in higher charge transfer efficiency and Full Well Capacity (FWC). The charge transfer path and two-step layer shape have been meticulously designed to ensure complete charge transfer. The unit pixel pitch is 4x4.45μm, with an optical FF of 37%, even without a microlens. All pixel characteristics and charge transfer efficiency have been validated through TCAD simulation. Furthermore, for the reliability of TCAD simulations, process comparison results with previous chips have been implemented.
Publisher
Ulsan National Institute of Science and Technology

qrcode

Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.