With the continuous development of autonomous driving technology, LiDAR (Light Detection and Ranging) systems have become an indispensable part of the modern automotive industry. LiDAR systems use light to detect and measure the distance to objects, providing a precise and reliable method for perceiving the surrounding environment. In this system, Single-Photon Avalanche Diodes (SPADs) play a crucial role.
This study, led by Professor M. Jamal Deen, a professor at McMaster University in Canada and the former President of the Royal Society of Canada, and conducted by researchers from the University of Electronic Science and Technology of China, aims to explore the modeling and design of SPADs in LiDAR applications. The authors, through in-depth analysis of the working principles of SPADs, reveal their vital role in LiDAR systems and propose a series of innovative design and optimization strategies.
(Note: Prof. M. Jamal Deen is a professor at McMaster University in Canada, the Senior Canada Research Chair in Information Technology, and a fellow and member of the Royal Society of Canada. He is a highly influential figure in the fields of nanoelectronics and bio-optoelectronic sensing internationally. His H Index exceeds 45, and his citations exceed 6800. He is a fellow of the IEEE (2002), a fellow of the Engineering Institute of Canada (EIC) (2003), a fellow of the American Physical Society (APS) (2004), a fellow of the Electrochemical Society (ECS) (2004), a fellow of the American Association for the Advancement of Science (AAAS) (2005), a fellow of the Royal Society of Canada (RSC) (2006), a fellow of the Canadian Academy of Engineering (CAE) (2007), a fellow of the Indian National Academy of Engineering (INAE), and a foreign fellow of the National Academy of Sciences India (NASI) (2012), among others.)
This paper primarily focuses on the rapidly evolving field of optical detection and ranging (LiDAR) systems in the automotive market, examining the crucial role of SPADs in achieving extended detection range, higher resolution, and rapid response times. With advanced complementary metal-oxide-semiconductor (CMOS) technology becoming increasingly available, SPAD offers an economically efficient solution.
Firstly, it provides a detailed review of recent SPAD applications within LiDAR systems, encompassing both commercial products and research efforts utilizing various CMOS technologies. SPADs manufactured using different technologies exhibit significant performance differences, emphasizing the importance of SPAD models for simulating critical performance before production. Subsequently, it offers a comprehensive overview of the evolution of SPAD models, covering fundamental knowledge and modeling processes. Leveraging 65-nanometer standard CMOS technology and insights from literature, an enhanced SPAD modeling process is introduced. This model considers the two-dimensional distribution of the electric field, enhancing the accuracy of dark count rate (DCR) predictions. To validate its effectiveness, a carefully designed SPAD employing TSMC 65-nanometer standard CMOS technology is calibrated and compared. Measurement results demonstrate negligible afterpulse probabilities (~0%) and satisfactory DCR levels (around 14 kHz at a 0.7 V excess bias). The wavelength dependency of measured photon detection probability (PDP) aligns with simulation results. Additional discussions explore discrepancies between simulation and measurement. Finally, important research challenges are presented based on simulation and measurement outcomes. To address these challenges, potential directions for optimizing SPAD models and designs are proposed, followed by concluding remarks.
SPAD is a photodiode capable of detecting single-photon events. It operates under reverse bias and utilizes the avalanche effect to amplify the photocurrent. When a photon enters the active region of the SPAD and generates a photo-generated charge, this charge is accelerated under the influence of a high electric field and collides with other charges, creating more charge carriers. This process rapidly triggers an avalanche, resulting in a large current pulse, enabling the detection of single-photon events. This characteristic makes SPAD well-suited for low-light conditions and an essential component in LiDAR systems.
In the document, Xuanyu Qian provides a detailed explanation of the basic structure and operational mechanisms of SPAD. Through mathematical modeling and simulation analysis, he emphasizes the impact of temperature, light intensity, and other external conditions on SPAD performance, and presents a range of methods for optimizing its capabilities. Additionally, he explores the potential applications of SPAD in autonomous vehicles and other modern applications, highlighting the potential for improving LiDAR system accuracy and reliability through enhanced SPAD design and performance.
In LiDAR systems, SPAD’s high sensitivity and rapid response time make it an ideal optical detector. It can operate under extremely low light conditions and accurately measure the time information of light signals, allowing for the calculation of light flight time and the distance to target objects. Furthermore, SPAD’s compact size and low power consumption characteristics make it suitable for portable and mobile applications, providing convenience for the integration of LiDAR systems.
To fully harness the performance of SPAD in LiDAR applications, it is crucial to subject it to precise modeling and optimization in design. This process entails in-depth analysis of SPAD’s optical characteristics, electrical attributes, and their interaction with external conditions, such as temperature and light intensity.
For a more in-depth understanding and optimization of SPAD’s performance, we need to conduct a detailed analysis of its photonic properties.
Generation and Transmission of Photogenerated Charge Carriers:
When photons enter the active region of SPAD, they can be absorbed by semiconductor materials, generating photogenerated electrons and electron-hole pairs. These photogenerated charge carriers are rapidly separated under the influence of the built-in electric field and commence their transmission within the semiconductor material.
Quantum Efficiency:
This is a parameter that describes SPAD’s responsiveness to incident photons. It signifies the ratio of the number of photogenerated charge carriers to the total number of incident photons. Enhancing quantum efficiency implies that SPAD can more effectively convert light energy into electrical signals, which is crucial for improving the detection sensitivity of LiDAR systems.
Charge Carrier Transmission Efficiency:
Once photogenerated charge carriers are generated, they need to be efficiently transported to the polycrystalline region of SPAD to trigger the avalanche amplification process. The efficiency of charge carrier transmission depends on the characteristics of the semiconductor material and the design of SPAD. By optimizing these parameters, it is possible to reduce recombination losses during carrier transmission and enhance the overall performance of SPAD.
Once photogenerated charge carriers reach the polycrystalline region, they accelerate under the influence of a high electric field, triggering the avalanche amplification process. This process is crucial for SPAD’s performance.
Avalanche Multiplication Factor: This is a parameter that describes the efficiency of the avalanche amplification process. It is defined as the ratio of the number of charge carriers produced at the end of the avalanche process to the initial number of photogenerated charge carriers. A high avalanche multiplication factor implies stronger signal amplification but may also result in higher noise levels.
Avalanche Breakdown Time: This refers to the time required for the avalanche process to fully develop into a measurable current pulse from the moment photogenerated charge carriers reach the polycrystalline region. This time parameter directly impacts SPAD’s time resolution, which is crucial for distance measurement accuracy in LiDAR systems.
The optical characteristics of SPAD vary with the wavelength of incident light, which is particularly significant in LiDAR applications.
Spectral Response: This describes SPAD’s responsiveness to light signals of different wavelengths. By optimizing the materials and structure of SPAD, high-sensitivity responses to specific wavelength ranges can be achieved, thereby enhancing the detection capabilities of LiDAR systems for target objects in specific environmental conditions.
Wavelength Dependence: Performance parameters of SPAD, such as quantum efficiency and avalanche multiplication factor, may change with the wavelength of incident light. Understanding and modeling this wavelength dependence is crucial for optimizing the overall performance and sensitivity of LiDAR systems.
Electrical Characteristics Analysis: This includes modeling SPAD’s capacitance, resistance, and other electrical parameters. These parameters directly influence SPAD’s response time and count rate, which are essential for improving the range accuracy and resolution of LiDAR systems.
Environmental Conditions Impact: External factors like temperature and light intensity significantly affect SPAD’s performance. Therefore, it is necessary to model these factors and design corresponding temperature compensation and light intensity compensation mechanisms to ensure that SPAD maintains its optimal performance under different conditions.
In Xuanyu Qian’s research, he not only conducted a detailed modeling analysis of SPAD but also proposed a series of optimization strategies to enhance SPAD’s performance in LiDAR applications.
Through in-depth analysis and optimization of SPAD in LiDAR applications, we can not only enhance the performance of LiDAR systems but also drive the development of autonomous driving technology and machine vision systems, creating more intelligent and safer transportation means for humanity. Xuanyu Qian’s research provides a valuable reference, demonstrating how innovative modeling and design methods can unlock the potential of SPAD in modern high-tech applications.
Headquarter
1F., No.96, Luke 5th Rd., Luzhu Dist., Kaohsiung City 821011 Taiwan
Email: qeservice@enli.com.tw
TEL:
+886-7-6955669(Gobal)
+1-929- 586-296(USA)
Shanghai Branch
Room 2A, Building 3, No. 100, Lane 1505, Zuchongzhi Road, Zhangjiang Town, Pudong New District, Shanghai, China
TEL: +86-21-31338780
Support
Investor area
Corporate Social Responsibility
