Akihide Sai
Research Subjects
Akihide Sai has a total of 61 co-authors. Here are the top co-authors by number of shared publications:
- Kentaro Yoshioka
Keio University, Tokyo, Japan
6 shared publications Daisuke Kurose
Toshiba Electronic Devices and Storage Corporation, Minato-ku 105-6691, Japan
4 shared publicationsHidenori Okuni
Toshiba Corporation, Kawasaki 212-8582, Japan
4 shared publicationsHirotomo Ishii
Toshiba Corporation, Kawasaki 212-8582, Japan
4 shared publicationsKatsuyuki Kimura
Toshiba,Kawasaki,Japan
4 shared publications- Tuan Thanh Ta
Toshiba,Kawasaki,Japan
4 shared publications
Latest Publications
- 12 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 80-81
1200x84-pixels 30fps 64cc Solid-State LiDAR RX with an HV/LV transistors Hybrid Active-Quenching-SPAD Array and Background Digital PT Compensation
This paper presents two essential techniques, Active-Quenching (AQ) -SPAD consisting of hybrid HV/LV transistors and Digital SPAD Characteristic Compensation (DSCC) circuit to realize high-performance and palm-size LiDAR. The hybrid AQ circuit shrinks a pixel area while ensures a high PDE. The high pixel density 2D-SPAD array realizes high image-resolution palm-size LiDAR by reducing light-receiving lens and RX unit size. The DSCC provides an on-chip Process/Temperature (PT) calibration without external components, which contributes to weatherability assurance and LiDAR miniaturization. These technologies downsize LiDAR RX to the world’s smallest size, 64cc, and realize a total 350cc-size LiDAR with the competitive performance as a mechanical one.
- 12 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 80-81
1200x84-pixels 30fps 64cc Solid-State LiDAR RX with an HV/LV transistors Hybrid Active-Quenching-SPAD Array and Background Digital PT Compensation
This paper presents two essential techniques, Active-Quenching (AQ) -SPAD consisting of hybrid HV/LV transistors and Digital SPAD Characteristic Compensation (DSCC) circuit to realize high-performance and palm-size LiDAR. The hybrid AQ circuit shrinks a pixel area while ensures a high PDE. The high pixel density 2D-SPAD array realizes high image-resolution palm-size LiDAR by reducing light-receiving lens and RX unit size. The DSCC provides an on-chip Process/Temperature (PT) calibration without external components, which contributes to weatherability assurance and LiDAR miniaturization. These technologies downsize LiDAR RX to the world’s smallest size, 64cc, and realize a total 350cc-size LiDAR with the competitive performance as a mechanical one.
- 15 September 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 55 (11) , 2866-2877
An Automotive LiDAR SoC for 240 × 192-Pixel 225-m-Range Imaging With a 40-Channel 0.0036-mm2 Voltage/Time Dual-Data-Converter-Based AFE
This article presents a 40-channel high-resolution automotive LiDAR system-on-chip (SoC), which utilizes the world's first dual-data converter (DDC). The proposed DDC consolidates the functions of ADC and TDC into a single circuitry and achieves acquisition of both high-precision time and voltage data from the input, realizing a 5x smaller analog front-end (AFE) area than prior arts. Such innovations lead us to a 40-channel AFE integration of the SoC without silicon cost increase, which characterizes our LiDAR system with 2x higher pixel resolution. To enhance the ADC's signal-to-noise and distortion ratio (SNDR) without sacrificing the area efficiency, a calibration-free and variation-tolerant voltage-controlled oscillator (VCO)-based ADC with a multiphase pulse density modulator (MP-PDM)-based feedback is proposed. Our proof-of-concept LiDAR system achieves a measurement range of 225 m at 70-klx direct sunlight, with a resolution of 240 x 192 pixels at 10 frames per second (FPS). To quantitatively measure the effect of the resolution improvement of our LiDAR, we evaluate the LiDAR's detection accuracy of small targets 75 m away. Due to the 2x higher vertical pixel resolution, our LiDAR can detect targets with 2x smaller height. The 3-D point cloud of the LiDAR captured outdoor in the wild is presented.
- 15 September 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 55 (11) , 2866-2877
An Automotive LiDAR SoC for 240 × 192-Pixel 225-m-Range Imaging With a 40-Channel 0.0036-mm2 Voltage/Time Dual-Data-Converter-Based AFE
This article presents a 40-channel high-resolution automotive LiDAR system-on-chip (SoC), which utilizes the world's first dual-data converter (DDC). The proposed DDC consolidates the functions of ADC and TDC into a single circuitry and achieves acquisition of both high-precision time and voltage data from the input, realizing a 5x smaller analog front-end (AFE) area than prior arts. Such innovations lead us to a 40-channel AFE integration of the SoC without silicon cost increase, which characterizes our LiDAR system with 2x higher pixel resolution. To enhance the ADC's signal-to-noise and distortion ratio (SNDR) without sacrificing the area efficiency, a calibration-free and variation-tolerant voltage-controlled oscillator (VCO)-based ADC with a multiphase pulse density modulator (MP-PDM)-based feedback is proposed. Our proof-of-concept LiDAR system achieves a measurement range of 225 m at 70-klx direct sunlight, with a resolution of 240 x 192 pixels at 10 frames per second (FPS). To quantitatively measure the effect of the resolution improvement of our LiDAR, we evaluate the LiDAR's detection accuracy of small targets 75 m away. Due to the 2x higher vertical pixel resolution, our LiDAR can detect targets with 2x smaller height. The 3-D point cloud of the LiDAR captured outdoor in the wild is presented.
- 24 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 54 (10) , 2730-2742
An 8 Bit 12.4 TOPS/W Phase-Domain MAC Circuit for Energy-Constrained Deep Learning Accelerators
- 24 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 54 (10) , 2730-2742
An 8 Bit 12.4 TOPS/W Phase-Domain MAC Circuit for Energy-Constrained Deep Learning Accelerators
- 12 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 27 (11) , 2575-2586
Digital Amplifier: A Power-Efficient and Process-Scaling Amplifier for Switched Capacitor Circuits
To realize high-resolution pipelined and pipelined-SAR analog-to-digital-converters (ADCs), an accurate residue amplifier is necessary. However, realizing such an amplifier in scaled CMOS is challenging due to the worsened transistor characteristics. Prior works focused on gain calibration techniques to mitigate the use of low-gain amplifiers, in return of system complexity and prolonged startups. In this paper, we introduce a digital amplifier (DA) technique to realize power-efficient and accurate amplification in scaled CMOS. DA cancels out all errors (i.e., gain error, nonlinearity, incomplete settling, power supply noise, and thermal noise) of the low-gain amplifier by feedback based on successive approximation. The DA accuracy can be arbitrary set by configuring the number of bits in the DA capacitor digital-to-analog-converter; the amplifier gain is decoupled from the transistor intrinsic gain which is suitable for scaled CMOS integration. We also show that the power efficiency can be enhanced over conventional opamp-based designs with relaxed settling error requirements of DA-based multiplying digital-to-analog-converters (MDACs). Moreover, the circuit design of DA-based MDACs is further discussed. Measurement results of the calibration-free 0.7-V 12-bit 160-MS/s pipelined-SAR ADC implemented in 28-nm CMOS are reported. Without calibration, the ADC achieves signal-to-noise-and-distortion-ratio=61.1 dB, figure-of-merit=12.8 fJ/conv., which is over 3x improvement compared with conventional calibration-free high-speed pipelined ADCs. In addition, we evaluate the DA's process scalability by comparing the measured results of the DA-based MDAC prototyped in 65- and 28-nm CMOS. We observe 2x-3x improvement in speed, power, and area mainly resulting from the DA's process scalability.
- 12 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 27 (11) , 2575-2586
Digital Amplifier: A Power-Efficient and Process-Scaling Amplifier for Switched Capacitor Circuits
To realize high-resolution pipelined and pipelined-SAR analog-to-digital-converters (ADCs), an accurate residue amplifier is necessary. However, realizing such an amplifier in scaled CMOS is challenging due to the worsened transistor characteristics. Prior works focused on gain calibration techniques to mitigate the use of low-gain amplifiers, in return of system complexity and prolonged startups. In this paper, we introduce a digital amplifier (DA) technique to realize power-efficient and accurate amplification in scaled CMOS. DA cancels out all errors (i.e., gain error, nonlinearity, incomplete settling, power supply noise, and thermal noise) of the low-gain amplifier by feedback based on successive approximation. The DA accuracy can be arbitrary set by configuring the number of bits in the DA capacitor digital-to-analog-converter; the amplifier gain is decoupled from the transistor intrinsic gain which is suitable for scaled CMOS integration. We also show that the power efficiency can be enhanced over conventional opamp-based designs with relaxed settling error requirements of DA-based multiplying digital-to-analog-converters (MDACs). Moreover, the circuit design of DA-based MDACs is further discussed. Measurement results of the calibration-free 0.7-V 12-bit 160-MS/s pipelined-SAR ADC implemented in 28-nm CMOS are reported. Without calibration, the ADC achieves signal-to-noise-and-distortion-ratio=61.1 dB, figure-of-merit=12.8 fJ/conv., which is over 3x improvement compared with conventional calibration-free high-speed pipelined ADCs. In addition, we evaluate the DA's process scalability by comparing the measured results of the DA-based MDAC prototyped in 65- and 28-nm CMOS. We observe 2x-3x improvement in speed, power, and area mainly resulting from the DA's process scalability.
- 1 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
A 12.4TOPS/W, 20% Less Gate Count Bidirectional Phase Domain MAC Circuit for DNN Inference Applications
A small gate count 8 bit bidirectional phase domain MAC (PMAC) circuit is proposed for DNN inference applications targeting IoT edge. PMAC consumes significantly smaller power than standard fully digital MACs, owing to its efficient analog accumulation nature based on Gated-Ring-Oscillator (GRO). Compared with the previous first PoC of PMAC, the bidirectional architecture proposed in this paper achieves 20% less gate count, which is comparable with fully digital MACs, and relaxes system design constraints by eliminating phase error originating in leakage current. Asynchronous readout technique and 2-step DTC for the better system throughput and compact implementation, respectively, are presented for the first time. The PMAC achieves peak efficiency of 12.4 TOPS/W in 28 nm CMOS.
- 1 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
A 12.4TOPS/W, 20% Less Gate Count Bidirectional Phase Domain MAC Circuit for DNN Inference Applications
A small gate count 8 bit bidirectional phase domain MAC (PMAC) circuit is proposed for DNN inference applications targeting IoT edge. PMAC consumes significantly smaller power than standard fully digital MACs, owing to its efficient analog accumulation nature based on Gated-Ring-Oscillator (GRO). Compared with the previous first PoC of PMAC, the bidirectional architecture proposed in this paper achieves 20% less gate count, which is comparable with fully digital MACs, and relaxes system design constraints by eliminating phase error originating in leakage current. Asynchronous readout technique and 2-step DTC for the better system throughput and compact implementation, respectively, are presented for the first time. The PMAC achieves peak efficiency of 12.4 TOPS/W in 28 nm CMOS.
- 24 September 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 53 (11) , 3026-3038
A 20-ch TDC/ADC Hybrid Architecture LiDAR SoC for 240 $\times$ 96 Pixel 200-m Range Imaging With Smart Accumulation Technique and Residue Quantizing SAR ADCThis paper presents a time-to-digital converter/analog-to-digital-converter (TDC/ADC) hybrid LiDAR system-on-chip (SoC) to realize reliable self-driving systems. The smart accumulation technique (SAT) is proposed to achieve both 200-m and high-pixel-resolution range imaging, which was untrodden with conventional LiDARs. The “smart” accumulation is realized by a simple object recognition strategy with small circuit overhead. When compared to conventional accumulations, the LiDAR range is enhanced without degrading the pixel resolution. Moreover, a TDC/ADC hybrid architecture is proposed to achieve a wide-distance-range LiDAR with a small silicon area and short-range precision. To minimize the ADC cost, a residue-quantizing noise-shaping (RQNS) SAR ADC is proposed. The prototype LiDAR SoC is fabricated in the 28-nm CMOS technology and integrated into the silicon photomultiplier (SiPM)-based LiDAR system. LiDAR measured with 240 $\times$ 96 pixels at 10 frames/s achieves a measurement range of 200 m with a 70-klx direct sunlight: the measurement range is 2 $\times$ longer than conventional designs. Furthermore, our LiDAR achieves 4 $\times$ higher effective pixel resolution compared to conventional designs using simple accumulation. A 3-D point-cloud image acquired with a real-life environment is presented.
- 24 September 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 53 (11) , 3026-3038
A 20-ch TDC/ADC Hybrid Architecture LiDAR SoC for 240 $\times$ 96 Pixel 200-m Range Imaging With Smart Accumulation Technique and Residue Quantizing SAR ADCThis paper presents a time-to-digital converter/analog-to-digital-converter (TDC/ADC) hybrid LiDAR system-on-chip (SoC) to realize reliable self-driving systems. The smart accumulation technique (SAT) is proposed to achieve both 200-m and high-pixel-resolution range imaging, which was untrodden with conventional LiDARs. The “smart” accumulation is realized by a simple object recognition strategy with small circuit overhead. When compared to conventional accumulations, the LiDAR range is enhanced without degrading the pixel resolution. Moreover, a TDC/ADC hybrid architecture is proposed to achieve a wide-distance-range LiDAR with a small silicon area and short-range precision. To minimize the ADC cost, a residue-quantizing noise-shaping (RQNS) SAR ADC is proposed. The prototype LiDAR SoC is fabricated in the 28-nm CMOS technology and integrated into the silicon photomultiplier (SiPM)-based LiDAR system. LiDAR measured with 240 $\times$ 96 pixels at 10 frames/s achieves a measurement range of 200 m with a 70-klx direct sunlight: the measurement range is 2 $\times$ longer than conventional designs. Furthermore, our LiDAR achieves 4 $\times$ higher effective pixel resolution compared to conventional designs using simple accumulation. A 3-D point-cloud image acquired with a real-life environment is presented.
- 1 February 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 92-94
A 20ch TDC/ADC hybrid SoC for 240×96-pixel 10%-reflection <0.125%-precision 200m-range imaging LiDAR with smart accumulation technique
Long-range and high-pixel-resolution LiDAR systems, using Time-of-Flight (ToF) information of the reflected photon from the target, are essential upon launching safe and reliable self-driving programs of Level 4 and above. 200m long-range distance measurement (DM) is required to sense proceeding vehicles and obstacles as fast as possible in a highway situation. To realize safe and reliable self-driving in city areas, LiDAR systems uniting wide angle-of-view and high pixel resolution are required to fully perceive surrounding events. Moreover, these performances must be achieved under strong background light (e.g., sunlight), which is the most significant noise source for LiDAR systems. To accomplish a 100m-range DM, an accumulation of the DM results through several pixels is utilized to improve the S/N ratio with 70klux background light [1]. Here, S is the number of photons reflected from the target and N as the number of background light photons. However, if the range is extended to 200m under similar condition of the laser power and frame rate (FPS), 16x more pixel accumulation is required. Such pixel accumulation leads to blurring the range image, and hence, a serious oversight in the surrounding events, such as a flying-out pedestrian, may occur, not suiting self-driving applications. Furthermore, the Time-to-Digital Converter (tDC) based ToF measurement is activated only when 2 or more photons are detected simultaneously [1], and thus, is not suitable for the 200m long-range DM where few photons are reflected from the target. On the other hand, ToF measurements using ADCs, which can continuously quantize the silicon photomultiplier (SiPM) output and can sense single-photon events, suits long-range measuring purposes well [2]. However, a number of accumulations should still be required to accomplish 200m-range DM, and hence, low resolution is inevitable. In addition, the SoC cost is critical. To enhance the short-range DM resolution by using ADCs, the required sampling rate is over 10GS/s; upon realizing a 20ch AFE, such an ADC array alone may occupy an area of over 10mm2 and consume huge power [3].
- 1 February 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 92-94
A 20ch TDC/ADC hybrid SoC for 240×96-pixel 10%-reflection <0.125%-precision 200m-range imaging LiDAR with smart accumulation technique
Long-range and high-pixel-resolution LiDAR systems, using Time-of-Flight (ToF) information of the reflected photon from the target, are essential upon launching safe and reliable self-driving programs of Level 4 and above. 200m long-range distance measurement (DM) is required to sense proceeding vehicles and obstacles as fast as possible in a highway situation. To realize safe and reliable self-driving in city areas, LiDAR systems uniting wide angle-of-view and high pixel resolution are required to fully perceive surrounding events. Moreover, these performances must be achieved under strong background light (e.g., sunlight), which is the most significant noise source for LiDAR systems. To accomplish a 100m-range DM, an accumulation of the DM results through several pixels is utilized to improve the S/N ratio with 70klux background light [1]. Here, S is the number of photons reflected from the target and N as the number of background light photons. However, if the range is extended to 200m under similar condition of the laser power and frame rate (FPS), 16x more pixel accumulation is required. Such pixel accumulation leads to blurring the range image, and hence, a serious oversight in the surrounding events, such as a flying-out pedestrian, may occur, not suiting self-driving applications. Furthermore, the Time-to-Digital Converter (tDC) based ToF measurement is activated only when 2 or more photons are detected simultaneously [1], and thus, is not suitable for the 200m long-range DM where few photons are reflected from the target. On the other hand, ToF measurements using ADCs, which can continuously quantize the silicon photomultiplier (SiPM) output and can sense single-photon events, suits long-range measuring purposes well [2]. However, a number of accumulations should still be required to accomplish 200m-range DM, and hence, low resolution is inevitable. In addition, the SoC cost is critical. To enhance the short-range DM resolution by using ADCs, the required sampling rate is over 10GS/s; upon realizing a 20ch AFE, such an ADC array alone may occupy an area of over 10mm2 and consume huge power [3].
- 25 October 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 51 (12) , 3125-3136
A 5.5 mW ADPLL-Based Receiver With a Hybrid Loop Interference Rejection for BLE Application in 65 nm CMOS
This paper presents a low-power hybrid-loop receiver (RX) with high-interference tolerance for Bluetooth low energy (BLE). The hybrid-loop structure based on an all-digital phase-locked loop enables the RX to both enhance the interference tolerance and digitize the frequency-modulated signal without an ADC. A novel single channel receiving method, which enables the conversion of the constellation from frequency shift keying to differential binary phase shift keying signal, is adopted to eliminate the Q-channel signal processing to reduce the power consumption. The prototype RX fabricated in a 65 nm CMOS technology consumes only 5.5 mW and fulfills the BLE requirements of the adjacent channel rejection and out-of-band blocker tolerance without exception. The sensitivity level is −90 dBm.
- 25 October 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 51 (12) , 3125-3136
A 5.5 mW ADPLL-Based Receiver With a Hybrid Loop Interference Rejection for BLE Application in 65 nm CMOS
This paper presents a low-power hybrid-loop receiver (RX) with high-interference tolerance for Bluetooth low energy (BLE). The hybrid-loop structure based on an all-digital phase-locked loop enables the RX to both enhance the interference tolerance and digitize the frequency-modulated signal without an ADC. A novel single channel receiving method, which enables the conversion of the constellation from frequency shift keying to differential binary phase shift keying signal, is adopted to eliminate the Q-channel signal processing to reduce the power consumption. The prototype RX fabricated in a 65 nm CMOS technology consumes only 5.5 mW and fulfills the BLE requirements of the adjacent channel rejection and out-of-band blocker tolerance without exception. The sensitivity level is −90 dBm.
- 29 November 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 47 (12) , 3160-3171
A 2-Gb/s Throughput CMOS Transceiver Chipset With In-Package Antenna for 60-GHz Short-Range Wireless Communication
This paper presents a fully integrated CMOS 60-GHz transceiver chipset for short-range and high-speed wireless communication. The target application of the short-range communication is less than 5-cm communication distance and more than 2-Gb/s throughput for file transfer. In order to achieve file transfer, physical (PHY) layers with error packet correction and media access control (MAC) layer with frame-exchanging function are implemented. The MAC is designed to have a high-efficiency feature due to short interval DATA/Acknowledgement (ACK) frame exchange. It is realized by fast transmitter (TX)/receiver (RX) switching. A bonding wire-based in-package antenna is adopted using a standard BGA package without any off-chip 60-GHz components. The proposed chipset is composed of two chips, an RF chip with in-package antenna, and a baseband chip including PHY and MAC layer. The fabricated chipset achieves 2.62-Gb/s PHY data rate, 2.07-Gb/s MAC throughput, and energy consumption of 651 pJ/bit in 3-cm distance.
- 29 November 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 47 (12) , 3160-3171
A 2-Gb/s Throughput CMOS Transceiver Chipset With In-Package Antenna for 60-GHz Short-Range Wireless Communication
This paper presents a fully integrated CMOS 60-GHz transceiver chipset for short-range and high-speed wireless communication. The target application of the short-range communication is less than 5-cm communication distance and more than 2-Gb/s throughput for file transfer. In order to achieve file transfer, physical (PHY) layers with error packet correction and media access control (MAC) layer with frame-exchanging function are implemented. The MAC is designed to have a high-efficiency feature due to short interval DATA/Acknowledgement (ACK) frame exchange. It is realized by fast transmitter (TX)/receiver (RX) switching. A bonding wire-based in-package antenna is adopted using a standard BGA package without any off-chip 60-GHz components. The proposed chipset is composed of two chips, an RF chip with in-package antenna, and a baseband chip including PHY and MAC layer. The fabricated chipset achieves 2.62-Gb/s PHY data rate, 2.07-Gb/s MAC throughput, and energy consumption of 651 pJ/bit in 3-cm distance.