A Reconfigurable 3-D-Stacked SPAD Imager With In-Pixel Histogramming for Flash LIDAR or High-Speed Time-of-Flight Imaging

Abstract

A 256 x 256 single photon avalanche diode (SPAD) sensor integrated in a 3D-stacked 90nm 1P4M/40nm 1P8M process is reported for flash light detection and ranging (LIDAR) or high speed direct time of flight (ToF) 3D imaging. The sensor bottom tier is composed of a 64x64 matrix of 36.72 m pitch modular photon processing units which operate from shared 4x4 SPADs at 9.18 m pitch and 51% fill-factor. A 16 x 14-bit counter array integrates photon counts or events to compress data to 31.4 Mbps at 30 fps readout over 8 I/O operating at 100 MHz. The pixel-parallel multi-event TDC approach employs a programmable internal or external clock for 0.56 ns to 560 ns time bin resolution. In conjunction with a perpixel correlator, the power is reduced to less than 100 mW in practical daylight ranging scenarios. Examples of ranging and high speed 3D TOF applications are given. Index Terms-3-D imaging, CMOS, direct time of flight (dTOF), histogramming, image sensor, light detection and ranging (LiDAR), single photon avalanche diodes (SPADs), time-to-digital converter (TDC), TDC sharing architecture, TOF.

High-sensitivity multispeckle diffuse correlation spectroscopy

“…Due to the recent investments in LiDAR technology for automotive and consumer electronics applications, high-performance large-pixel-count SPAD cameras with improved detection efficiency are rapidly becoming more available and less expensive. 30,37 While the first megapixel SPAD array was recently reported, 31 for mDCS to take full advantage of the extraordinary advances of SPAD camera technology, we envision the need for real-time data compression schemes implemented in the read-out FPGA of the SPAD camera or directly on chip. 30 Finally, our mDCS technique can also be implemented in the time domain to enhance sensitivity to deep tissue.…”

A 128 × 128 SPAD Motion-Triggered Time-of-Flight Image Sensor With In-Pixel Histogram and Column-Parallel Vision Processor

“…Flash LiDAR based on the principle of time of flight (ToF) is the least computationally demanding technique amongst other ranging methods such as triangulation, interferometry and structured light, and the flood illumination eliminates the mechanical complexity of a scanning emitter. Despite these system advantages, the integration of ToF image sensors in IoT nodes or mobile devices is challenging due to their data volumes exceeding hundreds of kilobytes per frame [12]- [15] to deliver sub-centimeter distance precision at high frame rate and spatial resolution. ToF cameras have been proposed that histogram time-correlated single photon counting data at each pixel to pre-process photon time stamps [12] and in doing so reduce the output data volume for each frame.…”

Numerical Model of SPAD-Based Direct Time-of-Flight Flash LIDAR CMOS Image Sensors

“…On the other hand, it suffers the presence of strong background light, which can be mitigated by the implementation of smart techniques at the pixel-level for the rejection of uncorrelated light. Recent advances in Single-Photon Avalanche Diode (SPAD) detectors in CMOS technology show that high-sensitivity, low-noise devices can be combined with dense logic in small areas taking advantage of advanced process nodes, down to 40 nm [ 8 ], even in combination with 3D stacking [ 9 ], and the race towards a fully autonomous car driving system has raised an enormous interest in SPAD-based LIDAR technology.…”

“…The performance of such systems are not easily predictable, in particular when the design of the single pixel is not trivial and aims at introducing smart features like background rejection [ 10 , 11 , 12 , 13 , 14 ], interference suppression [ 15 ], or partial on-chip memory/histogramming [ 16 , 17 , 18 , 19 ]. Additionally, expensive solutions, like 3D stacking [ 8 , 9 , 20 ] and microlenses [ 21 ], are to be carefully evaluated before being employed. Consequently, it is desirable to tailor the design in order to meet the required specifications avoiding the risk to obtain either poor or overkilling performance.…”