Systolic Processor ROIC for LiDAR Time-of-Flight Precision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing light ranging systems, such as LiDAR, face challenges in achieving high resolution time-of-flight measurements due to increased power dissipation and data volume with the number of pixels, as well as dead time lag between photon detections, which limits their precision and efficiency.
Innovation Solution
A systolic processor system with a Read-Out Integrated Circuit (ROIC) that uses systolic processing to capture and analyze time-of-flight data from Avalanche Photodiode arrays, sharing a common timer across rows to reduce power consumption and eliminate dead time, allowing for precise detection and analysis of photon arrival times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the number of pixels in the light detector array is increased to improve measurement precision, then the resolution of time-of-flight measurements is improved, but power dissipation and data volume increase
Solution Approach 1:
The system segments the time-of-flight measurement process into discrete time bins, with each pixel in the detector array independently timing photons within its assigned time bin. This segmentation allows parallel processing across multiple pixels without requiring centralized timing logic for each pixel, reducing overall power consumption while maintaining high measurement precision across the entire array.
Solution Approach 2:
The system uses a common timer that is activated only when needed (when photons are detected) rather than continuously running timers for each pixel. This partial action approach reduces power dissipation by eliminating unnecessary timing operations while still achieving high resolution measurements when photons are actually detected.
2Measurement precision
If the number of pixels in the light detector array is increased to improve measurement precision, then the resolution of time-of-flight measurements is improved, but the volume of data to be processed increases
Solution Approach 1:
The systolic processor array segments the data processing task, with each processing element handling data from a specific row of pixels. This segmentation allows data to be processed in parallel streams, reducing the total time to process large volumes of data from high-resolution arrays while maintaining measurement precision.
Solution Approach 2:
The system adds a temporal dimension to data processing by using a time-based architecture where data flows through the systolic processor array over multiple clock cycles. This dimensional transformation allows the system to handle large data volumes from high-resolution arrays by distributing processing over time rather than requiring all data to be processed simultaneously, effectively managing data volume without sacrificing measurement precision.
3Measurement precision
If traditional timing circuits are used for each pixel to achieve precise photon detection, then measurement precision is improved, but dead time lag between detections occurs
Solution Approach 1:
The system merges the timing function into a shared common timer that is accessed by multiple pixels through the systolic processor array, rather than having separate timing circuits in each pixel. This merging eliminates the dead time lag associated with individual pixel timers because the common timer can service multiple pixels in sequence without each pixel needing its own dedicated timing resources, while still maintaining precise photon detection through the coordinated operation of the systolic array.
4Measurement precision
If individual timing circuits are provided for each pixel to achieve precise time-of-flight measurement, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system merges multiple timing functions into a single common timer that is shared across all pixels in the array. This merging dramatically reduces device complexity by eliminating the need for individual timing circuits in each pixel, while the systolic processor array provides the coordination mechanism to maintain precise time-of-flight measurements using the shared timer resource.
Solution Approach 2:
The common timer is designed as a universal timing resource that can service any pixel in the array through the systolic processing architecture. This multi-functional timer replaces multiple specialized individual timers, reducing device complexity while maintaining the ability to achieve precise time-of-flight measurements across all pixels in the array.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the resolution and efficiency of time-of-flight measurements by reducing power dissipation and dead time, enabling higher density arrays and more precise distance measurements in LiDAR systems.
Implementation Method 1
A receiver circuit includes a Read-Out Integrated Circuit (ROIC) containing one or more light detectors, such as Avalanche Photodiode (APDs)
Implementation Method 2
Avalanche Photodiode (APDs) arranged in an array
Implementation Method 3
The ROIC is configured to measure at least a time of flight starting when at a first reference time and at a second reference point
Data Source
AI summary
A ROIC can perform systolic processing of light detectors. The ROIC performs the systolic processing of the light detectors to capture at least i) when, in time units, an initial photon of its reflected light pulse is captured by each of the light detectors in the array, ii) where geographically in terms of column and row address of the light detector capturing its photon is located in the array, iii) scan out data captured by the light detectors on the when in time units, and the where geographically that the photon was captured in a given light detector in the array, and then iv) analyze the data on the when and the where with an algorithm to know exactly when exactly, in terms of time units, the photon was captured relative to the input from the clock circuit in order to determine an objects characteristics.