Differential Correlator Filter for On-Chip ToF Peak Finding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional processing of time-of-flight (ToF) imager histograms is computationally intensive, requiring off-chip processing that increases complexity, cost, and introduces latency, making it difficult to integrate efficiently into ToF imagers.
Innovation Solution
A differential correlator filter is implemented as a finite-impulse response (FIR) filter to process histograms, comprising regions with specific coefficient configurations to efficiently extract target information, allowing on-chip processing and reducing computational complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional algorithms are used to process histograms from SPAD arrays, then accurate target detection is achieved, but computational complexity and processing time increase significantly
Solution Approach 1:
The histogram processing is segmented into distinct operational phases: peak detection phase using derivative calculation to identify local maxima, and distance calculation phase using median phase computation. This segmentation allows each phase to use optimized algorithms appropriate to its specific task, reducing overall computational complexity while maintaining detection accuracy.
Solution Approach 2:
The invention extracts only the essential features from the histogram data needed for target detection - specifically peak locations and phase information - rather than processing the entire histogram with complex algorithms. This extraction approach maintains measurement precision by focusing on critical data points while significantly reducing computational burden.
2Loss of information
If conventional histogram processing algorithms are used, then comprehensive target information is extracted, but processing latency and power consumption increase
Solution Approach 1:
The system performs preliminary processing by calculating the derivative of the histogram before peak detection, which pre-computes useful information that simplifies subsequent peak identification. This preliminary action reduces the computational work needed in the main processing stage, thereby reducing processing latency while maintaining complete target information extraction.
Solution Approach 2:
The invention changes the parameter representation from raw histogram counts to derivative values and phase angles. This parameter transformation simplifies the mathematical operations required for peak detection and distance calculation, reducing processing time and power consumption while preserving all necessary target information through the mathematical relationships between the transformed parameters and original data.
3Power
If off-chip processing is used for histogram analysis, then computational power is sufficient, but device integration and processing speed are limited
Solution Approach 1:
The patent implements dynamic processing capabilities directly on the sensor chip, allowing the system to adaptively process histogram data in real-time based on incoming photon events. This dynamic on-chip processing eliminates the need for static off-chip processing architectures, achieving sufficient computational power through efficient in-situ calculations while maximizing device integration and improving processing speed through reduced data transfer requirements.
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
The differential correlator filter enables efficient on-chip processing of ToF imager histograms, reducing device complexity, latency, and I/O demands, while directly providing target location information through zero-crossing points.
Implementation Method 1
A reflected photon may generate a carrier in the SPAD through the photo electric effect.
Implementation Method 2
The photon-generated carrier may trigger an avalanche current in one or more of the SPADs in an SPAD array.
Data Source
AI summary
A differential correlator filter includes: a pre-pulse region, where first filter coefficients in the pre-pulse region have negative values; and a pulse region including: a rising edge region adjacent to the pre-pulse region, where second filter coefficients in the rising edge region have positive values; an accumulation region adjacent to the rising edge region, where third filter coefficients of the accumulation region have positive values; and a falling edge region adjacent to the accumulation region, where fourth filter coefficients of the falling edge region have positive values, where the accumulation region is between the rising edge region and the falling edge region. The differential correlator filter further includes a post-pulse region adjacent to the pulse region, where the pulse region is between the pre-pulse region and the post-pulse region, where fifth filter coefficients of the post-pulse region have negative values.