Dual-Femtosecond Laser Ranging FPGA Peak Detection With Improved NAG
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing dual-femtosecond laser ranging systems face challenges with insufficient pulse peak detection accuracy and excessive hardware resource consumption, necessitating improvements in precision and resource utilization.
Innovation Solution
A data processing method using an improved Nesterov Accelerated Gradient (NAG) algorithm on an FPGA implementation, which includes discrete sampling, parameter optimization, and a modular pipeline design to reduce resource occupation while enhancing accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the BFGS algorithm is used for pulse peak detection, then ranging accuracy is improved, but FPGA resource consumption increases significantly
Solution Approach 1:
The patent transforms the BFGS algorithm into a NAG algorithm by changing key parameters and mathematical formulations. Specifically, it replaces the BFGS update rules with NAG's momentum-based update mechanism, and converts matrix operations into scalar operations, thereby reducing computational complexity and FPGA resource requirements while maintaining sub-micron ranging accuracy
Solution Approach 2:
The patent extracts and removes the matrix operation components from the BFGS algorithm that consume excessive FPGA resources. By eliminating the need for matrix multiplication and inversion operations, the implementation achieves significant resource reduction while preserving the essential optimization functionality through scalar-based NAG algorithm
2Measurement precision
If the BFGS algorithm is implemented on FPGA, then pulse peak detection accuracy is improved, but hardware resource occupation increases
Solution Approach 1:
The patent substitutes complex matrix mechanical operations with simpler scalar arithmetic operations. By replacing matrix-vector multiplications and matrix inversions with scalar-based momentum updates, the hardware implementation becomes significantly simpler while maintaining detection accuracy, thereby reducing FPGA resource occupation
Solution Approach 2:
The patent segments the optimization algorithm into discrete scalar operations that can be efficiently implemented in hardware. By breaking down the continuous matrix operations into step-by-step scalar calculations with explicit momentum updates, the implementation achieves better hardware utilization and reduced resource requirements
Data Source
AI summary
The disclosure is related to a data processing method and system based on an FPGA implementation using an improved NAG algorithm for a dual-femtosecond laser ranging system. It addresses low ranging accuracy and inefficient FPGA resource use. First, the system acquires pulse signal timestamps and values. Second, optimal learning rate and decay rate parameters are pretrained via grid search in software. Third, the improved NAG algorithm fits the pulse envelope on the FPGA using these parameters, computing the pulse peak time. Finally, the peak time is substituted into the optical distance formula to derive the distance. This approach significantly enhances ranging accuracy while reducing FPGA resource consumption, making it suitable for high-precision applications like atmospheric monitoring and lidar.