GNSS Correlation Computation Using Partial Sum Accumulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current correlation computation methods for GNSS receivers require high computational complexity when correlating multiple satellite PRN code sequences with a data sequence, especially when the code sequences are long or numerous, due to the need to try all possible time-shifted versions.
Innovation Solution
A correlation computation method and apparatus that utilize an accumulation-based circuit to obtain sums of all and selected data samples, and a processing circuit to derive correlation values by subtracting partial sums from the total sum, reducing computation complexity by sharing common data and computation across multiple code sequences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional correlation computation methods are used to correlate multiple satellite PRN code sequences with a data sequence, then accurate position and time synchronization can be achieved, but the computational complexity becomes excessively high
Solution Approach 1:
The patent segments the correlation computation process into distinct stages: (1) computing the sum of all data samples in the data sequence, (2) computing partial sums by selecting data samples according to code bits, and (3) deriving correlation values from the difference between total sum and partial sums. This segmentation allows each stage to be optimized independently, reducing overall computational complexity while maintaining synchronization accuracy.
Solution Approach 2:
The patent merges the computation of correlation values for multiple PRN code sequences by sharing common data samples and intermediate summation results. Instead of independently correlating each code sequence with the entire data sequence, the method combines computations by identifying and reusing common operations across multiple correlations, significantly reducing redundant calculations.
2Measurement precision
If all possible time-shifted versions of PRN code sequences are correlated with the data sequence to handle unknown and dynamic timing, then accurate synchronization can be achieved, but the number of multiplication and summation operations increases dramatically
Solution Approach 1:
The patent performs preliminary computation by pre-calculating the sum of all data samples in the data sequence before conducting correlation operations with multiple PRN code sequences. This preliminary sum is reused across all correlation computations, eliminating redundant summation operations and significantly improving computational efficiency when handling unknown and dynamic timing scenarios.
Solution Approach 2:
The patent extracts and reuses common computational elements across multiple correlation operations. Specifically, the sum of all data samples is extracted as a shared intermediate result that is applied to all PRN code sequence correlations, rather than recalculating it for each sequence. This extraction of common operations dramatically reduces the total number of multiplication and summation operations required.
3Reliability
If long PRN code sequences are used to improve satellite signal identification accuracy, then more satellites can be reliably identified, but the computational burden increases due to the longer code length
Solution Approach 1:
The patent applies partial action by computing only the necessary partial sums required for correlation, rather than processing the entire data sequence for each PRN code. By selecting and summing only the data samples that correspond to code bits, the method reduces computational effort while maintaining the reliability benefits of using long PRN code sequences for accurate satellite identification.
Data Source
AI summary
A correlation computation method includes: obtaining a first sum of all data samples included in a data sequence; obtaining a second sum of selected data samples that are selected from the data sequence according to code bits included in a first code sequence; and deriving, by a processing circuit, a first correlation value between the data sequence and the first code sequence from the first sum and the second sum.