Fast Fourier Transform Offset Estimation for GPS Signal Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining time or spatial offsets between signals require significant computational resources, particularly in applications like GPS receivers, where reducing processor speed, power, and time are essential.
Innovation Solution
The approach exploits the sparse nature of cross-correlation by using Fast Fourier Transforms and subsampling transform values, resulting in a reduced computation complexity of O(N√{square root over (log N)} for GPS locking, and O(N) for high SNR, by employing time aliasing and selective correlation computations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional FFT-based cross-correlation is used to determine time offset, then measurement precision is maintained, but computation resources (processor speed, power, time) are excessively consumed
Solution Approach 1:
The patent segments the cross-correlation computation by dividing the search space into coarse and fine stages. The coarse stage uses a reduced search space with step size of 4 chips, while the fine stage refines the estimate with step size of 1 chip. This segmentation reduces the total number of correlations computed while maintaining measurement precision through progressive refinement.
Solution Approach 2:
The patent applies partial action by computing correlations only at selected offset positions rather than all possible positions. The coarse search computes correlations at every 4th chip position, and the fine search computes correlations only in the vicinity of the coarse peak. This partial computation approach significantly reduces processing requirements while maintaining adequate measurement precision.
2Reliability
If full cross-correlation computation is performed to ensure accurate offset determination, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent segments the correlation computation into two phases: a coarse phase that quickly identifies the general region of the peak, and a fine phase that precisely locates the peak within that region. This segmentation maintains reliability by ensuring the true peak is not missed while reducing device complexity through selective computation only in the refined search region.
Solution Approach 2:
The patent performs preliminary action by conducting the coarse search first to identify the approximate peak location before performing the computationally intensive fine search. This preliminary identification of the search region allows the fine search to be confined to a small window, reducing overall complexity while maintaining reliability through the two-stage verification process.
Data Source
AI summary
An offset estimator (e.g., a time delay, a spatial image offset, etc.) makes use of a transform approach (e.g., using Fast Fourier Transforms). The sparse nature of a cross-correlation is exploited by limiting the computation required in either or both of the forward and inverse transforms. For example, only a subset of the transform values (e.g., a regular subsampling of the values) is used. In some examples, an inverse transform yields a time aliased version of the cross-correlation. Further processing then identifies the most likely offset of the original signals by considering offsets that are consistent with the aliased output.