Memory Optimized GNSS Correlator Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The European Galileo satellite navigation system's 4 ms long pseudorandom number (PRN) code poses challenges for correlator architecture, requiring higher MIPS processing and potentially larger memory, making it difficult to implement efficiently in hybrid correlators that also support other constellations like GPS and GLONASS.
Innovation Solution
The solution involves dividing the 4 k point correlation for Galileo into four 1 k point correlations, distributing processing evenly and reducing the need for per-channel partial sum memory by performing correlations in stages, allowing for a similar MIPS requirement across all frames and maintaining a correlator size comparable to traditional approaches.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the correlator processes the full 4 ms Galileo PRN code in a single cycle, then correlation accuracy is maintained, but processing complexity and memory requirements increase significantly
Solution Approach 1:
The patent divides the 4 ms Galileo PRN code correlation into multiple 1 ms segments. Each segment processes a portion of the correlation (e.g., first half and second half of the code), allowing the full correlation to be completed through multiple passes without requiring all data to be held in memory simultaneously. This segmentation reduces peak memory requirements and processing complexity while maintaining full correlation accuracy.
2Reliability
If the correlator uses a larger memory to store per-channel partial sums for Galileo, then correlation performance is improved, but device size and power consumption increase
Solution Approach 1:
The patent segments the correlation process into multiple passes over smaller data portions. Instead of requiring a single large memory to hold all partial sums for the full 4 ms code, the system uses smaller memory buffers that are reused across multiple processing passes. Each pass handles a segment of the correlation, and results are accumulated over time, reducing the peak memory requirement while maintaining correlation performance.
Solution Approach 2:
The patent employs periodic processing where the correlator performs multiple cycles of correlation on segmented data portions. The same memory structures are reused periodically across different processing passes, with data being written and read in a cyclic manner. This periodic action allows the system to maintain correlation performance without requiring proportionally larger memory, as the same physical memory resources are reused over time.
3Productivity
If the hybrid correlator is optimized for Galileo's 4 ms code, then Galileo correlation efficiency is improved, but compatibility with other constellations like GPS and GLONASS deteriorates
Solution Approach 1:
The patent uses a segmented correlation approach that can adapt to different code lengths. By dividing the correlation into standardized time segments (e.g., 1 ms intervals), the same hardware architecture can handle both the longer 4 ms Galileo codes and the shorter 1 ms GPS/GLONASS codes. The system simply adjusts the number of segments and processing passes based on the constellation being tracked, maintaining efficiency for Galileo while preserving compatibility with other systems.
Solution Approach 2:
The patent designs a universal correlator architecture that can process multiple GNSS constellations using the same segmented correlation technique. The correlator is configured to handle different code lengths and structures by adjusting processing parameters rather than requiring dedicated hardware for each constellation. This multi-functional design allows the system to optimize for Galileo when needed while maintaining full compatibility with GPS, GLONASS, and other systems.
Data Source
AI summary
A personal navigation device includes a correlator for processing GNSS signals from a constellation of satellites A signal is received from a navigation beacon containing a repeating code word, in which the code word includes a number N of samples corresponding to N phases, and in which reception of each code word occurs within a defined time period T. The sequence of N code samples is correlated with a known code word to determine a maximum value of correlation for a particular phase of the received signal. The correlation is performed using a correlator of size M, in which M is less than N, such that N/M=P complete correlations for a partial code phase are performed such that each correlation of a partial code phase is performed within a time period of approximately T/P. All P correlations of partial code phases are completed within time T.