Genetic Optimization of a Software-Defined GNSS Receiver

TL;DR

The paper tackles the challenge of maintaining robust GNSS PVT under high-dynamics by automating tracking-loop tuning using the Hybrid Genetic Optimizer (HyGO) within a software-defined GNSS receiver. It leverages a hardware-in-the-loop SDR testbed and realistic GPS L1 signals to optimize eight tracking- and integration-parameter settings for three dynamic scenarios (static, rocket, LEO). By scalarizing a normalized position- and velocity-error cost, HyGO discovers regime-specific configurations that balance tracking robustness and measurement accuracy, achieving significant error reductions across all cases. The work demonstrates that genetic-optimization-driven configuration of SDR GNSS receivers can deliver reliable navigation under extreme conditions and provides an open, reproducible framework for future research and deployment.

Abstract

Commercial off-the-shelf (COTS) Global Navigation Satellite System (GNSS) receivers face significant limitations under high-dynamic conditions, particularly in high-acceleration environments such as those experienced by launch vehicles. These performance degradations, often observed as discontinuities in the navigation solution, arise from the inability of traditional tracking loop bandwidths to cope with rapid variations in synchronization parameters. Software-Defined Radio (SDR) receivers overcome these constraints by enabling flexible reconfiguration of tracking loops; however, manual tuning involves a complex, multidimensional search and seldom ensures optimal performance. This work introduces a genetic algorithm-based optimization framework that autonomously explores the receiver configuration space to determine optimal loop parameters for phase, frequency, and delay tracking. The approach is validated within an SDR environment using realistically simulated GPS L1 signals for three representative dynamic regimes -guided rocket flight, Low Earth Orbit (LEO) satellite, and static receiver-processed with the open-source GNSS-SDR architecture. Results demonstrate that evolutionary optimization enables SDR receivers to maintain robust and accurate Position, Velocity, and Time (PVT) solutions across diverse dynamic conditions. The optimized configurations yielded maximum position and velocity errors of approximately 6 m and 0.08 m/s for the static case, 12 m and 2.5 m/s for the rocket case, and 5 m and 0.2 m/s for the LEO case.

Figures (8)

• Figure 1: Experimental configuration
• Figure 2: GNSS architecture used for the optimization loop, based on a FLL-assisted PLL.
• Figure 3: Ground truth. First row: trajectory; Second row: velocity; Third row: acceleration. The axes are represented such that () is x-axis, () is y-axis, () is z-axis in ECEF reference frame.
• Figure 4: Schematic overview of the HyGO genetic algorithm optimization loop for GNSS receiver parameter tuning. The iterative workflow comprises population initialization (with Latin Hypercube Sampling), cost evaluation via hardware-in-the-loop simulation, tournament selection, elitism, and convergence. Operational scenarios (static, rocket, LEO) are modeled.
• Figure 5: Distribution of the three main GNSS tracking loop parameters: PLL narrow, DLL narrow, and FLL bandwidth. Left: 3D representation of the full population, illustrating how individuals are spread across the parameter space. Right: Zoomed-in views for each scenario (static, rocket, LEO satellite), where the best individual from each generation is marked with a grayscale dot whose size and intensity increase with the generation number. These best individuals are connected by (), highlighting the trajectory of the optimization across generations. Individuals are represented as ( ) for static, ( ) for rocket, and ( ) for LEO satellite scenarios.