Mobile Device Positioning via Genetic Algorithm NLOS Error Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wireless positioning technologies face significant errors due to non-line-of-sight (NLOS) issues, which affect the accuracy of time and angle measurements in mobile device positioning, especially in environments where signals encounter reflections and diffractions, leading to positioning errors of up to 589 meters.
Innovation Solution
A method utilizing a genetic algorithm to determine the position of a mobile device in a wireless wide area network by iteratively calculating measurement circles and their radii, and using these to derive the best solution for positioning, thereby reducing the influence of NLOS errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional signal attenuation model or time of arrival measurement is used for positioning, then positioning can be achieved, but positioning accuracy deteriorates due to non-line-of-sight errors reaching up to 589 meters
Solution Approach 1:
The patent applies genetic algorithm to search for the optimal combination of base station selection and weighting factors, converting the harmful NLOS effects into a solvable optimization problem. By iteratively evolving the population of possible solutions, the system identifies the best configuration that minimizes positioning error even in presence of reflections and diffractions
Solution Approach 2:
The patent changes the parameter representation by encoding base station selection and weighting factors as chromosomal genes. The genetic algorithm modifies these parameters across generations through selection, crossover, and mutation operations, transforming the static positioning calculation into a dynamic parameter optimization process that adapts to NLOS conditions
2Measurement precision
If genetic algorithm iterations are performed to reduce NLOS errors, then positioning accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent implements a fixed number of genetic algorithm iterations (e.g., 100 generations) rather than continuing until perfect convergence. This partial action approach achieves sufficient positioning accuracy improvement while avoiding excessive computational burden that would occur with unlimited iterations
Solution Approach 2:
The patent segments the positioning problem into discrete generational steps of genetic algorithm execution. Each generation processes a population of candidate solutions independently, allowing the computational work to be divided into manageable segments that can be executed iteratively with controlled resource consumption
Data Source
AI summary
A method for positioning a mobile device in a wireless wide area network (WWAN) is provided. The method includes determining three measurement circles according to coordinates of three base stations and respectively calculating radiuses of the three measurement circles and distances between the three base stations. The method uses genetic algorithm to derive the best solution of a plurality of variables of an object function and estimates the position of the mobile device according to the best solution. Accordingly, non-line-of-sight (NLOS) errors are reduced, and more accurate positioning can be provided.