Location Spoofing Detection Using Confidence Scores
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current techniques fail to effectively identify and mitigate location spoofing by user equipment (UE) in multi-access edge computing (MEC) networks, leading to inaccuracies in location data used for critical services like vehicle-to-everything communications and fleet management, which can cause operational issues and safety concerns.
Innovation Solution
Implementing a system that utilizes both statistical models and machine learning models to process location data from UEs and the core network, determining a confidence interval and score to validate the authenticity of location data, thereby identifying and mitigating spoofing both on client devices and in transit.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current techniques are used to handle location data, then processing is simple and fast, but location spoofing cannot be effectively identified leading to data inaccuracies
Solution Approach 1:
The patent introduces statistical models and machine learning models as intermediary components between location data collection and service delivery. These models act as mediators that analyze location data from multiple UEs, compute confidence intervals and scores, and determine whether location spoofing is present before the data is used by services. This intermediary layer enables accurate spoofing detection without requiring changes to the underlying data collection infrastructure.
Solution Approach 2:
The system performs preliminary analysis of location data by computing confidence intervals and spoofing scores before the data is used by applications. The statistical and machine learning models evaluate location data in advance, determining the likelihood of spoofing before services make decisions based on location information. This preliminary action prevents spoofed data from affecting service operations.
2Reliability
If statistical models and machine learning models are implemented to identify spoofing, then location data integrity is improved, but computing resources and processing time increase
Solution Approach 1:
The patent applies partial action by using statistical models for initial assessment and reserving machine learning models for cases where statistical analysis is inconclusive or when higher confidence is required. The system computes confidence intervals using statistical methods first, and only invokes more computationally intensive machine learning models when necessary. This selective application of analysis methods achieves reliable spoofing detection while optimizing computing resource usage.
3Ease of operation
If location spoofing is not identified, then services operate without intervention, but safety-critical services like vehicle-to-everything communications may be compromised
Solution Approach 1:
The patent implements feedback mechanisms where the statistical and machine learning models continuously evaluate location data and provide information about spoofing likelihood. When spoofing is detected or suspected, the system provides feedback to services to adjust their operations or reject suspicious data. This feedback loop enables services to respond dynamically to potential spoofing threats while maintaining normal operation when data is reliable.
Data Source
AI summary
A device may receive user equipment (UE) location data identifying a latitude and a longitude of a UE, as reported by the UE, and may receive network location data identifying a latitude and a longitude of the UE, as reported by a core network. The device may process the UE location data and the network location data to determine a confidence score associated with an actual location of the UE. The device may process the UE location data and the network location data, when the confidence score is outside a confidence interval, to determine whether the UE location data is valid. The device may perform one or more actions based on determining whether the UE location data is valid.