Wireless Node Location Determination Using Probability Ratios
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless LAN (WLAN) systems face challenges in accurately determining the location of wireless nodes relative to defined areas due to factors like shadowing from walls and furniture, and multipath effects in RF environments, which complicates the creation of high-accuracy RF coverage maps.
Innovation Solution
A method using a location server that computes probability ratios based on received signal strength data to determine whether a wireless node is inside or outside a defined perimeter, with the option to bias these ratios and exclude certain regions to improve accuracy and security.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual site surveys and mathematical modeling techniques are used to create RF coverage maps, then coverage information can be obtained, but high accuracy is difficult to achieve due to shadowing and multipath effects
Solution Approach 1:
The patent replaces manual site surveys and complex mathematical modeling (ray tracing) with a probabilistic statistical approach using received signal strength measurements. Instead of using complex physics-based models to predict coverage, the system uses empirical measurements and probability theory to determine node location, substituting mechanical/computational complexity with statistical analysis that is more robust to shadowing and multipath effects
Solution Approach 2:
The system uses the wireless nodes themselves to perform measurements and provide location information. Each node autonomously measures received signal strengths from multiple access points and uses these measurements to determine its own location probability, eliminating the need for external surveyors and complex centralized modeling
2Measurement precision
If probability ratios are computed to determine node location, then location accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent divides the physical space into discrete regions (inside/outside the defined area) and computes probabilities for each region separately. By segmenting the location determination into distinct probability calculations for inside and outside regions, the system achieves accurate location classification without requiring complex continuous positioning algorithms
Solution Approach 2:
The system changes the parameter representation from precise continuous coordinates to discrete probability states (inside/outside). By transforming the location parameter from a continuous spatial coordinate to a discrete probabilistic state, the system achieves high accuracy in determining whether a node is inside or outside a area without the computational burden of precise continuous positioning
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In one embodiment, a method includes computing a probability surface corresponding to the location probability of the wireless node within a physical region based on the received signal strength data associated with a wireless node and an RF model of the physical region; computing, based on the probability surface, an aggregate probability (Pin) of the wireless node being inside a perimeter defined with the physical region! computing, based on the probability surface, an aggregate probability (Pout) of the wireless node being outside the perimeter; computing a probability ratio of the aggregate probabilities Pin to Pout; and determining whether the wireless node is inside or outside the perimeter based on a comparison of Pout and Pin.