Wi-Fi Indoor Localization via RSSI Clustering and Decision Trees
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wi-Fi based indoor localization systems face challenges in achieving accurate and robust localization without extensive supervised training and trilateration, particularly in complex indoor environments where signal propagation models are difficult to establish effectively.
Innovation Solution
A computer-implemented system using a model-driven approach that eliminates the need for supervision and training, employing k-means clustering, decision trees, and geometry modules to determine user location based on Received Signal Strength Indicators (RSSI) without requiring specific geometric positions of access points, allowing for self-learning and adaptability to environmental changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If trilateration technique is used for indoor localization, then geometric positioning can be achieved, but it requires Wi-Fi access points to be placed in specific geometric positions which reduces adaptability to different environments
Solution Approach 1:
The patent changes the fundamental parameters of the localization approach by replacing geometric trilateration with signal strength-based clustering. Instead of relying on precise geometric positions and distance calculations, the system uses RSSI values and k-means clustering to identify location patterns, making the system adaptable to any environment without requiring specific geometric configurations of access points
Solution Approach 2:
The patent substitutes the mechanical/geometric trilateration system with a data-driven clustering system. Rather than using physical geometric relationships between access points and user devices, the system employs k-means clustering on RSSI data to determine location, eliminating the need for specific geometric placements
2Measurement precision
If location fingerprinting with fine-grained supervised training is used, then high accuracy and resolution can be achieved, but extensive training and database generation are required which increases system complexity
Solution Approach 1:
The patent implements self-service by using unsupervised k-means clustering that automatically learns location patterns from RSSI data without requiring manual supervised training. The system self-organizes the data into clusters that correspond to physical locations, eliminating the need for extensive manual database generation and training while maintaining high accuracy
Solution Approach 2:
The patent extracts only the essential features needed for localization by using k-means clustering on RSSI data, removing the complexity of extensive supervised training databases while retaining the core functionality of accurate location determination
3Measurement precision
If signal propagation models are developed for complex indoor environments, then localization can be achieved, but it is extremely difficult to build an effective general model that coincides with real world situations
Solution Approach 1:
The patent performs preliminary clustering of RSSI data to identify location patterns before actual localization occurs. By pre-processing the data through k-means clustering and storing cluster centers and radii, the system prepares location information in advance, enabling fast and accurate localization without requiring complex real-time signal propagation modeling
Solution Approach 2:
The patent uses simple, computationally inexpensive k-means clustering instead of complex signal propagation models. This lighter approach achieves comparable or better accuracy in diverse environments without the computational burden of maintaining and applying sophisticated propagation models
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system achieves accurate and robust indoor localization without manual supervision or explicit training, enabling scalability and adaptability to new environments, and effectively handles regions that cannot be triangulated using trilateration.
Implementation Method 1
a measure of the power level that a Radio Frequency (RF) device, such as a Wi-Fi or a 3G client, receives from a radio infrastructure
Data Source
AI summary
The present disclosure envisages a computer implemented system and method for Wi-Fi based indoor localization. The system includes a repository for storing attributes of the floor plan of an indoor area with respect to the zones on the floor plan. A communicating module receives a threshold number of data points from user devices located in the area. These data points include a plurality of Received Signal Strength Indicators (RSSI) captured from the access points positioned in the area. A k-means clustering is then performed on the data points for grouping the data points into âkâ number of clusters and a decision tree is built by following a condition based approach. Distance values are then calculated pertaining to the RSSIs stored at the decision tree, and zone circles are plotted. Zone of user presence is then determined by correlating the plotted zone circles upon the floor plan using maximum overlap property.