Neural Network Indoor Positioning via Wireless Signal Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional GPS trilateration techniques face challenges in indoor environments due to signal blocking by obstacles and varying wireless signal propagation characteristics, leading to inaccurate device location determination.
Innovation Solution
The use of machine learning techniques, specifically neural networks, to estimate device location by learning environmental characteristics from wireless signal data, reducing human intervention through unsupervised or semi-supervised learning, and providing reliable distance estimates from anchor nodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If GPS trilateration techniques are used for location determination, then outdoor positioning can be achieved, but the system fails in indoor environments and areas with signal blocking
Solution Approach 1:
The system employs multiple positioning methods (GPS, Wi-Fi, cellular, Bluetooth) that can function across different environments. Each positioning technology serves as a complementary function, allowing the system to operate universally whether outdoors with GPS satellites visible or indoors using local wireless infrastructure.
Solution Approach 2:
When GPS signals are blocked by buildings or terrain, the system introduces intermediate positioning technologies such as Wi-Fi access points and cellular towers as mediators. These intermediaries provide reference points for trilateration when direct satellite-to-device communication is obstructed, enabling continuous positioning capability.
2Ease of operation
If conventional trilateration is used with wireless signals in indoor environments, then positioning can be provided, but accuracy deteriorates due to varying propagation characteristics
Solution Approach 1:
The system dynamically adjusts positioning parameters based on environmental conditions. When operating indoors, it changes from GPS-based parameters to Wi-Fi or cellular signal parameters, and further adjusts by selecting appropriate propagation models that account for indoor multipath effects, wall penetration losses, and signal reflection characteristics specific to the environment.
Solution Approach 2:
The positioning system continuously adapts to changing environmental conditions by dynamically selecting and reconfiguring positioning methods. As the device moves between indoor and outdoor environments or encounters different building structures, the system dynamically switches between GPS, Wi-Fi, cellular, and Bluetooth positioning, adjusting signal processing parameters in real-time to maintain accuracy.
3Measurement precision
If GPS signals are used for positioning, then satellite-based location determination is achieved, but signal blocking by obstacles prevents operation in many practical scenarios
Solution Approach 1:
The system introduces intermediate wireless infrastructure (Wi-Fi access points, cellular towers, Bluetooth beacons) as mediators between the positioning device and the environment. These intermediaries reflect, relay, or provide alternative positioning signals when direct satellite-to-device paths are blocked by buildings, terrain, or other obstacles, enabling positioning in urban canyons and indoor environments.
Solution Approach 2:
The positioning system implements multiple positioning functions simultaneously - GPS satellite-based positioning for outdoor open areas, Wi-Fi positioning for indoor environments, cellular positioning for broader coverage areas, and Bluetooth positioning for localized precision. This multi-functional approach ensures operational versatility across all environment types while maintaining high accuracy in each specific context.
Data Source
AI summary
Systems, methods, computer program products, and apparatuses to determine, by a neural network based on training data related to wireless signals exchanged by a device and a plurality of wireless access points in an environment, a respective distance between the device and each wireless access point, receive location data related to a respective location of each wireless access point of the plurality of wireless access points, determine a geometric cost of the neural network based on a geometric cost function, the respective distances, and the received location data, and train a plurality of values of the neural network based on a backpropagation and the determined geometric cost.