Predicting Wi-Fi Access Points via Cellular Signal Fingerprints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cellular network communications devices face battery drain and poor data transmission due to frequent scanning for Wi-Fi access points, especially when signal strength is low, which affects user experience and bandwidth availability.
Innovation Solution
A wireless availability prediction device gathers cellular fingerprints to create classes and trains a predictive model to classify new fingerprints, enabling devices to predict access point availability and quality, thereby conserving battery life and optimizing data offloading.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a Wi-Fi radio frequently scans for access points, then the device can find available access points for data offloading, but the device's battery is drained
Solution Approach 1:
The system performs preliminary actions by gathering cellular fingerprints and training a predictive model in advance. This model predicts access point availability before the device needs to scan, allowing the Wi-Fi radio to remain off until prediction indicates an access point is available, thus avoiding frequent scanning while still enabling timely data offloading
Solution Approach 2:
The system uses cellular network signals (which the device already receives for normal operation) to gather fingerprints and train the predictive model. This self-service approach leverages existing infrastructure without requiring additional active scanning or external resources, solving the energy efficiency problem while maintaining productivity
2Duration of action of moving object
If a device connects to access points with low signal strength, then data transmission can continue, but data loss and slow transmission occur resulting in poor user experience
Solution Approach 1:
The predictive model performs preliminary assessment of access point quality by analyzing cellular fingerprints and predicted signal strength before the device attempts connection. This allows the system to identify and connect to high-quality access points in advance, avoiding poor connections that would cause data loss and slow transmission
Solution Approach 2:
The system incorporates feedback mechanisms where successful handoffs generate new fingerprints that are added to the training set, continuously improving the predictive model's accuracy in assessing access point quality and reliability
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
Examples relate to predicting available access points. In one example, a computing device may: obtain a set of training fingerprints, each training fingerprint specifying, for a client device, i) a set of access points, and ii) cellular signal strength measurements for each cellular tower in a set of cellular towers; generate a plurality of classes based on the set of training fingerprints, each class specifying at least one access point, the access points of each class corresponding to the set of access points specified by at least one training fingerprint, and each combination being different from combinations specified by each other class in the plurality of classes; and train a predictive model to receive, as input, an input fingerprint specifying a cellular signal strength measurement for each cellular tower in a set of input cellular towers and produce, as output, at least one of the plurality of classes.