Wireless Roaming Access Point Selection by Predicted Data Rate
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network and access point selection methods for wireless devices rely on physical attributes like signal strength or SNR, which do not guarantee higher throughput, leading to suboptimal connections and unreliable connectivity.
Innovation Solution
A predictive data rate system using machine learning models to evaluate potential connections, predicting data rates based on physical and link layer measurements, allowing for informed network and access point selection to avoid buffering and timeouts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If network selection is based on physical attributes like signal strength or SNR, then the selection process is simple and fast, but the connection performance is suboptimal and unreliable
Solution Approach 1:
The system performs preliminary actions by conducting roam scans to gather physical and link layer measurements about candidate access points before actually connecting to them. This allows the device to predict future connection performance based on pre-collected data, resolving the contradiction by preparing information in advance without delaying the actual connection decision.
Solution Approach 2:
The patent introduces machine learning models as an intermediary between physical measurements and connection decisions. These models process the gathered measurements (signal strength, SNR, channel data) and translate them into predicted data rates, serving as a mediator that bridges simple physical attributes with reliable performance prediction.
2Measurement precision
If the device collects detailed physical and link layer measurements for all access points, then prediction accuracy is improved, but the time required for network selection increases
Solution Approach 1:
The system applies partial action by selectively gathering measurements only from access points that are both visible and potentially suitable, rather than exhaustively measuring all possible access points. The roam scan mechanism efficiently filters candidates based on basic criteria like signal strength thresholds, reducing the measurement burden while maintaining adequate prediction accuracy.
Solution Approach 2:
The patent implements continuity of useful action by performing roam scans periodically and continuously updating the database of access point measurements. This allows the system to leverage previously collected data for predictions, avoiding redundant measurements and reducing real-time selection time while maintaining high prediction accuracy through ongoing data accumulation.
3Reliability
If the system uses machine learning models to predict data rates, then connection reliability is improved, but device complexity increases
Solution Approach 1:
The machine learning models are trained and stored within the wireless device itself, enabling self-service prediction without requiring external server computations. The device uses its own processors to execute the models locally, transforming complex external dependency into internal self-sufficiency while maintaining high prediction reliability.
Solution Approach 2:
The patent applies parameter changes by adapting the machine learning models to different device configurations, firmware versions, and hardware capabilities. The models can be updated or selected based on specific device parameters, allowing the system to maintain high reliability across diverse devices while managing complexity through parameter-based model selection rather than universal complex implementations.
Data Source
AI summary
The technology automatically selects a wireless network connection for a wireless device such as a laptop or a mobile phone based on predicted data rate. The wireless device is configured to receive physical and link layer measurements of access points. Based on the measurements, the wireless device predicts a data rate for each access point prior to forming a network connection therewith. The predicted data rate estimates a rate at which data is transferred from a transceiver of the wireless device to the access point once the wireless network connection is formed. The wireless device selects an access point with a highest predicted data rate, and the transceiver can then be instructed to form a connection with the selected access point.