Wireless Quality Estimation for Base Station Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing wireless LAN connection methods rely solely on reception power of beacon signals, failing to account for interference, congestion, and bandwidth, leading to unpredictable throughput and quality after connection.

Innovation Solution

A communication quality estimation system using machine learning to predict throughput by analyzing past data of wireless environment information and communication quality, generating a model to estimate quality before connection.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If connection is determined only by reception power intensity, then connection establishment is simple, but communication quality (throughput) cannot be guaranteed

Engineering Contradiction:

Improveconnection establishment simplicityVSAvoidcommunication quality

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The system performs preliminary estimation of communication quality using a learning model before the terminal actually connects to the base station. The model predicts throughput based on wireless environment information (reception power, interference, congestion, bandwidth) collected during scanning, allowing the terminal to evaluate potential connection quality in advance and avoid connecting to low-quality networks.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

A learning model acts as an intermediary between the raw wireless environment information and the connection decision. The model processes multiple parameters (reception power, interference level, congestion status, bandwidth) and outputs a predicted throughput value, serving as a mediator that translates complex environmental data into a meaningful quality metric for connection selection.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Loss of information

If quality information is transmitted in beacon signal, then congestion can be recognized before connection, but direct throughput prediction is not possible

Engineering Contradiction:

Improvecongestion information availabilityVSAvoidthroughput prediction accuracy

Core Design Contradiction:

Loss of informationVSMeasurement precision

Solution Approach 1:

The system uses actual communication quality data (throughput measurements) from previously connected terminals as feedback to continuously train and improve the learning model. This feedback loop allows the model to learn the relationship between beacon signal information (congestion levels) and actual throughput, progressively improving prediction accuracy based on real-world performance data.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system transforms the available beacon signal parameters (congestion information, number of connected terminals) into a different parameter space that directly predicts throughput. Instead of using congestion information alone for decision-making, the learning model processes these parameters and outputs a predicted throughput value, changing the parameter representation from indirect congestion metrics to direct quality predictions.

Inventive Principle:

Principle #35Parameter changes

3Productivity

If signal processing techniques (beamforming, MIMO) are used, then throughput is improved, but reception power no longer matches throughput

Engineering Contradiction:

ImprovethroughputVSAvoidreception power correlation

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system replaces the traditional mechanical/physical measurement approach (reception power as the sole metric) with a data-driven computational approach (machine learning model). Instead of relying on the direct physical correlation between signal strength and quality, the system uses the learning model to compute predicted throughput based on multiple parameters, substituting the simple physical measurement with a complex computational prediction that accounts for modern signal processing techniques.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS20250350993A1Communication quality estimation system, communication quality estimation method and program

Publication Date: 2025.11.13 NIPPON TELEGRAPH & TELEPHONE CORP

AI summary

There is provided a communication quality estimation system including: a learning unit that performs learning of a model which receives wireless environment information as an input and outputs a communication quality, based on a plurality of pieces of data which are recorded each time one or more terminals connect to a certain base station and perform communication and include a set of the wireless environment information and the communication quality of the terminal on which the communication is performed; and an estimation unit that estimates a communication quality in a case where a terminal is connected to a certain base station by inputting, to the model on which learning is performed, the wireless environment information of the terminal for which the certain base station is set as a connection candidate. Thereby, it is possible to estimate a communication quality in a case where connection to a certain base station is established.