Position Estimation Using Channel Information and Machine Learning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional position estimation techniques require high costs due to the need for devices that simultaneously perform wireless communication with multiple base stations and have high time resolution.

Innovation Solution

A position estimation device that uses a wireless communication control unit to transmit commands for wireless signals to fixed terminals or position estimation targets, acquiring channel information for radio wave propagation and converting it into input features for a machine learning-based position estimation model.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional position estimation techniques are used with multiple base stations and high time resolution devices, then position estimation accuracy is improved, but device cost and complexity increase

Engineering Contradiction:

Improveposition estimation accuracyVSAvoiddevice complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces channel information as an intermediary that mediates between the wireless signal and position estimation. Instead of directly measuring position with complex timing devices, the system uses channel information (which reflects radio wave propagation characteristics) as a mediator to infer position through machine learning, thereby reducing device complexity while maintaining accuracy

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical/timing-based position estimation system with an information-processing system. Instead of using high-precision timing devices and complex signal processing hardware, the system substitutes a machine learning model that processes channel information to estimate position, thereby reducing hardware complexity

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If conventional position estimation techniques are used with multiple base stations and high time resolution devices, then position estimation accuracy is improved, but operational cost increases

Engineering Contradiction:

Improveposition estimation accuracyVSAvoidoperational cost

Core Design Contradiction:

Measurement precisionVSEase of manufacture

Solution Approach 1:

The patent employs standard wireless communication devices and general-purpose processors instead of expensive specialized position estimation equipment. By using readily available, lower-cost components that can be mass-produced, the system reduces operational cost while achieving accurate position estimation through software-based machine learning

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The patent changes the approach from time-based parameters to channel information parameters. Instead of measuring arrival times with high-precision timing, the system utilizes channel information parameters (such as channel state information) that can be obtained from standard wireless communication, thereby reducing operational cost

Inventive Principle:

Principle #35Parameter changes

3Ease of manufacture

If standard wireless communication devices are used instead of specialized high-performance devices, then device cost is reduced, but position estimation capability is limited

Engineering Contradiction:

Improvedevice costVSAvoidposition estimation accuracy

Core Design Contradiction:

Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent creates a virtual model of the physical environment through machine learning. By training a model with labeled position data and channel information, the system creates a computational copy of the position-channel relationship that can accurately estimate positions using standard devices, thereby compensating for the limited hardware capabilities

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent makes standard wireless communication devices multi-functional by enabling them to perform both wireless communication and position estimation using the same hardware components. The machine learning model allows the device to extract position information from channel information obtained during normal communication, eliminating the need for specialized position estimation hardware

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20240373392A1Position estimation apparatus, position estimation method, and program

Publication Date: 2024.11.07 NIPPON TELEGRAPH & TELEPHONE CORP

AI summary

The position estimation device includes a wireless communication control unit that determines a transmission command and transmits the transmission command to a wireless communication unit of a fixed terminal installed in the same environment as a host device or a wireless communication unit of a position estimation target, a wireless communication unit that receives a wireless signal or the wireless communication unit based on the transmission command and acquires channel information regarding radio wave propagation from the wireless signal, an input feature amount generation unit that converts the channel information into an input feature amount inputtable to a position estimation model, and a position estimation model using unit that estimates and calculates a position of the position estimation target by inputting the input feature amount to a position estimation model obtained by modeling a relationship between channel information and position information by machine learning.