Market Context — Why This Technology, Why Now

The global push for digital transformation (DX) and Industry 4.0 demands intelligent, adaptable sensing technologies. Simultaneously, increasing regulatory scrutiny on data privacy and the need for worker safety in automated environments are paramount. This technology addresses these converging pressures by offering a robust, privacy-preserving solution for real-time operational intelligence, driving efficiency and compliance across diverse industries.

Key Competitive Advantages
01

Enables high-precision non-contact motion estimation, even with obstructions, by fusing video and CSI data, ensuring stable data acquisition in challenging environments.

02

Enhances privacy protection and installation flexibility by utilizing camera-free CSI, reducing privacy risks and expanding monitoring system applicability across diverse environments.

03

Accelerates AI model development by up to 50% through automated training data generation via synchronized video and CSI, reducing time-to-market and development costs.

Market Opportunity
Smart Factories
$2.5B–$5B globally (AI est.)
The need for productivity improvement and safety management makes AI-driven anomaly detection and worker monitoring essential.
Industrial automation solution providers Large-scale manufacturing corporations Robotics and AGV manufacturers
Elder Care & Healthcare
$2.5B–$5B globally (AI est.)
Privacy-conscious non-contact sensing is highly sought after for elder monitoring and fall detection.
Smart home health device manufacturers Assisted living facility operators Remote patient monitoring system developers
Smart Offices & Buildings
$1.5B–$2.5B globally (AI est.)
CSI-based sensing contributes to optimizing space utilization through occupancy detection, traffic flow analysis, and enhanced security.
Commercial real estate management firms Building automation system integrators Smart office technology providers
IP Defensibility — Why Competitors Can't Replicate This
What This Patent Covers

This patent protects an innovative approach to object motion estimation through the synergistic use of video and CSI data for training data generation, and subsequent CSI-based learned model generation. It covers a broad scope across systems, methods, and programs, having overcome examiner rejections to establish clear and robust claims, indicating a highly reliable and defensible right.

Competitive White Space

This patent primarily covers the method of generating AI models from synchronized video and CSI data for motion estimation. White space exists in developing novel CSI hardware architectures, integrating this technology with other sensor modalities for enhanced environmental awareness, or creating advanced predictive maintenance algorithms based on the motion data.

Economic Impact
~$1M/year estimated operational cost reduction and 1.3x productivity increase per facility (est.).
estimated ROI · USD · AI analysis
ROI Calculation Logic

Traditional camera surveillance systems for equipment and worker monitoring in factories incurred annual operational costs of ~$33.5K/unit (AI est.). This technology could reduce operational costs by ~20% (~$6.5K/unit annually, AI est.) due to privacy considerations and reduced installation effort. Furthermore, real-time high-precision monitoring could shorten production line downtime by 50 hours annually, avoiding ~$1M in losses (AI est., assuming ~$20K/hour loss). This is estimated to improve productivity by 1.3x.

Speed to Market
4x faster than in-house development
This technology has established a logic for generating training data by synchronizing video and CSI data, with foundational algorithms for learned model generation already in place. This eliminates the need for licensees to conduct research and development from scratch, allowing them to focus on integration and validation with existing wireless communication infrastructure and video processing systems. The proof-of-concept phase is significantly shortened, potentially reducing time-to-market by approximately 2.7 years.
Competitive Positioning

X: Ease of Deployment & Cost Efficiency
Y: Motion Estimation Accuracy & Privacy Protection

Business Models & Applications
💻 Software License Provision
Provide AI model generation software incorporating this technology as SaaS or a package, enabling licensees to build learned models with their own data.
💡 Motion Estimation Solution Provision
Develop specialized motion estimation systems for specific industries (e.g., manufacturing, elder care) and offer them as customized solutions to licensees.
📊 Data Analysis & Consulting
Offer high-precision motion analysis services based on CSI and video data, developing a consulting business to support licensees' operational improvements and decision-making.
Adjacent Application Opportunities
🚗 Autonomous Driving & MaaS
In-Cabin Occupant Monitoring System
This system could use CSI data to non-contact detect occupant posture and abnormal behavior with high precision within autonomous vehicles. It could enable early detection of drowsy driving or sudden illness, enhancing safety and facilitating automatic intervention in emergencies, potentially reducing accident risks by over 15% while preserving in-cabin privacy.
⚽ Sports & Fitness
Non-Contact Form Analysis Coach
Analyzes subtle changes in an athlete's form in real-time using CSI data, quantifying body axis sway and center of gravity shifts often missed by cameras. This AI coaching tool could provide high-precision feedback, potentially improving athletic performance by 10-20% in private settings.
災害・セキュリティ
Rubbled Survivor Detection System
Detects subtle movements (e.g., breathing, heartbeat) of survivors trapped under rubble in disaster zones using wireless CSI. This system could estimate survivor location and condition even when cameras or acoustic sensors fail, potentially increasing rescue efficiency and success rates by over 30%.
Integration Roadmap — Estimated 12-Month Deployment
Technology Evaluation & PoC
Duration: 3 months
Evaluate compatibility with the licensee's existing systems (wireless communication environment, video data) and verify the basic performance and effects of this technology through a small-scale Proof of Concept (PoC).
Model Development & System Integration
Duration: 6 months
Based on PoC results, develop a learned model tailored to the licensee's specific requirements. Proceed with API integration and data flow design and integration into existing information processing systems.
Full-Scale Deployment & Operational Optimization
Duration: 3 months
Implement the developed system for full-scale operation in the field. Continuously collect data and gather feedback to improve model accuracy and optimize the overall system.
Technical Feasibility
This technology utilizes CSI data from existing wireless communication infrastructure and standard video data, requiring no significant new equipment investment. The patent claims specifically detail the method for generating training data by time-aligning CSI and video data. This logic can be integrated as a software module into existing information processing systems with high compatibility, as it is based on general-purpose data processing techniques, making the technical barrier relatively low.
Success Scenario
Upon adopting this technology, manufacturing lines could detect subtle equipment vibrations or unusual worker movements from CSI data in real-time, which traditional camera surveillance might miss. This could lead to early detection of equipment failures and reduced occupational hazard risks, potentially cutting downtime by ~20% annually and improving productivity by 1.3x.
Patent Record
APPLICATION NO.
特願2021-031824
REGISTRATION NO.
7624205
FILING DATE
2021/03/01
GRANT DATE
2025/01/22
EXPIRATION DATE
2041/03/01
PATENT HOLDER
国立大学法人静岡大学
Examination History
2024年02月13日
出願審査請求書
2024年12月03日
拒絶理由通知書
2024年12月20日
意見書
2024年12月20日
手続補正書（自発・内容）
2025年01月07日
特許査定