Multi-band Wi-Fi Fusion for Indoor Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional WLAN sensing methods for indoor localization face challenges due to measurement instability and coarse granularity of RSSI, and require access to PHY-layer interfaces and high computational power for CSI processing, while also necessitating dedicated hardware installations, which are costly and impractical for many applications.
Innovation Solution
The fusion of fine-grained CSI measurements from sub-6 GHz and mid-grained beam SNRs at 60 GHz using an autoencoder-based fusion network, which constructs a feature space for a class-dependent fingerprinting database without relying on external sensors, allowing for infrastructure-free monitoring and reducing the need for labeled training data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If CSI measurements are used for WLAN sensing, then measurement precision is improved, but device complexity and computational power requirements increase
Solution Approach 1:
The patent combines CSI measurements from sub-6 GHz bands with beam SNR measurements from mmWave band (60 GHz) into a unified multi-band feature space. This merging allows the system to leverage the complementary characteristics of different frequency bands: sub-6 GHz provides fine-grained channel information while mmWave provides spatial beam information, achieving enhanced sensing accuracy without requiring exclusive access to PHY-layer interfaces at all bands.
Solution Approach 2:
The system creates a universal sensing framework that can process multiple types of Wi-Fi measurements (CSI at sub-6 GHz and beam SNR at mmWave) through a single autoencoder-based fusion network. This multi-functional approach enables the system to utilize various measurement types from different frequency bands for common sensing tasks like indoor localization and occupancy detection, reducing the need for dedicated processing paths for each measurement type.
2Ease of operation
If RSSI measurements are used for indoor localization, then ease of operation is improved, but measurement precision deteriorates
Solution Approach 1:
The patent merges MAC layer RSSI measurements with PHY layer CSI and beam SNR measurements into a comprehensive multi-band feature space. The autoencoder fusion network processes these different granularities together, allowing the system to retain the ease of MAC layer access while incorporating the precision of PHY layer measurements. The fusion network learns to combine these heterogeneous measurements effectively, producing accurate localization results.
3Measurement precision
If dedicated sensors (UWB, LIDAR, radar) are installed for indoor localization, then measurement precision is improved, but device complexity and installation cost increase
Solution Approach 1:
The patent makes existing Wi-Fi infrastructure multi-functional by enabling it to perform sensing tasks (localization, occupancy detection, pose estimation) that traditionally required dedicated sensors. The autoencoder-based fusion network processes multi-band Wi-Fi measurements to achieve sensor-level accuracy, allowing Wi-Fi access points and client devices to serve both communication and sensing functions without requiring additional UWB, LIDAR, or radar hardware.
Solution Approach 2:
The system enables Wi-Fi devices to self-service sensing functions using their own built-in multi-band capabilities. The autoencoder fusion network is trained to extract meaningful features directly from Wi-Fi channel measurements, allowing the devices to perform sensing tasks autonomously without external dedicated sensors. This self-service approach eliminates the need for separate sensing hardware installations.
4Measurement precision
If multi-band Wi-Fi fusion is implemented, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent introduces an autoencoder-based fusion network as an intermediary component that bridges different types of measurements (CSI, beam SNR, RSSI) from multiple frequency bands. This intermediary fusion network automatically learns the relationships between different measurement types and constructs a unified feature space, abstracting away the complexity of manual feature engineering and multi-source data integration.
Solution Approach 2:
The system transforms different types of measurements with different granularities (sub-6 GHz CSI, mmWave beam SNR, MAC layer RSSI) into a unified parameter space through the autoencoder fusion network. The network learns to map these heterogeneous parameters into a common representation that preserves the essential information from each source while enabling consistent processing and comparison across all measurement types.
Data Source
Figure 1
Figure 2A~2B
Figure 2C~2E
AI summary
A system for fusion of Wi-Fi measurements from multiple frequency bands to monitor indoor and outdoor space is provided. The system includes a multi-band wireless network comprising a set of radio devices to provide coverage in an environment, wherein the set of radio devices are configured to establish wireless communication or sensing links over multi-band wireless channels, wherein the multi-band wireless channels use a first radio band at a millimeter wavelength and a second radio band at a centimeter wavelength. The system further includes a computing processor communicatively coupled to the set of radio devices and a data storage, wherein the data storage has data comprising a parameterized model, modules and executable programs. The computing processor is configured to receive measurement data over the multi-band wireless channels to obtain a set of heterogeneous sensor data, network transferred data, or wireless channel attribute data, fuse at least two types of measurements from the first and the second radio bands at one or more steps in a parameterized model to generate an estimated environmental state in the environment.