# US 2024/0004050 A1 · Ultra-wideband intelligent sensing system and method for car states detection

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2024/0004050 A1
- Document ID: US20240004050A1
- Status: Published application
- Kind: A1
- Application number: 18/465,650
- Filed: 2023-09-12
- Published: 2024-01-04
- Earliest claimed: 2019-03-29
- Assignees: Robert Bosch GmbH
- Applicants: Robert Bosch GmbH
- Inventors: Yunze ZENG; Avinash KALYANARAMAN; Sushanta Mohan RAKSHIT; Vivek JAIN
- Web page: https://patentgrep.com/patent/US20240004050A1

## CPC

- G01S 13/0209: {Systems with very large relative bandwidth, i.e. larger than 10 %, e.g. baseband, pulse, carrier-free, ultrawideband}
- G07C 5/08: Registering or indicating performance data other than driving, working, idle, or waiting time, with or without registering driving, working, idle or waiting time
- B60R 25/24: using electronic identifiers containing a code not memorised by the user
- G01S 7/41: using analysis of echo signal for target characterisation; Target signature; Target cross-section
- G01S 13/003: {Bistatic radar systems; Multistatic radar systems}
- G01S 13/04: Systems determining presence of a target (based on relative movement of target G01S13/56)
- G06N 20/00: Machine learning
- G07C 9/00309: {operated with bidirectional data transmission between data carrier and locks}
- H01Q 1/3241: {particular used in keyless entry systems}
- H01Q 5/25: Ultra-wideband [UWB] systems, e.g. multiple resonance systems; Pulse systems
- G07C 2209/63: Comprising locating means for detecting the position of the data carrier, i.e. within the vehicle or within a certain distance from the vehicle
- G01S: RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G07C: TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- B60R: VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- H01Q: ANTENNAS, i.e. RADIO AERIALS (radiators or antennas for microwave heating H05B6/72)

## Abstract

A system and method is disclosed for determining a particular vehicle state based on a UWB signal received at a plurality of receiving nodes. A plurality of channel-impulse responses (CIRs) may be computed from the UWB signal received from the plurality of receiving nodes. A plurality of peak-based features based on a selected position and amplitude may be extracted from the plurality of CIRs. A plurality of correlation-based features may be generated by correlating the plurality of CIRs to a corpus of reference CIRs relating to a plurality of vehicle states. A plurality of maximum likelihood vehicle matrices may be generated by correlating the plurality of CIRs to the corpus of reference CIRs relating to the plurality of vehicle states. The vehicle state may then be determined by processing the plurality of peak-based features and correlation-based features using the machine learning classification algorithm.

## Claims

### Claim 1

A system operable to determine a plurality of vehicle state, the system comprising:
a transmitting node operable to transmit an ultra-wide band (UWB) signal;
a plurality of receiving nodes operable to receive the UWB signal transmitted by the transmitting node;
a controller in communication with the transmitting node and the plurality of receiving nodes, wherein the controller is operable to:
receive first, second, and third UWB signals at the plurality of receiving nodes;
compute a plurality of channel-impulse responses (CIRs) from the UWB signals;
generate a plurality of vehicle matrices by correlating the plurality of CIRs to a corpus of reference CIRs relating to a plurality of vehicle states;
determine a first vehicle state corresponding to an occupancy of a vehicle based on the plurality of vehicle matrices;
determine a second vehicle state that is different than the first vehicle state based on the plurality of vehicle matrices; and
determine a third vehicle state that is different than the first and second vehicle states based on the plurality of vehicle matrices.

### Claim 2 (depends on 1)

The system of claim 1, wherein the second vehicle state corresponds to a change in a vehicular component.

### Claim 3 (depends on 2)

The system of claim 2, wherein the second vehicle state includes determining a position of a vehicle window and/or door.

### Claim 4 (depends on 2)

The system of claim 2, wherein the third vehicle state corresponds to an object outside the vehicle.

### Claim 5 (depends on 1)

The system of claim 1, wherein one or more vehicles states is determined using a machine learning algorithm.

### Claim 6 (depends on 1)

The system of claim 1, further comprising:
upsampling the plurality of CIRs; and
aligning the plurality of CIRs by shifting a first path within one of the plurality of CIRs to a reference tap.

### Claim 7 (depends on 6)

The system of claim 6, wherein the plurality of CIRs are upsampled using a Fast Fourier Transformation (FFT) algorithm that operates on a time domain of one of the plurality of CIRs.

### Claim 8 (depends on 6)

The system of claim 6, wherein the first path is identified as an event that occurs in the plurality of CIRs.

### Claim 9 (depends on 1)

The system of claim 1, wherein the vehicle matrices are generated during a machine learning process that includes a training phase and testing phase.

### Claim 10

A method of determining vehicle states, the method comprising:
selecting a transmission node and a plurality of receiving nodes based on a connectivity profile;
receiving a plurality of UWB signals at the plurality of receiving nodes;
computing a plurality of channel-impulse responses (CIRs) from the plurality UWB signals received from the plurality of receiving nodes;
extracting from the plurality of CIRs a plurality of peak-based features based on a selected position and amplitude; and
generating a plurality of correlation-based features by correlating the plurality of CIRs to a corpus of reference CIRs relating to a plurality of vehicle states; and
determining a vehicle state by processing the plurality of peak-based features and correlation-based features using a machine learning classification algorithm, the vehicle state including an occupancy of the vehicle, changes to a vehicular component, and an environment outside the vehicle.

### Claim 11 (depends on 10)

The method of claim 10, further comprising:
generating a plurality of maximum likelihood vehicle matrices by correlating the plurality of CIRs to the corpus of reference CIRs relating to the plurality of vehicle states; and
summing the plurality of maximum likelihood vehicle matrices to generate a top-K vehicle state matrix; and
determining the vehicle state by processing the plurality of peak-based features and correlation-based features using the machine learning classification algorithm within the top-K vehicle state matrix.

### Claim 12 (depends on 10)

The method of claim 10, wherein one or more peak-based features of the plurality of peak-based features may be determined based on a ratio of power of a first peak of the plurality of CIRs.

### Claim 13 (depends on 10)

The method of claim 10, wherein one or more peak-based features of the plurality of peak-based features may be determined based on a ratio of power of a top peak of the plurality of CIRs.

### Claim 14 (depends on 10)

The method of claim 10, wherein one or more peak-based features of the plurality of peak-based features may be determined based on a relative tap distance between a first peak of the plurality of CIRs.

### Claim 15 (depends on 10)

The method of claim 10, wherein one or more peak-based features of the plurality of peak-based features may be determined based on a relative tap distance between a top peak of the plurality of CIRs.

### Claim 16 (depends on 10)

The method of claim 10, wherein one or more peak-based features of the plurality of peak-based features may be determined based on a maximum peak power of the plurality of CIRs.

### Claim 17 (depends on 16)

The method of claim 16, wherein the one or more peak-based features may be determined based on a position of the maximum peak power of the plurality of CIRs.

### Claim 18 (depends on 10)

The method of claim 10, wherein the machine learning classification algorithm is generated using a machine-learning process that includes a training phase and a testing phase.

### Claim 19 (depends on 10)

The method of claim 10, wherein the machine learning classification algorithm is a Random Forest classifier.

### Claim 20

A system operable to determine a vehicle state, the system comprising:
a transmitting node operable to transmit an ultra-wide band (UWB) signal;
a plurality of receiving nodes operable to receive the UWB signal transmitted by the transmitting node;
a controller in communication with the transmitting node and the plurality of receiving nodes, wherein the controller is operable to:
compute a plurality of channel-impulse responses (CIRs) from a UWB signal received from the plurality of receiving nodes;
generate a plurality of maximum likelihood vehicle matrices by correlating the plurality of CIRs to a corpus of reference CIRs relating to a plurality of vehicle states; and
determine a position of an object outside a vehicle.
