# US 2019/0244125 A1 · System and method for optimal sensor placement

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

- Number: US 2019/0244125 A1
- Document ID: US20190244125A1
- Status: Published application
- Kind: A1
- Application number: 16/263,421
- Filed: 2019-01-31
- Published: 2019-08-08
- Earliest claimed: 2018-02-02
- Assignees: Koninklijke Philips N.V.
- Applicants: Koninklijke Philips N.V.
- Inventors: ALI AKBAR AHMAD SAMADANI; SAMAN PARVANEH
- Web page: https://patentgrep.com/patent/US20190244125A1

## CPC

- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G16H 40/67: for remote operation
- A61B 5/1123: {Discriminating type of movement, e.g. walking or running (A61B5/1116, A61B5/112 take precedence)}
- A61B 5/7264: {Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems}
- G06F 17/18: for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)}
- H04L 67/12: specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
- H04W 4/38: for collecting sensor information
- H04W 24/02: Arrangements for optimising operational condition
- A61B 5/0002: {Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network (endoradiosondes A61B5/07)}
- A61B 5/02416: {using photoplethysmograph signals, e.g. generated by infrared radiation (A61B5/14552 takes precedence)}
- A61B 5/1116: {Determining posture transitions}
- A61B 5/6887: {mounted on external non-worn devices, e.g. non-medical devices}
- A61B 5/7267: {involving training the classification device}
- A61B 2560/0223: of calibration, e.g. protocols for calibrating sensors
- H04W 88/18: Service support devices; Network management devices
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72)

## Abstract

A controller includes a memory that stores instructions and a processor that executes the instructions. The instructions cause the controller to execute a process that includes receiving sensor data from a first sensor and a second sensor. The sensor data includes a time-series observation representing a first activity and a second activity. The controller generates models for each activity involving progressions through states indicated by the sensor data from each sensor. The controller receives from each sensor additional sensor data including a time-series observation representing the first activity and the second activity. The controller determines likelihoods that the models generated a portion of the additional sensor data and calculates a pair-wise distance between each sensor-specific determined likelihood to obtain calculated distances. The calculated distances for each sensor are grouped, and a relevance of each sensor to each activity is determined by executing a regression model using the grouped calculated distances.

## Claims

### Claim 1

A controller for determining an arrangement of sensors for monitoring a space, comprising:
a memory that stores instructions; and
a processor that executes the instructions,
wherein, when executed by the processor, the instructions cause the controller to execute a process comprising:
receiving, from a first sensor of at least two sensors, a first sensor data comprising at least one time-series observation representing at least a first activity and a second activity;
receiving, from a second sensor of the at least two sensors, a second sensor data comprising at least one time-series observation representing the first activity and the second activity;
generating, by the processor, a first model for the first activity involving a first progression through a plurality of states indicated by at least a portion of the first sensor data;
generating, by the processor, a second model for the second activity involving a second progression through a plurality of states indicated by at least a portion of the first sensor data;
generating, by the processor, a third model for the first activity involving a third progression through a plurality of states indicated by at least a portion of the second sensor data;
generating, by the processor, a fourth model for the second activity involving a fourth progression through a plurality of states indicated by at least a portion of the second sensor data;
receiving, from the first sensor, a third sensor data comprising at least one time-series observation representing at least the first activity and the second activity;
receiving, from the second sensor, a fourth sensor data comprising at least one time-series observation representing at least the first activity and the second activity;
determining, using the processor, a likelihood that the first model generated at least a portion of the third sensor data, a likelihood that the second model generated at least a portion of the third sensor data, a likelihood that the third model generated at least a portion of the fourth sensor data, and a likelihood that the fourth model generated at least a portion of the fourth sensor data;
calculating, using the processor, a pair-wise distance between each sensor-specific determined likelihood to obtain calculated distances;
grouping, using the processor, the calculated distances for the likelihoods involving the first sensor, and grouping, using the processor, the calculated distances for the likelihoods involving the second sensor, to obtain grouped calculated distances, and
determining, using the processor, a first relevance of the first sensor and a second relevance of the second sensor for capturing the first activity and the second activity by executing a regression model using the grouped calculated distances,
wherein one of the first sensor and the second sensor is included in an arrangement of sensors used to monitor the space based on the first relevance and the second relevance determined using the processor and the other of the first sensor and the second sensor is excluded in the arrangement of sensors used to monitor the space based on the first relevance and the second relevance determined using the processor.

### Claim 2 (depends on 1)

The controller of claim 1, wherein the process executed by the controller further comprises:
performing one of: activating the first sensor based on the first relevance or removing the first sensor based on the first relevance.

### Claim 3 (depends on 1)

The controller of claim 1, wherein the process executed by the controller further comprises:
performing one of: activating the second sensor based on the second relevance or removing the second sensor based on the second relevance.

### Claim 4 (depends on 1)

The controller of claim 1, wherein the process executed by the controller further comprises:
receiving, from a third sensor of the at least two sensors, a fifth sensor data comprising at least one time-series observation representing the first activity and the second activity;
generating, by the processor, a fifth model for the first activity involving a fifth progression through a plurality of states indicated by at least a portion of the fifth sensor data;
generating, by the processor, a sixth model for the second activity involving a sixth progression through a plurality of states indicated by at least a portion of the fifth sensor data;
receiving, from the third sensor, a sixth sensor data comprising at least one time-series observation representing at least the first activity and the second activity; and
determining, using the processor, a likelihood that the fifth model generated at least a portion of the sixth sensor data, and a likelihood that the sixth model generated at least a portion of the sixth sensor data.

### Claim 5 (depends on 1)

The controller of claim 1, wherein the first sensor comprises a first group of sensors, and wherein the second sensor comprises a second group of sensors.

### Claim 6 (depends on 1)

The controller of claim 1, wherein the first sensor data comprises a first time-series observation representing a first activity and a second time-series observation representing a second activity.

### Claim 7 (depends on 1)

The controller of claim 1, wherein the first model comprises a probabilistic graphical model.

### Claim 8 (depends on 1)

The controller of claim 1, wherein the sensor-specific determined likelihood represents determined likelihoods associated with the first sensor and determined likelihoods associated with the second sensor.

### Claim 9 (depends on 1)

The controller of claim 1, wherein the regression model is a multinomial logistic regression model with a group LASSO penalty.

### Claim 10 (depends on 9)

The controller of claim 9, wherein the multinomial logistic regression model is a binomial logistic regression model with a group LASSO penalty.

### Claim 11 (depends on 1)

The controller of claim 1, wherein the regression model determines weights of the grouped calculated distances such that they best represent the first activity and the second activity.

### Claim 12 (depends on 1)

The controller of claim 1, wherein the process executed by the controller further comprises:
identifying, from the at least two sensors, a minimal set of sensors most salient for sensing the first activity and the second activity.

### Claim 13 (depends on 12)

The controller of claim 12, wherein the minimal set of sensors is selected as a subset of the at least two sensors.

### Claim 14 (depends on 1)

The controller of claim 1, wherein the first model, the second model, the third model and the fourth model each comprise a separate hidden Markov model.

### Claim 15 (depends on 1)

The controller of claim 1, wherein the arrangement is defined by characteristics of the space being monitored.

### Claim 16

A method for determining an arrangement of sensors for monitoring a space, comprising:
receiving, from a first sensor of at least two sensors, a first sensor data comprising at least one time-series observation representing at least a first activity and a second activity;
receiving, from a second sensor of the at least two sensors, a second sensor data comprising at least one time-series observation representing the first activity and the second activity;
generating, by a processor, a first model for the first activity involving a first progression through a plurality of states indicated by at least a portion of the first sensor data;
generating, by the processor, a second model for the second activity involving a second progression through a plurality of states indicated by at least a portion of the first sensor data;
generating, by the processor, a third model for the first activity involving a third progression through a plurality of states indicated by at least a portion of the second sensor data;
generating, by the processor, a fourth model for the second activity involving a fourth progression through a plurality of states indicated by at least a portion of the second sensor data;
receiving, from the first sensor, a third sensor data comprising at least one time-series observation representing at least the first activity and the second activity;
receiving, from the second sensor, a fourth sensor data comprising at least one time-series observation representing at least the first activity and the second activity;
determining, using the processor, a likelihood that the first model generated at least a portion of the third sensor data, a likelihood that the second model generated at least a portion of the third sensor data, a likelihood that the third model generated at least a portion of the fourth sensor data, and a likelihood that the fourth model generated at least a portion of the fourth sensor data;
calculating, using the processor, a pair-wise distance between each sensor-specific determined likelihood to obtain calculated distances;
grouping, using the processor, the calculated distances for the likelihoods involving the first sensor, and grouping, using the processor, the calculated distances for the likelihoods involving the second sensor, to obtain grouped calculated distances; and
determining, using the processor, a first relevance of the first sensor and a second relevance of the second sensor for capturing the first activity and the second activity by executing a regression model using the grouped calculated distances,
wherein one of the first sensor and the second sensor is included in an arrangement of sensors used to monitor the space based on the first relevance and the second relevance determined using the processor, and the other of the first sensor and the second sensor is excluded in the arrangement of sensors used to monitor the space based on the first relevance and the second relevance determined using the processor.

### Claim 17 (depends on 16)

The method of claim 16, further comprising:
arranging a minimized group of sensors to monitor the space based on the at least two sensors, wherein the minimized group excludes at least one of the at least two sensors.

### Claim 18

A system for determining an arrangement of sensors for monitoring a space, comprising:
a communication interface used to communicate over a communications network;
a user interface; and
a controller comprising a memory that stores instructions, and a processor that executes the instructions,
wherein, when executed by the processor, the instructions cause the system to execute a process comprising:
receiving, from a first sensor of at least two sensors, a first sensor data comprising at least one time-series observation representing at least a first activity and a second activity;
receiving, from a second sensor of the at least two sensors, a second sensor data comprising at least one time-series observation representing the first activity and the second activity;
generating, by the processor, a first model for the first activity involving a first progression through a plurality of states indicated by at least a portion of the first sensor data;
generating, by the processor, a second model for the second activity involving a second progression through a plurality of states indicated by at least a portion of the first sensor data;
generating, by the processor, a third model for the first activity involving a third progression through a plurality of states indicated by at least a portion of the second sensor data;
generating, by the processor, a fourth model for the second activity involving a fourth progression through a plurality of states indicated by at least a portion of the second sensor data;
receiving, from the first sensor, a third sensor data comprising at least one time-series observation representing at least the first activity and the second activity;
receiving, from the second sensor, a fourth sensor data comprising at least one time-series observation representing at least the first activity and the second activity;
determining, using the processor, a likelihood that the first model generated at least a portion of the third sensor data, a likelihood that the second model generated at least a portion of the third sensor data, a likelihood that the third model generated at least a portion of the fourth sensor data, and a likelihood that the fourth model generated at least a portion of the fourth sensor data;
calculating, using the processor, a pair-wise distance between each sensor-specific determined likelihood to obtain calculated distances;
grouping, using the processor, the calculated distances for the likelihoods involving the first sensor, and grouping, using the processor, the calculated distances for the likelihoods involving the second sensor, to obtain grouped calculated distances; and
determining, using the processor, a first relevance of the first sensor and a second relevance of the second sensor for capturing the first activity and the second activity by executing a regression model using the grouped calculated distances,
wherein one of the first sensor and the second sensor is included in an arrangement of sensors used to monitor the space based on the first relevance and the second relevance determined using the processor, and the other of the first sensor and the second sensor is excluded in the arrangement of sensors used to monitor the space based on the first relevance and the second relevance determined using the processor.

### Claim 19 (depends on 18)

The system of claim 18, wherein the space comprises an enclosed living space.
