Adaptive Sensor Data Compression for Distributed Object Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Distributed sensor networks face challenges in efficiently transmitting sensor data due to high bandwidth requirements in raw-sensor-data sharing (RDS) and data loss in object-based sensor data sharing (ODS), particularly in dense networks where communication technology limitations and incomplete sensor data hinder accurate object detection.

Innovation Solution

A computer-implemented method and device for encoding sensor data using adaptive data compression techniques based on detected object attributes, allowing for optimized data transmission by configuring compression parameters according to the specific characteristics of the objects in the sensor data, and providing object data to enable correct decoding at the receiver.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If raw-sensor-data sharing (RDS) is used to transmit sensor data, then the receiver device can access nearly all information acquired by the sensor for high accuracy processing, but high communication bandwidth is required which may not be satisfied by communication technology particularly in dense sensor networks

Engineering Contradiction:
Improveobject detection accuracyVSAvoidcommunication bandwidth
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent extracts only the essential information needed for object detection from the raw sensor data. Instead of transmitting complete raw sensor data, the system identifies and transmits only the relevant features and parameters required for accurate object detection, thereby reducing communication bandwidth requirements while maintaining detection accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies different processing and transmission strategies to different regions or types of sensor data based on their importance for object detection. Critical regions containing potential objects are processed with higher fidelity and transmitted with more detail, while less critical regions are compressed or aggregated, optimizing the balance between bandwidth usage and detection accuracy.

Inventive Principle:
Principle #3Local quality

2Quantity of substance

If object-based sensor data sharing (ODS) is used to transmit extracted objects and their attributes, then communication bandwidth requirements are reduced, but some useful data is lost and cannot be considered by the receiver device

Engineering Contradiction:
Improvecommunication bandwidthVSAvoiduseful data loss
Core Design Contradiction:
Quantity of substanceVSLoss of information

Solution Approach 1:

The patent performs preliminary processing at the sender device to extract and preserve critical object information before transmission. By pre-identifying and encoding essential object attributes and features in advance, the system ensures that all useful data needed for accurate object detection is captured and transmitted, preventing information loss that would occur with simpler ODS approaches.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary processing layer that bridges between raw sensor data and final object detection. This intermediary layer at the sender device performs intelligent data selection, feature extraction, and encoding to ensure that all relevant information is preserved in the transmitted data, allowing the receiver to reconstruct complete object information without losing useful data.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If object detection technique is applied to sensor data to detect objects and their attributes, then the data can be processed with high accuracy, but additional computational processing is required at the sensor device

Engineering Contradiction:
Improveobject detection accuracyVSAvoidsensor device computational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the computational workload between the sensor device and the receiver device. The sensor device performs only essential preliminary processing to identify and extract critical object information, while more complex detection and analysis operations are performed at the receiver device. This segmentation reduces the computational burden and complexity at the resource-constrained sensor device while maintaining high object detection accuracy.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4456435A1Transmitting sensor data in distributed sensor networks
Publication Date: 2024.10.30 ROBERT BOSCH GMBH
  • EP4456435A1 patent drawingFigure 1
  • EP4456435A1 patent drawingFigure 2
  • EP4456435A1 patent drawingFigure 3A

AI summary

Sensor data may be encoded for transmission to a receiver by applying an object detection technique to the sensor data to detect an object in the sensor data and to determine one or more attributes of the object, encoding the sensor data to obtain encoded sensor data, wherein the encoding comprises applying a data compression technique to the sensor data, wherein the data compression technique is configurable by one or more compression parameters, wherein applying the data compression technique to the sensor data comprises selecting at least one of the one or more compression parameters based on the one or more attributes of the object, generating object data indicative of the one or more attributes of the object, and sending the encoded sensor data and the object data to the receiver to enable the receiver to select one or more decompression parameters of a data decompression technique based on the one or more attributes of the object.