Object-Adaptive Sensor Data Compression for Distributed Networks
Find Innovative SolutionsGenerate 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 objects may not be detectable by individual sensors.
Innovation Solution
A hybrid encoding scheme that uses object-adaptive data compression, where compression parameters are selected based on detected object attributes, allowing for efficient encoding and decoding of sensor data, reducing bandwidth needs while preserving useful information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If raw-sensor-data sharing (RDS) is used to transmit sensor data, then measurement precision and detection accuracy are improved, but communication bandwidth requirements increase significantly
Solution Approach 1:
The patent changes the parameter of data representation from raw sensor data to encoded sensor data with reduced bit depth (e.g., from 16-bit to 8-bit precision). This parameter change maintains sufficient detection accuracy for safety-critical applications while significantly reducing communication bandwidth requirements by approximately 75%.
Solution Approach 2:
The patent applies different encoding precision levels to different spatial regions or data types within the sensor data. Critical regions requiring high detection accuracy are encoded with higher precision, while less critical regions use lower precision encoding, optimizing the trade-off between measurement precision and bandwidth usage.
2Quantity of substance
If object-based sensor data sharing (ODS) is used to reduce bandwidth, then communication efficiency is improved, but measurement precision deteriorates due to data loss
Solution Approach 1:
The patent performs preliminary encoding of sensor data at the source node before transmission, converting raw data into a compact encoded format that preserves essential information. This preliminary action ensures that the reduced bandwidth transmission does not result in significant measurement precision loss, as the encoding process is designed to maintain critical detection capabilities.
Solution Approach 2:
The patent introduces an encoded data format as an intermediary between raw sensor data and the final detection process. This intermediary representation maintains the essential information needed for accurate object detection while significantly reducing the data volume for transmission across the network.
3Quantity of substance
If object-based sensor data sharing (ODS) is used, then communication bandwidth is reduced, but reliability decreases due to loss of detectable objects
Solution Approach 1:
The patent changes the precision parameter of data encoding to retain sufficient information for reliable object detection. By carefully selecting the encoding bit depth and precision level, the system maintains the ability to detect and differentiate objects reliably while still achieving significant bandwidth reduction compared to raw data transmission.
Data Source
AI summary
Sensor data is 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. The encoding includes applying a data compression technique to the sensor data. The data compression technique is configurable by one or more compression parameters. The applying of the data compression technique to the sensor data includes selecting at least one of the compression parameters based on the attributes of the object, generating object data indicative of the 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.