Uplink Sensor Data Compression Using Environment Prediction Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The asymmetry in uplink and downlink time slots in 5G systems, where more data is uploaded than downloaded, leads to degraded networking performance for remote-controlled machines, particularly when realizing digital twins, as there are insufficient uplink time slots for transmitting large amounts of sensor data.
Innovation Solution
A wireless device determines prediction data, including a subset of sensor data and prediction parameters, which is transmitted to a network node to predict current and future sensor data, allowing efficient communication and reducing the data volume.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If more uplink time slots are allocated to transmit sensor data, then the data transmission capacity is improved, but the network asymmetry and resource allocation efficiency deteriorate
Solution Approach 1:
The patent extracts and transmits only the essential elements needed for data reconstruction: a subset of actual sensor data and prediction parameters. The remaining sensor data is generated through prediction algorithms at the network node, eliminating the need to transmit all original data and resolving the contradiction between transmission volume and resource efficiency.
Solution Approach 2:
The patent transforms the transmission approach by changing the parameters being transmitted from raw sensor data to prediction parameters and data subsets. This parameter transformation enables the network node to reconstruct full sensor data streams with reduced uplink transmission requirements, maintaining productivity while improving resource allocation.
2Measurement precision
If all sensor data is transmitted to ensure complete information, then the measurement precision and data accuracy are improved, but the data transmission volume and network load increase
Solution Approach 1:
The patent creates a copy of the sensor data transmission process by separating actual measurements from transmitted data. Instead of transmitting all sensor data, the system transmits a subset and prediction parameters, allowing the network node to generate copies of the remaining data through prediction algorithms, thus maintaining accuracy while reducing transmission volume.
3Productivity
If prediction parameters and data subsets are transmitted instead of all sensor data, then the data transmission efficiency is improved, but the complexity of prediction algorithms increases
Solution Approach 1:
The patent introduces prediction parameters as an intermediary between the wireless device and the prediction algorithm at the network node. These parameters serve as a bridge that enables efficient data reconstruction without requiring complex algorithms at the transmitting device, thus improving transmission efficiency while managing overall system complexity.
4Quantity of substance
If a subset of sensor data is transmitted to reduce data volume, then the transmission bandwidth requirement is reduced, but the information completeness and prediction accuracy may deteriorate
Solution Approach 1:
The patent applies preliminary action by selecting and transmitting a carefully curated subset of sensor data that contains the most critical information for accurate prediction. This preliminary selection ensures that the transmitted data subset, combined with prediction parameters, provides sufficient information for the network node to reconstruct complete sensor data streams accurately, preventing information loss while reducing transmission volume.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method performed by a wireless device, for handling a plurality of sensor data streams related to the wireless device in a wireless communications network is provided. The wireless device obtains (401) sensor data related to the plurality of sensor data streams. Based on the obtained sensor data, the wireless device determines (403) prediction data indicative of how to predict current and/or future sensor data of the plurality of sensor data streams. The prediction data comprises any one or more out of a subset of sensor data relating to a subset of the plurality of sensor data streams, and at least one prediction parameter indicative of a predictability of the current and/or future sensor data of the plurality of sensor data streams. The wireless device transmits (404) an indication of the determined prediction data to a network node.