Point Cloud Data Prediction for Lower Air Interface Overhead
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The large amount of point cloud data exchanged between sensing and processing devices in communication scenarios leads to significant air interface overheads, necessitating efficient data compression methods to reduce transmission costs.
Innovation Solution
A data processing method that predicts and transmits location information of matching data subsets, along with differences and coefficients, using prediction models to minimize data transmission, thereby reducing air interface overheads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If all point cloud data is transmitted from sensing device to processing device, then data completeness is improved, but air interface overhead increases significantly
Solution Approach 1:
The patent extracts and transmits only the essential components needed for data reconstruction. Specifically, it transmits discrete point data from the point cloud along with prediction model parameters, rather than transmitting the entire point cloud dataset. This extraction approach maintains data completeness for reconstruction purposes while dramatically reducing the quantity of data transmitted over the air interface.
Solution Approach 2:
The patent applies preliminary action by performing prediction model processing at the sensing device before transmission. The device predicts future point cloud states based on historical data and transmits only the necessary prediction parameters and correction data, rather than transmitting all raw point cloud data. This preliminary processing reduces air interface overhead while ensuring the processing device can reconstruct complete point cloud information.
2Quantity of substance
If point cloud data is compressed to reduce transmission size, then air interface overhead is reduced, but data processing complexity increases
Solution Approach 1:
The patent uses copying by transmitting a simplified representation (copy) of the point cloud data through prediction models. Instead of compressing the actual point cloud data points, the system creates a predictive model copy that can generate the necessary data at the receiving end. This approach reduces air interface overhead while avoiding complex compression algorithms, as the prediction model serves as a compact representation that can be efficiently transmitted and executed.
3Quantity of substance
If prediction models are used to predict point cloud data, then data transmission volume is reduced, but prediction accuracy may deteriorate
Solution Approach 1:
The patent implements feedback by transmitting discrete point data from the actual point cloud that serves as ground truth corrections to the prediction model. The processing device receives both the prediction model parameters and discrete point data, uses the discrete points to verify and correct prediction accuracy, and can update the prediction model accordingly. This feedback mechanism ensures prediction accuracy is maintained while still achieving significant reduction in data transmission volume compared to sending complete point cloud data.
Data Source
Figure 1(a)~3
Figure 4~6
Figure 7(a)~7(b)
AI summary
Embodiments of this application provide a data processing method and a communication apparatus, to reduce air interface overheads during data transmission. The method includes: When devices at a transmit end and a receive end both predict an actual sampled data set at a current moment based on a same algorithm by using a reconstructed data set of a first moving object at a previous moment, a sending device may obtain partial sampled data in the actual sampled data set at the current moment through prediction by using the reconstructed data set of the first moving object at the previous moment, and indicate, to a receiving device by using first information, location information of the partial sampled data obtained through prediction. In this way, there is no need to send all sampled data at the current moment, so that a transmission amount of sampled data can be reduced, to reduce air interface overheads.