Event Data Timestamp Compression for Faster Stream Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Event cameras generate large event data streams with high storage requirements and processing speed limitations due to the large number of bits occupied by timestamps.
Innovation Solution
A data processing method that compresses timestamps in event data streams using time differences and flag information, along with noise removal techniques to reduce storage space and improve processing efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If complete timestamps are used in event data streams, then time information integrity is maintained, but storage space requirements increase and processing speed decreases
Solution Approach 1:
The patent segments the timestamp representation into two parts: a reference timestamp that can be shared across multiple events, and a time difference value that represents the offset from the reference. This segmentation allows the system to store only the difference values (which are smaller) while maintaining the ability to reconstruct complete timestamps when needed, thus reducing storage space while preserving time information integrity.
Solution Approach 2:
The patent changes the parameter representation from absolute timestamps to relative time differences. By storing time differences (Δt) instead of complete timestamps, the patent reduces the number of bits required per event while maintaining the ability to recover absolute time information through addition of the reference timestamp, thereby resolving the contradiction between storage efficiency and information integrity.
2Loss of information
If complete timestamps are used in event data streams, then time information integrity is maintained, but processing speed decreases
Solution Approach 1:
The patent segments the timestamp processing into a reference component and a difference component. During processing, the system can work with the smaller difference values directly for operations like sorting and filtering, only reconstructing complete timestamps when absolutely necessary. This segmentation accelerates processing speed while maintaining time information integrity through the preserved reference timestamp.
Solution Approach 2:
The patent transforms the time parameter from absolute timestamps to relative time differences, which are computationally more efficient to handle. Operations such as comparing event times, detecting temporal patterns, and sorting become faster when working with small difference values rather than large absolute timestamp values, thereby improving processing speed while preserving the ability to recover exact timing information.
3Quantity of substance
If timestamp compression is applied to event data streams, then storage space is reduced, but device complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-selecting and storing a reference timestamp that serves as a baseline for subsequent events. This reference timestamp is established beforehand and reused for multiple events, allowing the system to store only time differences rather than complete timestamps. This preliminary setup simplifies the overall structure compared to storing full timestamps for every event, reducing storage space without significantly increasing processing complexity.
Solution Approach 2:
The patent uses a copying mechanism where the reference timestamp is effectively copied and applied to multiple events through simple addition of time differences. Instead of storing and processing complete timestamp values for each event, the system copies the reference timestamp concept and applies it efficiently to generate absolute times when needed, reducing storage requirements while maintaining processing simplicity.
Data Source
AI summary
A data processing method includes receiving an event data stream, where the event data stream includes at least a first event data item and a second event data item, the first event data item includes a first timestamp for obtaining the first event data item, the second event data item includes a second timestamp for obtaining the second event data item, and the second event data item is obtained most recently before obtaining the first event data item; and obtaining a compressed event data stream corresponding to the event data stream, where the compressed event data stream includes at least a first compressed event data item corresponding to the first event data item, and the first compressed event data item includes first time information.


