Multi-Buffer Packet Compression for Loss-Resilient Decompression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In acknowledged mode (AM) transmission in LTE systems, data packet loss during uplink data compression (UDC) leads to decompression errors at the receive end due to misaligned buffers, resulting in a large number of lost data packets, as the receive end cannot correctly decompress data packets without the lost packet.
Innovation Solution
Implementing multiple buffers for data packets, where each buffer is determined based on the data packet's serial number or randomly, with indication information, to ensure that data packets not corresponding to the lost packet can be decompressed independently, reducing the impact of packet loss.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a single buffer is used for data packet compression, then the compression efficiency is improved, but the reliability of data transmission deteriorates when packet loss occurs
Solution Approach 1:
The patent divides the single buffer into multiple buffers (first buffer, second buffer, third buffer, etc.), each responsible for compressing specific data packets based on their serial numbers. This segmentation ensures that when one buffer experiences packet loss, other buffers can continue to operate independently, thereby improving transmission reliability without sacrificing compression efficiency.
Solution Approach 2:
The patent changes the buffer management approach by introducing multiple buffers with different data packet ranges. Each buffer is assigned to handle specific serial number ranges of data packets, allowing the system to maintain compression efficiency while improving reliability through parallel processing of multiple packet streams.
2Reliability
If multiple buffers are used for data packet compression, then the reliability of data transmission is improved, but the device complexity increases
Solution Approach 1:
The patent applies local quality by assigning specific ranges of data packet serial numbers to specific buffers. Each buffer has a dedicated responsibility area (e.g., first buffer for packets 1-N, second buffer for packets N+1-2N), allowing independent operation and simplifying the management logic for each buffer while maintaining overall system reliability.
Solution Approach 2:
The patent introduces a dimensional approach by organizing buffers based on serial number ranges rather than a single sequential buffer. This creates a multi-dimensional buffer structure where buffers are distributed across different data packet ranges, enabling parallel processing and improving reliability without requiring complex synchronization mechanisms.
3Quantity of substance
If data packets are compressed based on previous packets in the buffer, then the compression ratio is improved, but the loss of a single packet affects all subsequent packets
Solution Approach 1:
The patent segments the data packets into multiple groups, each handled by a separate buffer. This segmentation isolates the dependency chains so that packet loss in one buffer does not propagate to other buffers. Each buffer maintains its own compression context for its assigned packet range, preventing error propagation while maintaining high compression ratios.
Solution Approach 2:
The patent introduces buffer boundaries as intermediaries that separate the compression contexts of different packet ranges. These boundaries act as mediators that prevent error propagation from one packet range to another, allowing each buffer to independently manage its compression process without affecting other buffers.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
This application relates to a communication method and apparatus. The communication method includes: A first device configures a plurality of buffers used to compress data packets, where each of the plurality of buffers is used to compress one or more data packets. The first device compresses a first data packet by using a first buffer, where the first buffer is one of the plurality of buffers. The first device sends a compressed first data packet to a second device. According to the method provided in embodiments of this application, a quantity of lost data packets can be decreased.