Flow-Level Compression Selection for Wireless Data Links
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital data compression methods in wireless networks are obsolete due to changing network conditions and the need to support multiple types of data flows, leading to inefficiencies and increased transmission delays, as they assume stable conditions and are not adaptable to heterogeneous data streams.
Innovation Solution
A method that analyzes data flows to extract characteristics for selecting the most suitable compression algorithm based on both flow-specific and link-specific conditions, using a performance scoring system to dynamically choose the best algorithm for each stream, implemented at the OSI model's layer 2 to optimize compression without relying on the content of IP headers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single compression algorithm is negotiated for the entire link using PPP CCP protocol, then the negotiation is simple and establishes a baseline compression mode, but the compression efficiency deteriorates when network conditions change or heterogeneous data flows are present
Solution Approach 1:
The patent segments the single link-level compression negotiation into multiple flow-level compression decisions. Instead of negotiating one compression mode for the entire PPP link, the system classifies data into different flows (IP, non-IP, TCP, UDP, etc.) and independently selects compression algorithms for each flow based on its characteristics and current network conditions, thereby resolving the contradiction between negotiation simplicity and compression efficiency.
Solution Approach 2:
The patent introduces dynamic adaptation of compression algorithms. Rather than using a static compression mode negotiated once at link establishment, the system continuously monitors network conditions (packet loss rate, delay, bandwidth) and flow characteristics, then dynamically selects the most appropriate compression algorithm for each flow, maintaining optimal compression efficiency despite changing conditions.
2Productivity
If compression mode re-negotiation is performed to adapt to changing conditions, then compression efficiency can be improved, but transmission delays increase and data packets are lost during re-negotiation
Solution Approach 1:
The patent avoids link-level re-negotiation by implementing flow-level compression selection. When network conditions change, the system adjusts compression algorithms for individual flows without triggering a complete PPP link re-negotiation, thus maintaining compression efficiency while avoiding transmission delays and packet losses associated with link re-negotiation.
Solution Approach 2:
The patent performs preliminary classification of data flows at the beginning of transmission and establishes compression parameters for each flow class. This preliminary action allows the system to adapt to changing conditions by reclassifying or re-evaluating flows without needing to re-negotiate the entire link, thereby preventing transmission delays.
3Reliability
If robust compression algorithms like predictor are used in unreliable wireless networks, then packet loss resistance is improved, but compression ratio deteriorates compared to algorithms like deflate
Solution Approach 1:
The patent changes the selection of compression algorithm parameters based on network conditions and flow characteristics. Instead of using a fixed robust algorithm like predictor for all cases, the system evaluates current packet loss rates and selects algorithms with appropriate robustness levels. For example, it may choose deflate for flows with low packet loss and predictor for flows experiencing high packet loss, thereby optimizing both reliability and compression ratio dynamically.
Solution Approach 2:
The patent applies different compression algorithms to different data flows based on their specific characteristics and current network conditions. Rather than using a uniform robust algorithm across all traffic, the system tailors the compression approach to each flow's requirements, applying stronger error resistance only where necessary, thus maintaining high compression ratios for flows that don't require robustness while ensuring reliability for sensitive flows.
4Duration of action of stationary object
If compression algorithms process already compressed data, then data transmission continues, but data size increases and transmission delay increases due to unnecessary compression/decompression
Solution Approach 1:
The patent performs preliminary detection of data characteristics and compression status before applying compression algorithms. By classifying flows and analyzing data patterns in advance, the system can identify already-compressed data and skip unnecessary compression operations, thereby maintaining transmission continuity while avoiding the delay and size increase caused by redundant compression/decompression cycles.
Data Source
Figure 1~3
Figure 2
AI summary
The method involves classifying digital data to be transmitted into a set of distinct flows, and analyzing each flow to determine characteristic quantities. Transmission conditions on a physical link are analyzed to determine another set of characteristic quantities. The two characteristic quantities are combined through a determined performance to calculate the performance rating for each of compression algorithm implemented by a mobile terminal. The compression algorithm with better performance rating is selected for each flow.