Cached Adaptive Transforms for Data Stream Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data compression methods, such as JPEG and MPEG, use fixed transforms that lack adaptability to input streams, requiring side information to be sent, which negates the gains of adaptive transforms and is inefficient in terms of storage and bandwidth.
Innovation Solution
The method employs cached adaptive transforms that adapt to the data stream by using a baseline transform and caching custom orthogonal basis vectors, allowing for efficient encoding and decoding without sending the custom basis, thereby reducing bandwidth consumption and improving compression efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If fixed transforms are used for data compression, then device complexity is reduced and ease of operation is improved, but adaptability to input streams deteriorates and compression efficiency is limited
Solution Approach 1:
The system pre-computes and caches multiple custom orthogonal transforms during an initialization phase before actual data compression begins. This preliminary action allows the transforms to be ready for immediate use without requiring real-time computation, thus maintaining ease of operation while achieving adaptability through pre-adapted transforms specific to different data types and compression requirements
Solution Approach 2:
The system dynamically selects appropriate cached transforms based on the characteristics of the input data stream. Rather than using a single fixed transform, the system adapts by choosing from multiple pre-computed transforms that best match the current data properties, achieving both adaptability and operational simplicity through dynamic selection rather than dynamic computation
2Productivity
If adaptive transforms are used for data compression, then compression efficiency is improved, but device complexity increases due to the need to compute and manage custom transforms
Solution Approach 1:
Custom orthogonal transforms are pre-computed and stored in cache during an initialization phase, eliminating the need for complex real-time transform computation during data compression. This preliminary action reduces device complexity during operation while maintaining high compression efficiency through the use of optimized pre-adapted transforms
Solution Approach 2:
Instead of computing unique custom transforms for each data stream, the system creates copies of pre-computed transforms from a cached library. These copied transforms can be reused across multiple data streams with similar characteristics, reducing computational complexity while maintaining compression efficiency through effective transform selection
3Measurement precision
If custom orthogonal basis vectors are sent to decoder, then decoding accuracy is improved, but bandwidth consumption increases
Solution Approach 1:
The system extracts and transmits only the essential index information that identifies which cached transform should be used at the decoder, rather than sending the complete custom orthogonal basis vectors. This extraction approach maintains decoding accuracy by ensuring the correct transform is selected while dramatically reducing bandwidth consumption by transmitting minimal identifying information instead of full transform data
Solution Approach 2:
The cached transform library acts as an intermediary between the encoder and decoder. The encoder selects and references transforms from this cache, transmitting only lightweight references (indices) to the decoder. The decoder then uses these references to access the corresponding transforms from its own cached copy, ensuring decoding accuracy while minimizing bandwidth usage through the intermediary cache mechanism
4Measurement precision
If transform information is sent to decoder, then decoding precision is improved, but storage requirements increase
Solution Approach 1:
The system creates and maintains cached copies of custom orthogonal transforms in both encoder and decoder memory. This copying approach allows the transmission of minimal transform identification information while ensuring both sides have access to the full-precision transforms needed for accurate encoding and decoding, thereby maintaining decoding precision while reducing storage requirements compared to transmitting complete transform data
Data Source
AI summary
A method and apparatus for cached adaptive transforms for compressing data streams, computing similarity, and recognizing patterns have been disclosed.In one embodiment of the invention an encoder and decoder begin with a baseline transform. As data is transferred an algorithm is arranged so that the encoder and decoder adapt toward a superior basis than the baseline, with a corresponding reduction in the encoding bit rate. That is the algorithm adapts to the incoming data stream and can use a custom basis. We deliberately avoid having to send the custom basis itself (when possible), because sending the basis vectors consumes precious bandwidth and may defeat the goal of compression. The encoder and decoder can bootstrap themselves into using one or more better bases.In one embodiment of the invention there is no beginning baseline transform shared between the encoder and the decoder.