Test Vector Compression Using Column-Wise Sliding Windows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current lossless compression algorithms, such as the LZ77 series, are not convenient for implementation on FPGA and ASIC in the field of ATE testing, as they require a dictionary and are not suitable for real-time compression without one.
Innovation Solution
A lossless compression method for test vectors that converts data into a binary data stream with rows and columns, compressing each column independently using a sliding window approach without a dictionary, allowing for parallel processing on FPGA and efficient compression and decompression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of substance
If LZ77 series lossless compression algorithm is used, then compression ratio is improved, but device complexity increases due to dictionary requirement
Solution Approach 1:
The patent extracts and removes the dictionary component from the compression system. Instead of using a dictionary-based approach like LZ77, the invention uses a sliding window that operates directly on the incoming data stream, eliminating the need for pre-stored dictionary structures while maintaining compression functionality.
Solution Approach 2:
The patent segments the compression process into fixed-size windows that slide through the data stream. Each window processes a specific segment of data independently, allowing for parallel processing on FPGA/ASIC without requiring complex global dictionary structures.
2Loss of substance
If dictionary-based compression is used, then compression effect is improved, but ease of operation deteriorates for real-time processing
Solution Approach 1:
The patent prepares the sliding window structure in advance with predetermined size parameters before processing begins. This preliminary configuration allows the system to immediately start compressing incoming data streams without dynamic dictionary building delays, enabling real-time operation.
Solution Approach 2:
The sliding window dynamically moves through the data stream, continuously updating its position and contents as new data arrives. This dynamic behavior allows the system to adapt to incoming data in real-time without requiring static dictionary structures, facilitating real-time compression processing.
3Loss of substance
If LZ77 algorithm is implemented, then compression ratio is improved, but productivity decreases for FPGA/ASIC implementation
Solution Approach 1:
By segmenting the data stream into fixed-size windows that can be processed independently and in parallel, the patent enables efficient utilization of FPGA/ASIC parallel processing capabilities. Each window can be compressed simultaneously without waiting for dictionary operations, significantly improving processing throughput.
Solution Approach 2:
The patent replaces the mechanical dictionary lookup operations with a more efficient sliding window mechanism that uses simple bit-shift and comparison operations. This substitution eliminates complex memory access patterns associated with dictionaries and enables faster, more hardware-friendly processing suitable for high-speed FPGA/ASIC implementation.
Data Source
AI summary
A lossless compression method for a test vector has following steps: S01: converting the test vector into a data stream with A rows and B columns, the data stream is expressed in binary; S02: compressing the data stream a column by a column sequentially to form compressed words and uncompressed words corresponding to test datum of each of the B columns; a compression method for the test datum of each of the B columns comprises: setting a window with a width of 1 bit and a depth of M rows, and sliding the window down a row by a row from top of test datum of a column, and forming the compressed words and the uncompressed words corresponding to the test datum of each of the B columns; S03: converging the compressed datum of the test datum of each of the B columns to form a compressed data stream.