Interval-Based Table Lookup Circuit for Accuracy-Storage Tradeoffs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing table lookup systems face challenges in achieving high fitting accuracy while minimizing storage space, particularly when the number of entries in the lookup table is insufficient or excessive, leading to reduced accuracy or excessive storage requirements.
Innovation Solution
The system divides the input range of operators into multiple intervals, employing different table lookup processing modes based on the characteristics of each interval, such as linear or non-linear intervals, using a two-step lookup process for non-linear intervals to improve accuracy while reducing storage space.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the quantity of entries in the lookup table is increased, then fitting accuracy is improved, but storage space occupied increases
Solution Approach 1:
The patent divides the input range of the operator into multiple intervals, with each interval having its own lookup table. This segmentation allows each lookup table to cover a smaller range with fewer entries, reducing individual storage requirements while maintaining overall accuracy across the full input range through interval-specific optimization.
Solution Approach 2:
Different lookup tables are created for different intervals based on the local characteristics of the operator in each interval. This allows each lookup table to be optimized for its specific interval, using appropriate entry quantities and distributions that match the local behavior, thereby improving fitting accuracy without uniformly increasing storage across all intervals.
2Volume of stationary object
If the quantity of entries in the lookup table is decreased, then storage space is reduced, but fitting accuracy deteriorates
Solution Approach 1:
By segmenting the input range into intervals, the patent enables each lookup table to use fewer entries while collectively covering the full range with sufficient precision. The segmentation transforms a single large-accuracy-requiring table into multiple smaller tables that together achieve the same or better accuracy with reduced total storage.
Solution Approach 2:
The patent applies different numbers of entries in different intervals based on local requirements. Some intervals may use more entries where the operator varies rapidly, while other intervals use fewer entries where the operator is more stable, avoiding the excessive storage that would result from uniformly high entry counts across all intervals.
Data Source
AI summary
Embodiments of this disclosure disclose a table lookup processing system and method, a medium, a device, and a chip. The system includes: a memory; a processor, coupled to the memory; and a table lookup processing circuit coupled to the memory and the processor. The table lookup processing circuit is configured to: determine table lookup input data based on to-be-processed data from the processor or to-be-processed data from the memory; determine a target input interval to which the table lookup input data belongs from at least one preset input interval; perform table lookup processing on the table lookup input data based on a target interval type corresponding to the target input interval, to obtain a table lookup result corresponding to the table lookup input data; determine a processing result corresponding to the to-be-processed data based on the table lookup result; and write the processing result into the memory.