Extremum Search Instruction Merging Value and Location Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining minimum or maximum values in multi-dimensional arrays are resource-intensive, requiring additional storage and multiple steps that incur significant computational overhead and latency.
Innovation Solution
A method that maintains a reference extreme value and its location identifier, updating them during the search process to simultaneously determine the global extremum and its location without additional searches through the array.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the first approach of calculating extreme values row by row is used, then the extreme value can be determined systematically, but additional temporary storage space is required and the operation incurs considerable computational overhead and latency
Solution Approach 1:
The patent combines the extreme value calculation and location tracking into a single integrated process. Instead of separately calculating extreme values for each row and then searching for the global extremum, the invention maintains location identifiers throughout the calculation process, merging what were previously separate operations into one unified procedure that eliminates the need for additional storage and repeated searches.
Solution Approach 2:
The patent performs preliminary tracking of location identifiers during the extreme value calculation process. By maintaining and updating location information as extreme values are calculated row by row, the system prepares the necessary location data in advance, eliminating the need for a subsequent search through the array to find where the global extremum occurs.
2Measurement precision
If the global extremum is determined first and then searched for in the two dimensional array, then the extreme value is identified, but the search requires reading each row and comparing data elements which increases latency
Solution Approach 1:
The patent performs preliminary tracking of location identifiers during the extreme value calculation process. By maintaining and updating location information as extreme values are calculated row by row, the system prepares the necessary location data in advance, eliminating the need for a subsequent search through the array to find where the global extremum occurs.
Solution Approach 2:
The patent maintains continuous tracking of location identifiers throughout the entire extreme value calculation process. Rather than completing the calculation and then performing a separate search operation, the location information is continuously updated and maintained alongside the extreme value calculations, ensuring that location data is always current without requiring additional time for a separate search phase.
3Measurement precision
If multiple stages are used to determine extreme values and their locations, then accuracy is maintained, but the number of instructions and computational resources increase significantly
Solution Approach 1:
The patent combines the extreme value calculation and location tracking into a single integrated process. Instead of separately calculating extreme values for each row and then searching for the global extremum, the invention maintains location identifiers throughout the calculation process, merging what were previously separate operations into one unified procedure that eliminates the need for additional storage and repeated searches.
Solution Approach 2:
The patent creates a multi-functional process that simultaneously calculates extreme values and tracks their locations. The same computational loop that determines extreme values also updates location identifiers, making the process universal rather than requiring separate specialized operations for value calculation and location finding.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An apparatus, system and method of determining an extremum are disclosed. A reference location identifier and a reference extremum are coupled. An input extremum of an input data set is determined and a corresponding location identifier of the input extremum is also determined. The input extremum is compared with the reference extremum to determine an output extremum and output location identifier, based on the comparison.