State Machine Engine Instruction Insertion for Parallel Pattern Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional computing systems face inefficiencies in pattern recognition tasks due to the increasing volume of data and number of patterns, leading to bottlenecks in processing and delays in data stream analysis, as they lack a hierarchical processing structure similar to the biological brain's neuron layers.
Innovation Solution
A state machine engine employing finite state machine lattices arranged in a hierarchical structure, allowing multiple FSMs to analyze data in parallel, with outputs from lower levels serving as inputs to higher levels, enabling rapid analysis of complex patterns across high-speed data streams.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional von Neumann based computers are used for pattern recognition, then the system structure is simple, but the processing efficiency is low and computing bottlenecks occur
Solution Approach 1:
The system segments pattern recognition into multiple hierarchical levels, with lower levels analyzing raw signals and higher levels analyzing outputs from lower levels. This segmentation allows parallel processing of different pattern types simultaneously, improving efficiency without requiring a complete redesign of the computing architecture.
Solution Approach 2:
The patent introduces a hierarchical dimension to pattern recognition processing, organizing neurons into multiple levels where each level processes patterns at increasing complexity. This dimensional organization enables simultaneous processing of multiple patterns at different hierarchical levels, resolving the bottleneck in conventional sequential processing.
2Adaptability or versatility
If the number of patterns to search for increases, then pattern recognition capability improves, but the delay before the system is ready to search the next portion of data stream increases
Solution Approach 1:
The system divides the data stream processing into multiple parallel channels, each dedicated to searching for specific patterns. This segmentation allows the system to maintain high adaptability by configuring different channels for different patterns while avoiding sequential processing delays, as all channels operate simultaneously on the same data stream.
Solution Approach 2:
The system performs preliminary configuration of multiple pattern search channels in advance, so that when data arrives, all pattern recognition tasks are already prepared and can execute in parallel. This preliminary setup eliminates the need to reconfigure the system for each new pattern search, reducing time loss while maintaining versatility.
3Productivity
If hardware is designed to search data stream for patterns by distributing data among multiple circuits, then processing capacity increases, but the system cannot process adequate amounts of data in the given time
Solution Approach 1:
The patent introduces a hierarchical temporal dimension to parallel circuit processing, where lower levels process raw data simultaneously and pass results to higher levels. This hierarchical organization allows the system to process adequate amounts of data at high speed by structuring parallel circuits in levels rather than simple parallel arrays, enabling efficient data flow through multiple processing stages.
Solution Approach 2:
The system performs preliminary processing at lower hierarchical levels before data reaches higher levels. This preliminary action filters and prepares data in advance, reducing the processing burden on subsequent circuits and enabling the entire system to handle adequate data volumes at high speed without bottlenecks.
Data Source
Figure 1~7
Figure 2
Figure 3
AI summary
State machine engines are disclosed, including those having an instruction insertion register. One such instruction insertion register may provide an initialization instruction, such as to prepare a state machine engine for data analysis. An instruction insertion register may also provide an instruction in an attempt to resolve an error that occurs during operation of a state machine engine. An instruction insertion register may also be used to debug a state machine engine, such as after the state machine experiences a fatal error.