Network Device Control Using Pre-learned Parameters for Fast Startup
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network devices face challenges in quickly starting up due to the need for adaptive learning of running parameters based on current working conditions, which slows down data transmission.
Innovation Solution
A method and apparatus for a network device control system that collects and analyzes working data to determine trends, allowing for the quick startup of network devices by using pre-learned device running parameters associated with different working conditions, eliminating the need for adaptive learning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the network device performs adaptive learning on running parameters based on current working conditions, then data transmission quality is ensured, but startup time increases to hundreds of milliseconds
Solution Approach 1:
The patent pre-collects working data including measurement values, performance values, and device running parameters under various working conditions before actual operation. When the network device needs to start, it directly queries the pre-collected data based on current working conditions and determines appropriate running parameters without performing adaptive learning, thus achieving quick startup while ensuring transmission quality
Solution Approach 2:
The patent creates a database of pre-collected working data that replicates the relationship between working conditions, performance values, and optimal running parameters. Instead of performing actual adaptive learning during startup, the system copies the appropriate running parameters from the pre-collected database based on matching working conditions, eliminating the time-consuming learning process
2Productivity
If the network device performs adaptive learning to determine running parameters, then optimal performance is achieved, but the learning process takes hundreds of milliseconds
Solution Approach 1:
The system performs the learning process in advance by collecting working data under various conditions before actual operation. The pre-collected data includes the relationship between working conditions and optimal running parameters. During actual operation, the system directly queries this pre-learned data rather than performing learning, eliminating the hundreds of milliseconds learning delay while maintaining optimal performance
Solution Approach 2:
The patent implements a dynamic query mechanism that matches current working conditions (measurement values) with pre-collected data to determine appropriate running parameters. The system can dynamically select from multiple pre-collected parameter sets based on the current working state, achieving both speed and optimality without the learning delay
Data Source
AI summary
A network device control method and a control apparatus. The control apparatus may collect first working data, where the first working data includes a first measurement value, a first performance value, and a first device running parameter. The control apparatus determines whether the collected first working data meets a trend of a first set, where the trend of the first set is that an increasing/decreasing trend of measurement values in a plurality of groups of working data in the first set is consistent with or contrary to an increasing/decreasing trend of performance values in the plurality of groups of working data.