Network Management Center Predicting Behavioral Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network management systems in wireless communication networks, such as Wi-Fi, face issues like continuous triggering of recommended settings leading to poor user experience, lack of clear effective solutions for recurring problems, uncertainty in network reactions, and potential conflicts across networks.
Innovation Solution
A network management center that predicts behavioral patterns in wireless communication networks by analyzing metric data vectors, comparing them to a library of known patterns, and applying recommended settings to anticipate and address future network behavior, ensuring more accurate and effective problem-solving.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If recommended settings are continuously triggered to address network problems, then problem remediation is improved, but user experience deteriorates due to continuous disturbances
Solution Approach 1:
The system performs preliminary analysis by comparing current network state against historical behavioral patterns before triggering recommended settings. This allows prediction of whether a problem is temporary or persistent, enabling action only when necessary rather than continuous triggering.
Solution Approach 2:
The system incorporates feedback mechanisms by continuously monitoring network metrics and comparing them against predicted behavioral patterns. This feedback loop allows the system to adapt triggering decisions based on actual network response, avoiding unnecessary continuous triggering that degrades user experience.
2Adaptability or versatility
If multiple recommended settings are applied for the same problem, then problem coverage is improved, but solution effectiveness deteriorates due to lack of clear preference
Solution Approach 1:
The system applies local quality by matching specific network conditions and problem types to the most appropriate recommended setting from the library. Rather than applying all possible settings, it selects the one best suited to the specific behavioral pattern identified, ensuring targeted effectiveness.
3Speed
If recommended settings are triggered without predicting future network behavior, then response speed is improved, but accuracy of problem-solving deteriorates
Solution Approach 1:
The system performs preliminary prediction of future network behavior by comparing current metrics against historical patterns before triggering recommended settings. This advance analysis ensures accurate problem identification while maintaining relatively fast response by using pre-established pattern matching rather than complex real-time simulation.
4Device complexity
If network operates in open loop without tracking network reactions, then system complexity is reduced, but operational reliability deteriorates
Solution Approach 1:
The system implements feedback by monitoring network metrics before and after triggering recommended settings, comparing actual network reactions against predicted outcomes based on behavioral patterns. This feedback mechanism improves operational reliability by verifying whether problems are actually remediated without requiring complex closed-loop control systems.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
A network management center (400) for predicting behavioural patterns in a wireless communication network (420) is discloses comprising a monitoring module (401) for obtaining a first set of metric data vectors (100) for different time periods, a predicting module (402) arranged for predicting a set of future metric data vectors (130) from the first set of metric data vectors (100), a pattern analysing module (404) to compare the set of future metric data vectors (100) with a library (403) of different behavioural patterns in the communication network (420) and to relate the set of future data vectors (130) to a respective behavioural pattern in the library (403) when the future data vectors (130) are similar to the metric data vectors of the respective behavioural pattern.