Network Anomaly Detection via Uplink Downlink Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods fail to accurately detect and effectively process network anomalies in mobile communication networks, particularly in distinguishing between uplink and downlink performance issues, leading to reduced network throughput and stability.
Innovation Solution
A method where a base station periodically collects and analyzes performance index information, compares it with preset thresholds to determine link status, and performs corresponding processes to restrict or adjust scheduling resources, thereby identifying and mitigating anomalies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the base station uses traditional anomaly detection methods, then the detection process is simple, but the network anomaly cannot be detected completely and correctly
Solution Approach 1:
The patent segments the anomaly detection process into multiple dimensions by separately analyzing uplink and downlink performance indexes. Different performance indicators (throughput, error rate, packet loss rate) are monitored independently for each direction, allowing precise identification of whether the anomaly originates from uplink or downlink, thereby significantly improving detection accuracy without excessive complexity
Solution Approach 2:
The patent introduces a new dimension of analysis by comparing performance indexes from both uplink and downlink directions simultaneously. This dimensional expansion allows the system to distinguish the direction of anomaly occurrence, transforming a single-dimensional detection approach into a multi-dimensional one, which enhances detection completeness and accuracy
2Reliability
If no targeted processing is performed on detected anomalies, then the processing system remains simple, but the network performance degradation cannot be effectively mitigated
Solution Approach 1:
The patent applies local quality by implementing direction-specific processing strategies. When an uplink anomaly is detected, only uplink resources are adjusted; when a downlink anomaly is detected, only downlink resources are adjusted. This localized processing approach maintains network reliability by targeting the specific problematic direction while avoiding unnecessary complexity in the overall processing system
Solution Approach 2:
The patent implements dynamic processing by adjusting scheduling resources based on real-time anomaly detection results. The system dynamically switches between normal scheduling mode and anomaly mitigation mode, adjusting resource allocation strategies according to the detected anomaly type and direction, thereby improving network performance stability without requiring a permanently complex processing architecture
3Reliability
If the base station adjusts scheduling resources for anomaly mitigation, then the impact on normal links is reduced, but the scheduling system becomes more complex
Solution Approach 1:
The patent segments the scheduling resource adjustment into uplink-specific and downlink-specific strategies. When an anomaly is detected in one direction, only the scheduling resources for that direction are adjusted, while the other direction continues normal scheduling. This segmentation protects normal links from unnecessary interference and reduces the complexity burden on the overall scheduling system
Solution Approach 2:
The patent applies partial action by adjusting only the necessary scheduling resources related to the detected anomaly rather than reconfiguring the entire scheduling system. This targeted adjustment mitigates the impact on normal links while keeping the scheduling system complexity manageable by avoiding comprehensive system reconfiguration
Data Source
Figure 1~2
Figure 3~5
AI summary
Disclosed are a network anomaly detection and processing method and device and a computer storage medium. The method comprises the following steps: periodically collecting statistics on performance index information of an access link; determining a state of the link according to the collected statistical performance index information and a threshold corresponding to the performance index information; and processing the link according to the determined link state.