Cognitive Network Management via AI Golden Records
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional approaches to managing modern wireless communication networks are slow, inconsistent, and insufficient in responding to dynamic changes in network characteristics due to their complexity and the need for manual updates across multiple vendors and operators in disaggregated, cloud-native architectures.
Innovation Solution
The implementation of cognitive-defined network management (CDNM) using a multi-modal AI engine to normalize raw network data into 'golden record' data for real-time analysis and orchestration, enabling automated and dynamic management of network functions across a disaggregated wireless network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional manual approaches are used to manage disaggregated wireless networks, then human administrators can analyze and orchestrate network functions, but the management becomes slow, inconsistent, and unable to respond to dynamic changes
Solution Approach 1:
The system implements self-service through automated network management where the network infrastructure autonomously collects, normalizes, and analyzes its own operational data using AI/ML models. The orchestrator automatically executes workflows and adjusts network functions without human intervention, enabling the system to self-manage dynamic changes in real-time
Solution Approach 2:
The patent replaces manual mechanical administration with automated digital systems. Human administrators no longer manually collect and analyze data; instead, automated agents continuously gather operational data, AI/ML models process and normalize this data, and orchestration systems automatically execute management workflows, substituting human mechanical operations with automated computational processes
2Reliability
If human administrators manually collect and analyze network data, then network performance can be monitored, but the process becomes error-prone and insufficient to keep up with rapid changes
Solution Approach 1:
The system implements continuous monitoring and analysis through automated agents that continuously collect operational data from network elements. The AI/ML models continuously process this data stream in real-time, and the orchestrator continuously executes workflows, ensuring uninterrupted and consistent network management actions without the gaps inherent in manual periodic updates
Solution Approach 2:
The system implements feedback loops where operational data is continuously collected from the network, analyzed by AI/ML models to detect changes and anomalies, and used to automatically adjust network functions. The orchestrator monitors execution results and feeds this information back into the system for continuous optimization, creating a closed-loop control mechanism that improves reliability and reduces response time
3Adaptability or versatility
If disaggregated architecture with multiple vendors is implemented, then network flexibility and efficiency improve, but data heterogeneity and interoperability complexities increase
Solution Approach 1:
The system implements universality through a vendor-agnostic orchestrator and standardized data models that can work with network elements from multiple vendors. The AI/ML models are trained on normalized data from diverse sources and can process heterogeneous data formats uniformly, enabling the system to manage multi-vendor disaggregated networks with a single unified platform that adapts to different vendor-specific implementations
4Productivity
If real-time data normalization is implemented, then dynamic orchestration capability improves, but computational processing requirements increase
Solution Approach 1:
The system implements preliminary action by pre-training AI/ML models offline using historical operational data from multiple vendors. This pre-processing creates ready-to-use models that can quickly normalize and analyze real-time data with minimal computational overhead during actual network operations, reducing the energy required for real-time processing while maintaining high orchestration capability
Data Source
AI summary
Techniques are described for cognitive defined network management (CDNM) that seek to perform real-time collection and analysis of raw network data from across a disaggregated wireless network and to dynamically orchestrate network management functions substantially in real time, accordingly. For example, a multi-modal artificial intelligence (AI) engine is trained to normalize the heterogeneous raw network data into homogeneous so-called “golden record data.” A repository of historical golden records can be maintained for generating data models for use in training AI network management applications. An orchestrator can operate to directing execution of pre-developed network management workflows based on results obtained from querying the trained AI network management applications with newly received (real-time) golden records.