Dynamic Edge Device Operation via Network Slicing and Analytics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional edge computing systems are not adaptive to dynamic changes in communication techniques like IoT and 5G due to their static nature, leading to inefficient network bandwidth allocation and lack of effective network slicing mechanisms.
Innovation Solution
A system that dynamically assigns operations to devices in a communication network by clustering devices based on behavioral and contextual attributes, selecting appropriate network slices, and configuring analytics models using neural-based deep learning for intelligent edge device operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If edge computing systems use pre-configured data analytics models and static algorithms, then device operation is simplified and easier to manage, but the system cannot adapt to dynamic changes in communication techniques like IoT and 5G
Solution Approach 1:
The patent implements dynamic operation of edge devices by transitioning from static pre-configured models to dynamically selected analytics models and algorithms. The system continuously monitors contextual attributes and behavioral patterns, then adapts device operations in real-time based on changing network conditions and device requirements, enabling responsiveness to IoT and 5G dynamics.
Solution Approach 2:
The system changes operational parameters by dynamically selecting different analytics models, algorithms, and network slices based on contextual attributes. Instead of fixed configurations, the system adjusts model selection, algorithm parameters, and network resource allocation according to real-time device behavior and network conditions, achieving adaptability through parameter variability.
2Productivity
If edge computing systems use static behavior and pre-configured models, then system management is easier, but network bandwidth allocation becomes inefficient
Solution Approach 1:
The patent implements feedback mechanisms by continuously monitoring contextual attributes and behavioral patterns of edge devices. The system uses this feedback to dynamically adjust analytics model selection, algorithm configuration, and network slice allocation, optimizing bandwidth utilization based on actual device performance and network conditions rather than static pre-configuration.
Solution Approach 2:
The system transitions from static bandwidth allocation to dynamic resource management by continuously adapting network slice selection and analytics model deployment based on real-time device behavior and network conditions. This dynamic approach optimizes bandwidth utilization for diverse IoT and 5G applications while maintaining automated management through AI-driven decision-making.
3Adaptability or versatility
If edge devices have pre-configured analytics models for specific operations, then device operation is straightforward, but the system lacks effective network slicing mechanisms for dynamic requirements
Solution Approach 1:
The patent applies segmentation by dividing the network into multiple virtual network slices, each optimized for specific types of traffic and device requirements. The system dynamically assigns edge devices to appropriate network slices based on their behavioral attributes and contextual attributes, enabling effective network slicing for diverse IoT and 5G applications while maintaining manageable complexity through automated assignment.
Data Source
AI summary
The present disclosure relates to a system (102) for assigning dynamic operation of devices in a communication network (106). The system (102) receives one or more behavioral attributes and one or more contextual attributes associated with one or more devices (232) in a communication network (106). The system (102) further determines one or more clusters (234) associated with each device from the one or more devices (232). The system (102) further determines, dynamically, one or more network slices, from a set of network slices associated with the one or more clusters (234). The system (102) further determines, dynamically, one or more analytics models associated with the one or more clusters (234). The system (102) further assigns dynamic operation of the one or more clusters (234) based on the one or more contextual attributes, the one or more network slices and the one or more analytics models.