E2E DNN Reconfiguration for Edge Mobility and Cloud Handover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of managing edge computing resources is increased by device mobility, as UEs move between coverage areas with and without edge compute servers, requiring dynamic redirection of application processing between edge and cloud-based services, which complicates network management.
Innovation Solution
Adapting end-to-end deep neural networks (DNNs) to dynamically reconfigure based on changes in edge compute server participation modes, optimizing processing for edge or cloud-based computing environments by modifying parameter and architecture configurations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the wireless network dynamically redirects application processing between edge compute servers and cloud-based services to accommodate device mobility, then processing flexibility is improved, but network management complexity increases
Solution Approach 1:
The system dynamically adapts the E2E DNN configuration based on real-time participation mode changes of edge compute servers. The network entity monitors ECS availability and automatically reconfigures the neural network architecture and parameters to optimize processing for either edge or cloud-based computing environments, enabling flexible redirection without manual intervention
Solution Approach 2:
The patent changes key parameters of the E2E DNN including architecture configuration, model weights, and processing parameters based on the participation mode of edge compute servers. By pre-configuring multiple E2E ML configurations corresponding to different ECS participation modes, the system可以快速 switch between edge and cloud processing without complex real-time optimization
2Loss of time
If edge computing resources are utilized to reduce data transfer latency, then network response time is improved, but system complexity increases due to dynamic redirection requirements
Solution Approach 1:
The system pre-configures multiple E2E ML configurations corresponding to different ECS participation modes before runtime. When an ECS becomes unavailable, the network entity can immediately switch to a pre-prepared cloud-based configuration rather than performing complex real-time reconfiguration, reducing latency while managing complexity
Solution Approach 2:
The patent creates and maintains multiple copies of the E2E DNN configuration optimized for different operating modes (edge computing mode and cloud-based mode). These configuration copies are stored and can be rapidly switched between, avoiding the need for complex real-time transformation and reducing both latency and management complexity
Data Source
AI summary
Techniques and apparatuses are described for adapting an end-to-end, E2E, machine-learning, ML, configuration for processing communications transferred through an E2E communication. A network entity directs a user equipment (UE) and a base station participating in the E2E communication to implement the E2E communication by forming at least a portion of an E2E deep neural network, DNN, based on a first E2E ML configuration. The network entity determines to update the first E2E ML configuration based on a change in a participation mode of an edge compute server (ECS) in the E2E communication. The network entity identifies a second E2E ML configuration based on the change in participation mode and directs the UE or the base station to update the portion of the E2E DNN using the second E2E ML configuration.