Delegating Cloud Roles to IoT Devices via Trajectory Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
IoT devices often experience service disruptions due to unstable or unavailable network connections, preventing them from performing cloud-dependent functions and data processing tasks.
Innovation Solution
A method that predicts network disconnection based on trajectory paths and environmental factors, allowing software modules and data to be delegated to IoT devices, enabling them to continue processing insights and maintaining services even when disconnected from the cloud.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all cloud node functions are performed on IoT devices to maintain service during network unavailability, then service continuity is improved, but device complexity and resource requirements increase
Solution Approach 1:
The system performs preliminary actions by predicting network unavailability in advance and proactively synchronizing required software modules and data to the IoT device before the network actually fails. This allows the device to maintain service continuity without needing to handle all cloud functions, as only essential modules are pre-loaded based on predicted needs rather than all possible functions being replicated on the device.
Solution Approach 2:
Instead of uniformly replicating all cloud node functions on every IoT device, the system applies local quality by selectively synchronizing only the specific software modules and data required for each device's current operational context and predicted trajectory. This differentiated approach reduces unnecessary complexity on devices that don't require all cloud functions.
2Reliability
If network quality information is continuously monitored and trajectory prediction is performed to predict network unavailability, then service disruption is reduced, but computational resources and processing time increase
Solution Approach 1:
The system performs trajectory prediction and network quality assessment in advance, during periods when network conditions are stable and computational resources are abundant. By predicting future network unavailability before it occurs and preparing the necessary software modules ahead of time, the system avoids the need for real-time computational intensive analysis during critical service periods, thus reducing overall processing time and resource consumption.
3Reliability
If software modules and data are synchronized to devices before network unavailability, then service continuity is improved, but network bandwidth and communication overhead increase
Solution Approach 1:
The system synchronizes only the specific software modules and data required for each device's predicted operational needs, rather than transferring all possible cloud resources. By tailoring the synchronization content to each device's specific trajectory and requirements, the system minimizes unnecessary network bandwidth consumption while ensuring service continuity for the actual functions needed.
Solution Approach 2:
The synchronization is performed in advance during periods of stable network connectivity, allowing batch transfers of required modules before network unavailability occurs. This proactive approach consolidates network usage into controlled windows rather than requiring continuous or reactive data transfer, optimizing overall network bandwidth efficiency.
Data Source
AI summary
Provided are techniques for delegating cloud-side roles to devices. Network quality information is received for a network from one or more devices. It is determined that the network is unavailable in an area from which the network quality information was received. For a new device, a trajectory path of the new device is predicted. It is determined that the new device is about to enter the area based on the trajectory path. One or more software modules and data are identified. The one or more software modules and the data are sent to the new device, wherein the new device executes the one or more software modules to generate one or more insights from the data.