IoT Task Allocation via LSTM Space Complexity Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
IoT devices face inefficiencies in processing large data sizes due to limited processing power, leading to prolonged processing times, and existing systems lack an effective strategy to optimize task distribution between IoT devices and edge devices.
Innovation Solution
A computer-implemented method that utilizes a long short-term memory (LSTM) neural network to predict space complexity data size, allowing for the allocation of task processing between IoT devices and edge devices, thereby minimizing overall processing time by determining the optimal offload value K to distribute data processing resources effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If IoT devices process large data sizes locally, then processing capability is maintained, but processing time increases and device performance degrades
Solution Approach 1:
The patent segments the data processing task into two parts: local processing at the IoT device and remote processing at the edge device. The system dynamically determines the optimal split ratio based on predicted space complexity, allowing critical functions to remain local while offloading computationally intensive tasks to reduce processing time.
Solution Approach 2:
The patent introduces an edge device as an intermediary between the IoT device and the cloud. This intermediary handles complex data processing tasks that are too resource-intensive for the IoT device but require more processing power than the device can provide, thereby reducing processing time without sacrificing capability.
2Loss of time
If tasks are offloaded to edge devices, then processing time is reduced, but system complexity increases
Solution Approach 1:
The patent uses parameter changes to simplify the offloading decision process. By predicting space complexity as a key parameter and using it to determine the offloading ratio, the system transforms a complex multi-factor decision into a more manageable parameter-based decision, reducing system complexity while maintaining optimal processing time.
Solution Approach 2:
The system implements feedback mechanisms where the predicted space complexity and actual processing outcomes are used to adjust future offloading decisions. This feedback loop allows the system to learn and optimize its behavior, reducing the need for complex manual configuration and simplifying the overall system management.
3Productivity
If more data is processed at edge devices, then overall processing efficiency improves, but communication overhead increases
Solution Approach 1:
The patent applies partial offloading rather than complete offloading. By processing some data locally and only offloading the necessary portion based on predicted space complexity, the system achieves improved processing efficiency without the full communication overhead that would result from offloading all data to edge devices.
Data Source
AI summary
Methods and systems are disclosed for allocating tasks between apparatus in an IoT system in a manner to generally minimize the total amount of time to execute the tasks. At least one embodiment includes a computer-implemented method for allocating task processing between an internet of things (IoT) device and an edge device. The computer-implemented method includes collecting data from one or more sensors to execute a task having data size Xt; predicting a space complexity data size Xc for the task based on data size Xt, and allocating data for processing between the IoT device and edge device as a function of Xc. In at least one embodiment, the space complexity data size Xc is determined by applying Xt to the input of a long short-term memory neural network.