ML Proxy Offloading for IoT Data Collection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Implementing machine learning in IoT devices is hindered by the logistical and cost difficulties of collecting and processing large volumes of data, requiring significant processing power and complex software configurations.
Innovation Solution
A machine-learning proxy system that offloads ML processing from IoT devices to a dedicated proxy device, using a local area network for communication, which includes processors, memory, and software to perform ML processing, allowing IoT devices to focus on data collection and simplifying the necessary infrastructure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If machine learning processing is implemented directly on IoT devices, then processing capability is improved, but device complexity and cost increase significantly
Solution Approach 1:
The patent introduces a cloud-based machine learning processing service as an intermediary between IoT devices and data processing. Instead of implementing ML directly on resource-constrained IoT devices, the system uses a cloud service that receives data from IoT devices, performs ML processing, and returns results. This mediator approach allows IoT devices to leverage powerful ML capabilities without requiring complex local processing hardware or software.
2Measurement precision
If significant volumes of data are collected for machine learning, then learning accuracy is improved, but data collection and transmission complexity increase
Solution Approach 1:
The patent extracts the complex data collection, storage, and ML processing functions from the IoT device itself and places them in the cloud environment. The IoT device is simplified to primarily generate and transmit data, while the cloud service handles data aggregation, preprocessing, storage, and ML model training. This extraction of complex functions resolves the contradiction by maintaining high learning accuracy through comprehensive data collection while removing the burden of complex data management from IoT devices.
3Power
If numerous processors and software are deployed for data processing, then processing power is improved, but implementation cost increases
Solution Approach 1:
The patent merges multiple processing functions (data collection, preprocessing, storage, ML model training, and inference) into a unified cloud-based service. Instead of deploying separate processors and software components on each IoT device, the system consolidates these functions in a shared cloud infrastructure that serves multiple IoT devices. This merging approach provides substantial processing power for ML while reducing per-device implementation costs through resource sharing and centralized management.
Data Source
AI summary
An apparatus and method for providing ML processing for one or more ML applications operating on one or more Internet of Things (IoT) devices includes receiving a ML request from an IoT device. The ML request can be generated by a ML application operating on the IoT device and include input data collected by the first ML application. A ML model to perform ML processing of the input data included in the ML request is identified and provided to an ML core for ML processing along with the input data included in the first ML request. The ML core produces ML processing output data based on ML processing by the ML core of input data included in the ML request using the ML model. The ML processing output data can be transmitted to the IoT device.