IoT Resource Sharing via Performance-Based Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current connected objects in IoT networks are designed independently, leading to high manufacturing costs and complexity due to individual resource dimensioning, which hinders efficient resource sharing and utilization.
Innovation Solution
A method and system for resource sharing between connected objects within a local network, utilizing a management unit to control and manage resource access through a list of shared resources, selecting the most efficient object to perform tasks based on past performance indicators.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If each connected object is sized in hardware and software resources independently, then each object can perform its dedicated function reliably, but the manufacturing cost and complexity of each connected object increases
Solution Approach 1:
The patent merges resource management capabilities into a centralized management unit that coordinates resources across multiple connected objects. Instead of each object having complete resource stacks independently, the management unit consolidates resource allocation, load balancing, and coordination functions, reducing individual object complexity while maintaining system-wide reliability through centralized control and monitoring
Solution Approach 2:
The management unit serves multiple functions: resource allocation, load balancing, performance monitoring, and coordination across different connected objects. This universal component handles diverse resource management tasks for various object types (sensors, actuators, processors), reducing the need for each individual object to have specialized resource management capabilities
2Ease of manufacture
If each connected object is sized in hardware and software resources independently, then each object can operate autonomously, but the manufacturing cost increases due to over-provisioning
Solution Approach 1:
The patent implements dynamic resource allocation where the management unit continuously monitors system state and adjusts resource distribution in real-time. Resources are dynamically assigned to objects based on current workload, performance requirements, and availability, allowing the system to adapt to changing conditions without requiring each object to be over-provisioned for worst-case scenarios
Solution Approach 2:
The management unit changes resource allocation parameters dynamically based on system conditions. It adjusts the amount of computing power, memory, and other resources assigned to each object according to actual performance needs, enabling cost-effective manufacturing while maintaining adaptability through parameter optimization rather than fixed over-provisioning
3Device complexity
If resources are shared between connected objects, then manufacturing cost and complexity are reduced, but efficient resource selection and coordination becomes more difficult
Solution Approach 1:
The management unit implements feedback mechanisms that continuously monitor resource usage, object performance, and system state. This feedback information is used to make informed decisions about resource allocation and to select the most appropriate objects for specific tasks based on their current performance metrics and resource availability, simplifying the coordination process through data-driven decision-making
Data Source
Figure 1~2
Figure 3~4
AI summary
The invention relates to a computer-implemented method (100) for sharing resources between connected objects forming a local network, implementing a management unit connected to said connected objects and controlling resource sharing using a shared resource list (LRP), and comprising at least one iteration of the following steps: - emission (102) by a first connected object of a request to use a resource of another connected object for the execution of a target operation; and - selection (112), by said management unit, of a second connected object possessing said resource to perform said target operation on behalf of said first object;characterized in that, when several connected objects, called candidate objects, can perform said target operation, the second object is selected from among said candidate objects based on at least one performance indicator, previously measured for each of said candidate objects, during past performance(s) of operation(s) identical or similar to said target operation. It also concerns a system implementing such a process.