Internet of Things node selection method and system for privacy protection, equipment and terminal

Abstract

The invention belongs to the technical field of Internet of Things, and discloses an Internet of Things node selection method and system for privacy protection, equipment and a terminal, the Internet of Things node selection method comprises the steps: modeling a selection problem of an unknown node as a multi-arm gambling machine problem of a multi-node game, and modeling node capability attributes as capability measurement parameters; carrying out a training process on capability measurement parameters based on a Thompson sampling algorithm to realize iterative selection of a node group; in the iteration process, updating the node capability parameters, wherein the nodes use a random response mechanism to disturb feedback values of the capability parameters of the nodes, so that the parameter updating process meets the localized differential privacy constraint. According to the method, the optimal node group can be selected through multiple iterations without context background, and the capability sensitive attributes of the nodes are not leaked. Experimental results show that compared with the prior art, the differentiated node selection method has high efficiency and is suitable for diversified perception tasks.

Description

technical field [0001] The invention belongs to the technical field of the Internet of Things, and in particular relates to a privacy-protected Internet of Things node optimization method, system, device and terminal. Background technique [0002] Currently, with the rapid development of information technology, the Internet of Things (IoT) paradigm has become one of the main attractions for researchers and enterprises. Many see this as the next major technological revolution. IoT is based on the ability of different objects to interact and collaborate to achieve a common goal with minimal human interaction. Such objects include sensors, RFID, actuators and mobile phones. In sensing applications, an important enabler of the IoT paradigm is the Wireless Sensor Network (WSN). A WSN is a network composed of a large number of small, cheap, and battery-powered devices (i.e., nodes equipped with sensors). Typically, these nodes process, store and communicate data, however, IoT ...

AI Technical Summary

Problems solved by technology

[0008] (3) For the problem of unknown node selection, existing schemes use reinforcement learning methods to select multiple nodes to perform tasks together in each round. The condition or threshold for the end of the iteration

[0009] The difficulty in solving the above problems and defects is as follows: the traditional IoT node selection mechanism mainly screens through the information of IoT nodes, which not only exposes private information such as sensitive attributes of node capabilities, but also does not conform to the actual scenario, that is, the platform should not own IoT nodes. any prior knowledge of the node

Images