Differential private multi-source wireless signal fingerprint fusion indoor positioning method under edge computing architecture

Abstract

The invention discloses a differential private multi-source wireless signal fingerprint fusion indoor positioning method under an edge computing architecture. The method comprises the steps that (1),edge equipment adds Laplace noise to own information strength RSS data and then randomly sends the RSS data to nearby edge nodes; (2), after receiving the RSS data, the edge node aggregates the RSS data of WiFi and Bluetooth collected at the same position, uniformly calibrates the aggregated RSS data and sends the calibrated RSS data to an edge server; (3), the edge server integrates received noise marks and unmarked samples, performs feature fusion of differential privacy protection on RSS data of WiFi and BLE by utilizing graph Laplace manifold constraints, and sends all data sets subjectedto privacy protection processing to a cloud server; and (4), learning parameters are learned by the cloud server, machine learning model training satisfying differential privatives is carried out, and a safe and credible indoor positioning model is generated. According to the invention, proven privacy protection can be provided, and high positioning precision and low resource consumption can be ensured.

Description

technical field [0001] The present invention relates to the field of indoor positioning services. In order to obtain better positioning services, users voluntarily provide their own or collected data to participate in the training of positioning models, and after appropriate differential private disturbance layer by layer, a safe and reliable model is generated on the cloud server. Indoor positioning model, so as to protect the user's location privacy. Background technique [0002] In the traditional cloud-centric computing method, all the data collected by mobile devices will be uploaded and stored on the cloud server for centralized computing and processing. However, with the rapid development of technologies and fields such as the Internet of Things, crowd sensing, and social networks. Ubiquitous mobile devices and sensors continue to generate massive amounts of data, and hundreds of millions of users generate huge amounts of interaction when enjoying Internet services, ...

AI Technical Summary

Problems solved by technology

The index set includes: several hash tables, the parameters of the corresponding function group of each hash table, and the position coordinates of each fingerprint marked with the fingerprint serial number, but the disadvantage of this invention is that the attacker can use violence to The attack method obtains the approximate information of the database, that is, the attacker forges as many WiFi fingerprints as possible within the legal range, and uses this scheme to obtain the positioning results of all WiFi fingerprints. When the positions of the reference points in the index set are the same, these positioning results and the corresponding The database composed of pseudo WiFi fingerprints is very similar to the original database

Images