Fusion representation network model training method, fingerprint representation method and equipment thereof

Abstract

The invention provides a fusion representation network model training method, a fingerprint representation method and equipment thereof, and the training method comprises the steps: carrying out the feature extraction of channel state information data through a multi-layer perceptron network, and obtaining a channel state information feature map; performing feature extraction on the image data of each orientation by using a convolutional neural network with the same weight to obtain a feature map of the image of each orientation; fusing the feature maps of the images of all orientations of the same training sample to obtain a multi-azimuth feature map; splicing the channel state information feature map and the multi-azimuth feature map to construct fusion representation; correspondingly constructing a fusion fingerprint database by using the channel state information and the fusion representation of the image, and performing parameter optimization by using a set measurement index to obtain a fusion representation network model; and setting a measurement index for measuring the distance between the feature fingerprints in the fusion fingerprint database. Through the scheme, the feature distinction degree can be improved, and the positioning accuracy is improved.

Description

technical field [0001] The invention relates to the field of positioning technology, in particular to a fusion representation network model training method, a fingerprint representation method and a device thereof. Background technique [0002] With the development of informatization and intelligence in society, information such as navigation and positioning occupy an increasing proportion in daily life, and the location-based service industry has been widely used in various fields. [0003] For complex indoor environments, a variety of positioning methods have been proposed. According to the signal source, they can be divided into positioning technologies based on wireless fidelity (Wireless-Fidelity, Wi-Fi), Bluetooth, ultra-wideband and other radio signals, and positioning technologies based on infrared, Positioning techniques for non-radio signals such as ultrasonic, visual and inertial systems. [0004] Among the many positioning sources, Wi-Fi and visual signal source...

AI Technical Summary

Problems solved by technology

However, most of the heterogeneous and heterogeneous feature fusion algorithms in the existing fusion positioning system still stay on feature combination or screening, and it is difficult to fundamentally change the problem of insufficient discrimination of fingerprint features

Images