Optimal label selection method of RFID equipment-free human body tracking system

Abstract

The invention discloses an optimal label selection method for an RFID equipment-free human body tracking system, which comprises the following steps: S1, region division: dividing a monitoring region into N positions; s2, feature extraction and calculation: extracting a mean value and a variance of T RSSIs sampled at each position within a period of time, and establishing a mapping relationship between the distribution of the RSSIs in the period of time and the positions of the RSSIs; s3, constructing a deep learning model, and analyzing a corresponding position sequence according to the RSSI sequence of the T, namely a real movement track of a human body; and S4, label layout mode selection: preferentially selecting labels according to the classification accuracy of the deep learning model for the positions. Through the deep learning model, while the human body tracking precision is maintained, the number of labels is reduced, the flexibility of the model is improved, precision reduction is avoided, the performance of processing a long-path sequence is improved, and the gap between training and inference of a position sequence prediction task is filled up.

Description

technical field [0001] The invention relates to the technical field of personnel positioning, in particular to a preferred labeling method of an RFID device-free human body tracking system. Background technique [0002] With the increasing aging of modern society, the guardianship of the elderly is a major issue that has to be faced today. How to enable modern smart devices to better support the independent living of the elderly has received more and more attention from the industry and academia. The core of this problem is how to accurately locate the human body in a complex living environment. Although traditional methods based on Compared with non-equipment technology, the technology of equipment is less affected by the environment, and it will be easier to implement, but to use these systems, users must be required to carry corresponding sensors or communication equipment, and for the elderly, equipment forgetting and The problem of loss, and the user also needs to cons...

AI Technical Summary

Problems solved by technology

[0005] Difficulties to solve: 1) Due to the influence of the environment, the power signal has a large range of changes and is unstable. If it is directly used to train the deep learning model, the tracking effect will be poor; 2) At present, there are many types of deep learning models, and the effects of different structures and training methods Different, how to choose the two according to the current task is a difficult point; 3) The existing research does not consider the problem of tag usage, most of which are because the system itself is limited, and reducing tags will cause system tracking or positioning errors to increase, and will also cause reading How to use a relatively small number of tags is the difficulty of this research

The traditional tracking system is usually a whole with a high degree of coupling. It is a very difficult problem to control the number of tags, which is one of the reasons why this problem is ignored.

Images