Sensor data high-precision classification method based on phase space

Abstract

The invention discloses a sensor data high-precision classification method based on a phase space. The method comprises the following steps of: 1) embedding a univariate time sequence X collected by a sensor into an m-dimensional phase space through adopting a time delay embedding method; 2) determining optimal values of tau and m by maximizing the resolution of theta (tau, m); 3) linearly mapping all elements in the theta (tau, m) to an integer interval [0, 255] to obtain a maximum track matrix image theta max; and 4) classifying the maximum track matrix image theta max obtained by time sequence conversion through using a classifier. According to the sensor data high-precision classification method, the track matrix is directly used as an image instead of mapping a time sequence to an RPS image, so that phase space projection is not needed, and information loss caused by projection is avoided; and reconstruction parameters are determined by maximizing the resolution of the trajectory matrix image theta (tau, m), the optimization method is very simple, and the classification performance of a classifier can be optimized.

Description

technical field [0001] The technical field of sensor data processing of the present invention particularly relates to a classification method of sensor data. Background technique [0002] As sensor devices become more prevalent in our daily lives, various types of sensor data can be effectively utilized in numerous applications. Therefore, classification of sensor data has become an essential requirement in these applications. Most sensor data is time series data, which refers to the data sequence obtained in a continuous time period, so the classification problem of these sensor time series data is actually a time series classification (TSC) problem. With the rapid growth of sensor time series data, time series classification has become a fundamental task in many practical applications. [0003] For several years, TSC methods for large amounts of sensor data have emerged, and these methods can be divided into two categories: traditional methods and deep learning methods. ...

AI Technical Summary

Problems solved by technology

[0006] In view of this, the object of the present invention is a high-precision classification method for sensor data based on phase space, to solve the problem that when the method based on phase space is used to classify sensor data, the existing method may cause RPS image Technical problems such as loss of original data information and reduced classification performance

Images