Step counting method based on autocorrelation operation

Abstract

The invention provides a step counting method based on autocorrecting computing which comprises the following steps: collecting N three-axis acceleration data in a current time window by a three-axis acceleration sensor of a pedometer, calculating the module values of the N three-axis acceleration data and storing the calculated module values; collecting and calculating the module values of N three-axis acceleration data in a next time window, and taking the first L module values to be put behind the N module values to obtain N+L merged data, denoting the N+L merged data as sample data x (n); performing filtering and autocorrecting computing on the sample data x (n) to detect the number of wave crests of the sample data x (n); taking the number of wave crests as the step number detected in the current time window, and adding the step number and the current total step number together to obtain new current total step number; taking the next time window of the current time window as a new current time window and repeating the steps until the movement course is finished. According to the step counting method based on autocorrecting computing provided by the invention, through autocorrecting computing, the periodic features of periodic signals are highlighted in signals containing a great deal of noise, and the effective step number is detected according to the counted number of crest values of autocorrelation function, and the step counting accuracy is effectively improved.

Description

technical field [0001] The invention relates to the technical field of consumer application electronics, in particular to a step counting method capable of counting the steps of a user. Background technique [0002] With the rapid development of wearable electronic devices, pedometers have been widely used. Pedometers are a daily exercise progress monitor that can count the number of steps people walk, estimate the distance people walk, and calculate the energy consumed by exercise. Calories, so that people can monitor their fitness intensity, exercise level and metabolism at any time. [0003] There are many types of sensors used in pedometers. For example, the Chinese patent document "Pedometer Using Magnetic Sensor and Measurement Method" (CN101086450A) discloses a pedometer and measurement method using a magnetic sensor. In this method, a bar magnet is attached to one foot, a receiving device is installed on the other foot, and steps are counted by detecting whether the...

AI Technical Summary

Problems solved by technology

[0008] 1. When using the pedometer, the position of the pedometer is different, the range of the user's movement is different, and the movement of the user's two legs is asymmetrical, resulting in a great difference in the acceleration peak value generated by different users, so it is difficult to set Set an effective peak threshold to judge whether the peak appears to represent a valid step, resulting in inaccurate step counting

[0009] 2. In actual use, because the user's movement is very complicated and the device does not necessarily move with the human body, there is a lot of noise in the measured three-axis acceleration data, and even several peaks appear in one step valley, such as figure 2 As shown, because the distribution of these peaks is relatively scattered, its low-frequency components overlap with the low-frequency components required for step counting, and cannot be effectively removed by filtering means, so it will inevitably lead to step counting errors

Images