Design And Implementation Of Low Power Pedometer Based On ADXL362

With the rapid development of the Internet of things and people to their health gradually attention,pedometer as a wearable device of quantitatively monitoring the movement of people has walked into the thousands of households.This paper introduces a low power pedometer based on ADXL362,the pedometer is fixed by way of embedded in the sole,future is expected to use the piezoelectric materials to take the mechanical energy into electrical energy to supply power for the pedometer by the foot movement.Firstly,based on the STM32 development platform,a set of acceleration data acquisition system is built.The system uses USB to TTL module to upload the acceleration data to the computer.Using Serial_Digital_Scope V2 open source software,we get real-time display acceleration data waveform,and this software can save the data as Dat.file,at the same time completed the Dat.file data protocol analysis.Secondly,the working principle of FIR filter is studied,and the FIR filter parameters are designed by using the window function on MATLAB software.Finally,considering the actual filtering effect and time efficiency of the filter,a class of special filters in FIR filter,called sliding average filter,are selected.Based on the analysis of gait,the extraction strategy of swinging region and support region in a complex cycle is proposed.And adapt to different motion modes frequency to solve the problem of fixed threshold cannot,proposes an adaptive peak step algorithm,MATLAB simulation test proves the feasibility of the algorithm.Finally,the pedometer which this paper designed and smart running shoes which mi technology designed in 2017 compared step accuracy in four modes of movement: walking,running,going upstairs and downstairs.The test results show that the pedometer of this paper have reached the step accuracy above 99% in walking and running mode,and the accuracy of about smart shoes of mi technology;But in the upstairs and downstairs test,the pedometer accuracy of this paper was above 98%,and smart shoes of mi technology in the upstairs mode step precision is less than 85%,downstairs mode is about 94%.Therefore,the design of the pedometer step counting accuracy in the downstairs and upstairs mode is obviously higher than that of smart running shoes of mi technology.But in terms of power optimization of hardware circuit control or step pedometer algorithm,this paper have a lot of improve space.