Research And System Implementation Of Pedometer Detection Method Based On Mobile Phone Sensor

As the pace of work in city life continues to accelerate and people's eating habits change,urban health problems continue to intensify.Real-time detection of the number of steps can help people control their exercise in real-time,maintain a moderate daily exercise,enhance their physical fitness,and thus leads to people's high immunity,prevention of chronic diseases and occupational diseases purposes.Although the existing step-by-step method shows a good accuracy,but the irregular pace of recognition is not high,and the anti-interference is not prominent enough,some non-walking action on the step test results in significant interference and causes false high data.According to the demand of daily motion detection and anti-jamming,a kind of step-by-step detection method with good anti-jamming is designed in this paper.It includes the selection of walking characteristics,the determination of threshold and the design of process and the step achieve.The specific research work is as follows:(1)By analyzing the changes in the acceleration of the human body walking,the three characteristics that are most relevant to the walking action are extracted: the amplitude feature to show the range of the intensity of the movement;the angle characteristic to reflect the changes in the direction of the movement of the human body;distance characteristics to show the movement of the body when the displacement changes.Furthermore,the feature calculation method is designed for the three features.(2)The determination of the three characteristics of amplitude,angle and distance is essential to improve the anti-interference of the step method.Firstly,the initial threshold setting principle is established by comparing the characteristics of the interference action and the time domain of the walking action.Then,the initial threshold of the three features is determined by experiment and the threshold correction method is designed.(3)Based on the study of the interference filtering method,a step-by-step method is designed to realize the step-by-step system through the Android platform,and the pedometer and other step-by-step method are compared and tested.The experimental results show that the system has qualified recognition efficiency in the normal walking situation,the accuracy rate of the identification of the continuous step is increased by 10%-18%,and the error rate of non-walking behavior reduced to 6% to 12% which reflects good anti-jamming.