Step length estimation method based on MEMS inertial sensor and FM broadcast signal

Abstract

The invention discloses a step length estimation method based on an MEMS (Micro Electro Mechanical System) inertial sensor and an FM (Frequency Modulation) broadcast signal, which combines inertial sensor characteristics and FM broadcast signal characteristics through a support vector machine regression method to realize step length estimation, and comprises the following steps of: realizing step number detection of pedestrians based on an MEMS accelerometer; multiple features are extracted based on MEMS inertial sensor signals and FM broadcast signals, dimensionality reduction is performed on the extracted multiple features by using a principal component analysis method so as to eliminate redundant features, finally, the multiple features are fused by adopting SVR, and step length estimation is realized after two stages of training and prediction. According to the method, the MEMS inertial sensor and the FM broadcast signal are used for step length estimation, and priori information such as height or a parameter adjustment process is not needed; by introducing FM broadcast signal characteristics which are related to distance change and are not influenced by the movement speed or people, the negative influence of MEMS inertial device deviation and noise on the performance is reduced, and the practicability of the algorithm for different speeds and different people is also ensured.

Description

technical field [0001] The invention relates to the technical field of indoor navigation and positioning, in particular to a step size estimation method based on MEMS (Micro-Electro-Mechanical System) inertial sensors and FM (Frequency Modulation) broadcast signals. Background technique [0002] With the widespread use of smartphones in recent years, indoor positioning has become more and more widely used, including museum guides, shopping guides, search and rescue, mobile advertising, and targetable social networks. After decades of development, the Global Navigation Satellite System (GNSS) has become a standard solution for outdoor positioning. However, indoors, people move more frequently than outdoors. Frequent human activities and a large number of walls, doors, furniture and other factors cause GNSS signals to be seriously affected by attenuation and multipath indoors, and the conditions for unobstructed straight-line transmission are often difficult to achieve. . Th...

AI Technical Summary

Problems solved by technology

This method has strong flexibility and basically does not require parameter adjustment process. However, most of the existing research requires rich training set data, so the preliminary work is relatively large; in addition, the existing research only uses the characteristics of inertial sensors such as acceleration. However, the device measurement will bring noise, especially when the low-cost MEMS inertial sensor is selected as the measurement tool, the measurement accuracy will be severely limited, and the acceleration characteristics are highly dynamic, and are less affected by people and the environment. is large, so the performance of the step size estimation is poor without the aid of other information

Images