A method of pedestrian flow monitoring based on received signal strength

Zhiyong Yang,Kaide Huang,Jing Wen

doi:10.1186/s13638-021-02079-y

Abstract

There is a wide demand for people counting and pedestrian flow monitoring in large public places such as scenic tourist areas, shopping malls, stations, squares, and so on. Based on the feedback from the pedestrian flow monitoring system, resources can be optimally allocated to maximize social and economic benefits. Moreover, trampling accidents can be avoided because pedestrian guidance is carried out in time. In order to meet these requirements, we propose a method of pedestrian flow monitoring based on the received signal strength (RSS) of wireless sensor networks. This method mainly utilizes the shadow attenuation effect of pedestrians on radio frequency (RF) signals of effective links. In this paper, a deployment structure of RF wireless sensor network is firstly designed to monitor the pedestrians. Secondly, the features are extracted from the wavelet decomposition of RSS signal series with a short time. Lastly, the support vector machine (SVM) algorithm is trained by an experimental data set to distinguish the instantaneous number of pedestrian passing through the monitoring point. In the case of dense and sparse indoor personnel density, the accuracy of the SVM model is 88.9% and 94.5%, respectively. In the outdoor environment, the accuracy of the SVM model is 92.9%. The experimental results show that this method can realize the high precision monitoring of the flow of people in the context of real-time pedestrian flow monitoring.

Highlights

Pedestrian flow monitoring is a research hotspot in the field of computer vision and intelligent monitoring [1, 2]
To evaluate the performance of the proposed received signal strength (RSS)-based pedestrian flow monitoring method, we compare the performance of the system using support vector machine (SVM) algorithm with the other two algorithms, the algorithm based on the model of Gaussian Mixture Model (GMM)–Hidden Markov Model (HMM) [32] and the gradient boosting decision tree (GBDT) algorithm
6 Conclusion and future work In this paper, a pedestrian flow monitoring method based on radio frequency signals is proposed

Summary

Introduction

Pedestrian flow monitoring is a research hotspot in the field of computer vision and intelligent monitoring [1, 2]. A large number of research results have sprung up in recent years, most of them mainly using individual characteristics such as shape, color, outline, or population characteristics to achieve pedestrian flow monitoring through a combination of SVMs, BP neural networks, CNNs [17, 18], and other machine learning algorithms [19,20,21]. We propose a method of pedestrian flow monitoring based on RSS of RF signal. By processing the received RSS data and using the SVM algorithm for classification modeling, real-time pedestrian flow monitoring can be achieved. The RSS of each wireless link varies with different numbers of people passing through the monitoring area This verifies the feasibility of pedestrian flow monitoring by using wireless link RSS based on a wireless sensor network. As can be seen from the waveform after wavelet filtering, the RSS fluctuation of the same effective link under the same condition tends to be stable, which is conducive to the establishment of the later model

SVM classification

Experiments and results

Findings

Conclusion and future work