Research On Urban Noise Monitoring System Based On Hybrid Sensing

Noise pollution is an emerging and challenging urban environmental problem that affects the physical and mental health of residents in many ways.Increasing noise pollution and the relentless pursuit of a good living environment by residents have become one of the important contradictions that cannot be ignored in urban development today.How to carry out timely,comprehensive and accurate monitoring and management of the urban acoustic environment has become an important problem to be solved urgently.Based on existing research,this paper designs a wide-range,high-precision,and energy-efficient hybrid sensing-based noise monitoring system.Its core idea is to combine wireless sensor networks and group intelligence sensing technology for noise monitoring.This hybrid sensing mode not only takes into account the advantages of fixed deployment of wireless sensor nodes for monitoring stability and accurate data collection,but also takes advantage of the flexible deployment of the mobile group intelligence sensing mode and the wide range of perception,combined with the advantages of the two.Collaborate to complete large-scale perception tasks.In order to improve the accuracy of the hybrid sensing monitoring system and reduce the cost of system consumption,this paper optimizes the system from three aspects.First,since mobile phone monitoring cannot have the same performance as a standard sound level meter,we propose an online distributed calibration method based on geometric mean regression to correct the collected data of the smartphone,thereby making it more accurate.Secondly,the mobile phone will be affected by the user's behavior when collecting data.According to this paper,it is proposed to identify the user's motion state to optimize the data collection system and reduce the measurement error of the collected data.This article also proposes an optimized solution for the mobile terminal collection process,the purpose is to reduce the system energy consumption of the mobile phone to perform the collection task.Finally,an experimental evaluation of the system was performed.The experiment showed that the average monitoring error after optimization using the calibration scheme and acquisition system proposed in this paper was 0.89 dBA,which was an average reduction of 85.38% compared to without optimization.The time overhead and energy consumption are reduced by 58% and 45% compared to the J48 algorithm,respectively.