Machine Learning Based Noise Monitoring For Construction Activities
Construction noise is one of the major sources of noise pollution. It not only affects the site, but also the
surrounding areas, businesses and residents. Depending on the extent, complexity, schedule, location and method of
construction, it creates varying extent of noise impact. There are many noise regulation such as BS 5228, BS 7580 in the UK
and 2002/49/EC across Europe to regulate this construction noise, however there is no standard mechanism to assess the
impact. This has created a scope to log, identify the noise and its magnitude using a low power noise monitoring equipment.
It should log and identify noise, and differentiate various noise types such as Public addressing, Construction activities -
demolition, and hammering, reversing truck warning signals. The system should also provide a noise map of the locality and
its surrounding area. This paper describes the development of a custom wireless sensor board for noise identification,
monitoring and localisation using a low-power DSP and microcontroller alongwith the development of frequency selective
and machine learning based approach to discriminate various types of noise sources. The system stores noise samples to a
local SD memory card as well for future analysis and its local processing and wirelessly transmits a summary of significant
noise events in real time. This paper describes the result of the test carried over an upcoming construction site in London.
The noise event and their efficiency has been compared with high resolution, beam forming technology based SeeSV-S205
audio camera. Future work on the system will include further testing on other congested sites.
Keywords— Machine Learning, Support Vector Machine, DSP, Construction Noise.