Ambient acoustic noise monitoring solution based on NB-IoT

Abstract

The increasing market for the Internet of Things (IoT) led to the Low-power widearea network (LPWAN) movement. This project investigates the use of the Narrowband Internet of Things (NB-IoT) network technology for acoustic sensing. The solution allows monitoring of eco-acoustics effects in biodiverse environments. This can lead to the development of policies and support enforcement of regulations and laws. This final project includes a prototype system using an acoustic sensor to record audio, a narrowband device to send and process the data, and a dashboard for remote monitoring. Communication between the narrowband device and the dashboard was achieved using the application layer protocol Message Queue Telemetry Transport (MQTT) with the use of an LTE base station. The performance of the technology was also evaluated to understand whether propagation modelling tools can be used during the design of such an IoT solution. Okumura Hata Propagation models were used to predict the values of the path loss and assess which model best predicts them. After collecting data from a base station, it was concluded that as the distance from the narrowband device to the base station became longer, the signal strength became weaker. In areas where buildings are evident, the ‘Large City’ model fit the best, and in areas where there were less high structures, the ‘Suburban Environment’ model fit best. MATLAB was used consistently to analyse and investigate the data collected throughout this study. Two experiments were conducted using the implemented system. The aim of these tests was not only to see the capabilities of the system but to also research the impact of noise on our environment. This was done by analysing sound levels to ensure quality sleep. The recommended noise level inside a bedroom for minimal sleep disturbance during the night-time is an average of 30dB over a period of 8 hours with closed windows, and 45dB with open windows. From the experiments, the average noise level with a closed window resulted to be 38.33dB and 46.33dB with an open window in the area tested. Hence, measures need to be taken to reduce this noise.