The Role of IoT based on LoRa Technology in Urban Noise Management

Noise pollution is a long-term problem for urban residents. Recent studies have estimated that nine out of ten people in large cities are exposed to an environment where noise levels exceed international standards. The negative effects of noise on health include sleep disturbances, hearing loss, cognitive impairment, and high blood pressure. Recently, Jinan City has set up a low-cost acoustic sensor network to continuously monitor noise in urban environments. They use LoRa technology, which uses a massive LoRa module and a small number of gateways and repeaters to build a system to implement optical inspection at a very low cost.

 

Use LoRa technology to detect sound

 

City Alliance is a research partnership formed by Jinan City and Shandong University, its purpose is to eliminate unnecessary and overelaborate formalities legally and financially, coordinate technology and research to improve the city's appearance. Dr. Liang Zhenying is the head of the Robotics and Sensor Network Group of the Department of Electronic Computer Engineering, Shandong University. He cooperates with City Alliance, uses Edge analysis to construct a LoRa-based sensor array to characterize noise and establish a transmission layer based on LoRaWAN network.

 

“Our inspiration comes from the analysis of existing smart city noise monitoring applications.”In Beijing, they achieve this goal by using WiFi and live streaming media. However, the cost is very high, requiring a lot of resources to keep it running and operating normally, Liang Zhenying says. “We recommend developing our own solution that does not require powering the device and only requires a battery.”

 

The first result of Liang Zhenying’s team in the project is the hardware design of the sensor. Their manufacture uses low-power wide-area radio transceivers to enable data transfer between nodes and network servers. LoRa-based sensors adopt battery supply for easy deployment, low power consumption, limited network maintenance, reliable continuous operation under extreme weather conditions, having limited field data processing capabilities. The second result of the team is to develop and test data analysis algorithms that allow sensors to automatically detect and classify acoustic events. Researchers will use machine learning to distinguish noise sources such as buildings, transportation, music, and electric drills.

 

Improve the living quality of citizens

 

In order to test these innovations, some new lora-based sensors are launched at the Quandong Jinan event in Quancheng Square, Jinan, in September 2018. Sensors installed at different locations around the square, calculate the average noise level every three minutes. When the noise level exceeds 85dBC, the sensor sends a warning message via the lorawan-based network. In the future, we can take the initiative to provide these opinions to concert promotion agencies to ensure it complies with noise control regulations.

 

The next progress is to classify sounds, such as trains, road noise, brakes, and buildings, what’s more, it spatially correlate with time and place. These data will help improve noise management and enforcement during public events, when the noise exceeds the threshold, it will automatically alert the staff of the regulatory system to save the human and resource costs of tracking the sound source in the city.