Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Filters applied. Clear all
. 2013 Nov 27;10(12):6380-96.
doi: 10.3390/ijerph10126380.

Monitoring Street-Level Spatial-Temporal Variations of Carbon Monoxide in Urban Settings Using a Wireless Sensor Network (WSN) Framework

Affiliations
Free PMC article

Monitoring Street-Level Spatial-Temporal Variations of Carbon Monoxide in Urban Settings Using a Wireless Sensor Network (WSN) Framework

Tzai-Hung Wen et al. Int J Environ Res Public Health. .
Free PMC article

Abstract

Air pollution has become a severe environmental problem due to urbanization and heavy traffic. Monitoring street-level air quality is an important issue, but most official monitoring stations are installed to monitor large-scale air quality conditions, and their limited spatial resolution cannot reflect the detailed variations in air quality that may be induced by traffic jams. By deploying wireless sensors on crossroads and main roads, this study established a pilot framework for a wireless sensor network (WSN)-based real-time monitoring system to understand street-level spatial-temporal changes of carbon monoxide (CO) in urban settings. The system consists of two major components. The first component is the deployment of wireless sensors. We deployed 44 sensor nodes, 40 transmitter nodes and four gateway nodes in this study. Each sensor node includes a signal processing module, a CO sensor and a wireless communication module. In order to capture realistic human exposure to traffic pollutants, all sensors were deployed at a height of 1.5 m on lampposts and traffic signs. The study area covers a total length of 1.5 km of Keelung Road in Taipei City. The other component is a map-based monitoring platform for sensor data visualization and manipulation in time and space. Using intensive real-time street-level monitoring framework, we compared the spatial-temporal patterns of air pollution in different time periods. Our results capture four CO concentration peaks throughout the day at the location, which was located along an arterial and nearby traffic sign. The hourly average could reach 5.3 ppm from 5:00 pm to 7:00 pm due to the traffic congestion. The proposed WSN-based framework captures detailed ground information and potential risk of human exposure to traffic-related air pollution. It also provides street-level insights into real-time monitoring for further early warning of air pollution and urban environmental management.

Figures

Figure 1
Figure 1
System overview: The wireless sensor network comprises two major components: (1) front-end: geo-sensor network and (2) back-end: monitoring and control platform. The sensor network is responsible for measuring the concentration of air pollutants in the study area. The monitoring and control platform receives monitoring data through a map-based system to provide real-time information of the sensor data. The data gateway nodes connect the front end and back end by wireless Internet.
Figure 2
Figure 2
Study area: The wireless sensor network was distributed along an arterial connecting a residential area and the central business district and a circle intersecting another arterial connecting the south and north of the city. The yellow dots present the air pollutant sensors, the grey dots symbolize the transmitters and the green diamonds are the data gateway nodes of the wireless sensor network.
Figure 3
Figure 3
Instruments: A sensor node includes a semi-conductor carbon monoxide sensor, a sensor data processing module and a wireless module. The semi-conductor sensor is used to measure the concentration of air pollutants with variations of voltage. The processing module transfer the voltages gathered by the CO sensor to the CO concentration. The wireless module transmits the monitoring data among sensor, transmitter, and gateway nodes. A 12-V lead-acid rechargeable battery was used to provide system power.
Figure 4
Figure 4
Settings of MiCS-5525 sensors at an EPA station for calibrating CO concentration before deploying wireless sensors.
Figure 5
Figure 5
Segmented regression was employed to transfer the voltage of the sensor to the concentration of carbon monoxide. Every sensor has its own regression equation. The method identifies a break-point to separate the regression model into two parts: the flatter trend line represents no association, and the steeper one explains the linear relationship between the concentration and the sensor voltage.
Figure 6
Figure 6
Map-based monitoring and control platform.
Figure 7
Figure 7
Comparisons of hourly average CO concentration between Sensor No.3 and the EPA station over different time of a day from July to September 2010.
Figure 8
Figure 8
The hourly interquartile range (IQR) of CO concentrations of the sensor No. 2 and No. 3 at 24 hours a day from July to September 2010 (right panel). Sensor No. 3 represents a location affected by the difference of traffic flows on weekdays and weekends, but sensor No. 2 reflects a location that always accumulates high traffic volume from downtown to the residential areas (left panel). The traffic flows from downtown (blue lines) converge on the arterial and are diverted to different roads. The difference between weekdays and weekends only has a slight effect on sensor No. 2. The greater IQR of sensor No. 3 could be explained by the traffic flow (red line) only appearing during weekdays.
Figure 9
Figure 9
Street-level spatial-temporal variations around the circle: a clear on/off-peak traffic pattern is captured by WSN. In the morning, many vehicles drive from residential to downtown areas for work. This congestion induces a high concentration of air pollutants near a traffic signal. In the evening, the traffic flow to residential areas increases and it results in high concentrations at the contrast side.

Similar articles

See all similar articles

Cited by 3 articles

References

    1. Postolache O.A., Pereira J.M.D., Girao P.M.B.S. Smart sensors network for air quality monitoring applications. IEEE T. Instrum. Meas. 2009;58:3253–3262. doi: 10.1109/TIM.2009.2022372. - DOI
    1. Corti A., Senatore A. Project of an air quality monitoring network for industrial site in Italy. Environ. Monit. Assess. 2000;65:109–117. doi: 10.1023/A:1006493031179. - DOI
    1. Ferreira F., Tente H., Torres P., Cardoso S., Palma-Oliveira J.M. Air quality monitoring and management in Lisbon. Environ. Monit. Assess. 2000;65:443–450. doi: 10.1023/A:1006433313316. - DOI
    1. Chiang C.T., Wang C.S., Huang Y.C. A monolithic cmos autocompensated sensor transducer for capacitive measuring systems. IEEE T. Instrum. Meas. 2008;57:2472–2486. doi: 10.1109/TIM.2008.925020. - DOI
    1. Riza N.A., Sheikh M. All-silicon carbide hybrid wireless-wired optics temperature sensor network basic design engineering for power plant gas turbines. Int. J. Optomechatronics. 2010;4:83–91. doi: 10.1080/15599611003650008. - DOI

Publication types

Feedback