1 Abstract—Anaerobic decomposition of landfill garbage produces gases like CH 4 and H 2 S at subsurface. These gases diffuse through the foundation and come out to pollute the indoor building environment above. CH 4 is highly flammable and causes explosion whereas H 2 S causes equipment failure and significant pulmonary ailments in human beings over long term exposure. This work focuses on 3D ad-hoc wireless sensor network (WSN) based localization and risk assessment of such a gas source buried at subsurface. Experiments are performed to emulate the gas diffusion process at underground. Experiments also help to determine gas propagation constants over the surface. Higher precision of the three- step maximum likelihood (ML) based source localization method enhances the accuracy of source risk assessment process. Simulation results indicate that the proposed localization algorithms with limited sensor node mobility offer higher accuracy of estimation using lesser network resources. Finally, we propose an analysis for remote source concentration measurement. This help estimate severity of threat possessed by the buried landfill gas source. Also a method is proposed for 3D ad-hoc WSN based source location estimation with obstructed paths for surface gas propagation. Keywords—Buried landfill source, gas source localization, risk assessment, 3D ad-hoc wireless sensor networks. I. INTRODUCTION UE to space crunch in megacities, new buildings are constructed over not so old former garbage landfill sites. Anaerobic decomposition of the wastes at moist and humid subsurface produce hazardous gases like CH 4 , CO 2 , H 2 S, benzene, trichloroethene, toluene, xylene etc. Under the effect of negative pressure, these gases diffuse through the soil and concrete foundation and come out to pollute the building Manuscript received October 28, 2011. This work is supported by the Crompton Greaves Limited, Kanjurmarg, Mumbai, India. Saurav Mitra is a Ph. D. research scholar with the Indian Institute of Technology Bombay, Powai, Mumbai, Maharashtra, India (phone: +91-22- 2576-7866; fax: +91-22-2572-3707; e-mail: sauravm@ee.iitb.ac.in). Siddhartha P. Duttagupta is assistant professor with the Indian Institute of Technology Bombay, Powai, Mumbai, Maharashtra, India (phone: +91-22- 2576-7866; fax: +91-22-2572-3707; e-mail: sdgupta@ee.iitb.ac.in). Kushal Tuckley is director and chief of research and development of AGV Systems Pvt. Ltd., Goregaon (E), Mumbai, MH, India (Phone: +91-22-2685- 3634; fax: +91-22-2572-3707; e-mail: kushal.tuckley@agv-systems.in). Samsul Ekram is assistant general manager of Crompton Greaves Limited, Kanjurmarg, Mumbai, Maharashtra, India (Phone: +91-22-6755-9083; fax: +91-22-6755-9008; e-mail: samsul.ekram@cgglobal.com). environment above [1]-[3]. In Grand Riviera Princess Hotel at Playa Del Carmen, Mexico, accumulation of methane from nearby swamp caused huge explosion at basement. It rendered many deaths and severe financial damage [4]. In another incident at Malad office of Mindspace BPO, Mumbai subsurface H 2 S infiltrated through the floor and led to incessant crashing of computers, and electrical equipments [1]. In less aerated offices, industries, and residential buildings, the ailing effects of such gases on human beings can create nausea, headache, and serious lung diseases like asthma, bronchitis, and even cancer [2]. In such scenarios, it is of immediate necessity to track the source of contamination and identify the degrees of threat that the gas sources posses towards employees and equipments within [5]. This helps the proper authorities to deal with the crisis and restraining the hazardous effects through preventive measures. Landfill gases (LFG) generally contains CO 2 and CH 4 as dominant emission elements (45-50% each), and H 2 S (3-5%) and other volatile organic compounds (VOC-1%) as trace elements. For its distinct rotten smell, presence of H 2 S in the indoor environment is easily felt even for ppm level concentration. But CH 4 concentration determination at source is highly challenging as CH 4 is odorless and colorless. 5 to 15% of CH 4 in indoor environment produces a highly flammable mixture in presence of O 2 /air [2]. Over a large surface area, manual localization of such gas sources and remote monitoring of gas effusion rate would be highly cumbersome and ineffectual. But with the help of dedicated self-powered wireless sensor nodes, this type of critical event monitoring becomes lot easy. These multi sensor nodes when networked together can provide quick real-time detection, identification and threat assessment with higher accuracy. Examples include monitoring of the accidental spillage of toxic waste by ships, factories etc that contaminate the sea and the air, early detection of fires, nuclear radiation monitoring, and other environmental monitoring and so on [6]. Environmental monitoring and toxic gas source localization is one of the most promising fields of research since it enhances human safety in adverse residential and industrial environments [7]. One of the common issues has been that most detection of chemicals with mobile robots has been based on experimental setups where the distance between the source and the sensor following an odor trail has been minimized to limit the influence of turbulent transport [8]- [10]. In many cases mobile robots are also used for localization of toxic chemical plume source for explosive 3D ad-hoc sensor networks based localization and risk assessment of buried landfill gas source Saurav Mitra, Siddhartha P. Duttagupta, Kushal Tuckley, Samsul Ekram D INTERNATIONAL JOURNAL OF CIRCUITS, SYSTEMS AND SIGNAL PROCESSING Issue 1, Volume 6, 2012 75