Residential Building Natural Gas Leak Monitoring System Based On Wireless Sensor Network

Nowadays, natural gas is a major fuel for domestic cooking and heating in manycities. On the one hand natural gas has brought us convenience; the other handnatural gas leak issue has been a threat to people’s normal work and life. Explosioncaused by gas leak takes place every year. The explosion will bring significanteconomic losses, casualties, and bad society effect, because it usually triggers a greatarea of the collapse.To solve the above-mentioned problem, this paper designs a sensornetwork-based intelligent natural gas leak monitoring system which focuses on datacollection protocol and backend Java Web monitoring system, aiming to achievereal-time alarm message delivery and node status maintenance. The data collectionprotocol consists of a direction-limited flooding algorithm and a periodic statusacquisition scheme. The direction-limited flooding algorithm is designed to deliverthe gas leak alarm message. The alarm messages are only flooded along thedirection nearer to the sink node, thus the number of broadcasts is limited to avoidbroadcast storm. The periodic status acquisition scheme is explored to maintain thenode status of existence. Each node collects HELLO packets from its neighbors todeduce the node failure. A node failure alarm message is delivered to the sink nodewhen a node is detected failure. On the other side we used a B/S architecture, SSHframework as the main frame, Java, JS, Dojo, CSS, HTML5, JQuery, MySQL tosupport server-side applications to efficient, stable, convenient and intelligentmonitoring and alarming on natural gas leak.This paper implements a gas leak monitoring prototype system. The proposeddata collection protocol and backend Java Web monitoring system are implementedin the system. The experiments are designed and carried out to verify the datacollection protocol. We first measure the link quality across walls in buildings underdifferent Wi-Fi interference. The measurement results show that the protocol canwork by selecting frequency channel carefully. The experiment results validate thedata collection protocol.