SEMANTIC NEIGHBORHOOD SENSOR NETWORK FOR SMART SURVEILLANCE APPLICATIONS

Smart situation detection is a key feature of today’s surveillance systems. The human operator has to be relieved from overseeing all activities of a monitored area simultaneously. By using smart sensors the operator can be notified about uncommon situations and evaluate only these situations. A key feature of such a system is the flexibility of smart node interaction and the ease of setup. This paper introduces a new method to establish communication paths in large-scale multi-modal sensor networks. Nodes use wireless communication allowing high-range, medium-bandwidth, and secure communication. Some of the nodes also possess wired Ethernet connections to allow the transport of video streams to the supervisor and to make the network truly scalable in terms of number of nodes without scarifying bandwidth. In a stepwise procedure the nodes first discover neighbor nodes by means of wireless connection strength and then try to find overlaps on the semantic level by utilization of the loopy belief propagation algorithm. This procedure is executed various times with increased wireless transmission power what ensures scalability. Finally, all nodes know about their neighbor nodes with overlapping sensor fields and can establish direct communication between them.