No abstract

The advent of technology has facilitated the development of networked systems of extremely small, low power devices that combine programmable general purpose computing with multiple sensing and wireless communication capability. This networked system of programmable sensor nodes, together f o r m a sensor network, poses unique challenges o n how information collected by and stored within the sensor network could be queried and accessed, and how concurrent sensing tasks could be executed internally and programmed by external users. In this paper, we describe a sensor information networking architecture, called SINA, that facilitates querying, monitoring, and tasking of sensor networks. W e model a sensor network as a collection of massively distributed objects, and SINA plays the role of a middleware that facilitates adaptive organization of sensor information. The S I N A kernel provides a set of configuration and communication primitives that enable scalable, robust, and energy-eficient organization of and interactions among sensor objects. O n top the S I N A kernel is a programmable substrate that follows the spreadsheet paradigm and provides mechanisms t o create associations among sensor nodes. Users then access information within a sensor network using declarative queries and perform tasks using programmable scripts. Issues concerning interworking between stationary sensor networks and mobile nodes will also be addressed.

No abstract

Absmcr-As sensor nodes are embedded into physical environments and becoming integral parts of our daily lives, sensor networks will become the important nerve systems that monitor and actuate our physical environments. We define the process of monitoring the status of a sensor network and figuring out the problematic sensor nodes .seiz.sor fzework diupmis. However, the high sensor node-to-manager ratio makes it extremely difficult to pay special attention to any individual node. In addition, the response implosion problem, which occurs when a high volume of incoming replies triggelvd by diagnosis queries cause the central diagnosing node to become a bottleneck, is one major obstacle to be overcome. In this paper, we describe approaches to addressing the response implosion problem in sensor network diagnosis. We will also present simulation experiments on the performance of these approaches, and discuss presentation schemes for diagnostic results.

No abstract