Fire erupts in a warehouse in an industrial section of town. A wireless sensor network installed in the building feeds detailed data to fire crews arriving on the scene, describing the location, characteristic and etiology of the fire, and predicting its future path. The result: firefighters are able to work quickly and safely to bring the blaze under control.An unexpected late summer rainfall soaks a vineyard in northern California. The vineyard manager logs onto her notebook computer and accesses data from soil moisture sensors installed throughout the vineyard, so she can determine the impact of the rainfall and adjust irrigation schedules if needed.The day after a moderate earthquake jolts the city of San Francisco, building inspectors check on the structural integrity of an office building in the financial district. Sensors embedded in the walls of the building to monitor and record vibration data confirm that the structure is safe to enter.
The Power of Wireless Sensor Networks
The scenarios above hint at the enormous potential of wireless sensor networks. By delivering detailed data about the physical world, sensor networks can vastly enrich our understanding of how the world works, opening the door to entirely new computing applications. For example, sensor networks could provide precise information about crops in real time, enabling agribusiness to reduce water, energy, and pesticide usage and enhancing environment protection. In industrial settings, they could be used to monitor equipment continuously and predict equipment failure, or the need for maintenance, with far greater precision, enabling companies to avoid costly equipment failures or shutdowns of production lines. These are just two of the growing number of potential applications of sensor network technology.
Intel Research, working with the academic community and industry collaborators, is actively exploring the potential of wireless sensor networks. The research is already demonstrating the potential of this new technology to enhance public safety, reduce the cost of doing business, and bring a host of other benefits to business and society.
How They Work
Wireless sensor networks (also called ad hoc or multi-hop networks) are formed by small nodes or "motes"- tiny, self-contained, battery-powered computers with radio links that enable the motes to self-organize into ad hoc networks, communicate with each other and exchange data. Motes form the building blocks of wireless sensor networks. Data from individual nodes moves across the network by wirelessly hopping from node to node (hence the "multi-hop" label), typically flowing towards a super-node, or server, with greater processing power.
Intel Motes currently run an open source operating system called TinyOS, developed by researchers at UC Berkeley. Each mote links up with its neighbors from the moment it is turned on. Although each sensor has limited power and processing capabilities, a collection of motes that spontaneously organize into a network can perform tasks no ordinary computer system could.
Using Heterogeneous Sensor Networks to Improve Performance
As sensor networks increase in size, their performance can degrade. Here's why: In a typical wireless sensor network, data gathered from each node is collected at a single server that acts as the "gateway" to an IP network. As data hops from mote to mote across the network to get to the gateway, some data gets lost, and the problem increases with the size of the network. In addition, when a node sends a data packet to a neighboring node, and the neighbor has to forward it, this takes energy. The bigger the network, the more nodes that must forward data, and the more energy that is consumed. The end result: as the network grows, performance declines.
To address these performance issues, Intel has pioneered techniques and protocols to leverage heterogeneity in sensor networks. Our approach allows a sensor network to automatically leverage resources in the environment, such as a corporate network infrastructure, a wall power socket, or a node with a high-performance processor, at a subset of nodes to increase network lifetime and data fidelity.
For example, a sensor network might utilize a corporate 802.11 network (or an ad hoc 802.11 network). High-end nodes, such as Intel XScaleŽ -based nodes, are connected by the wireless network and overlaid on a sensor network. The structure is analogous to a highway overlaid on a roadway system. Sensors can enter and exit the 802.11 "highway" at multiple interchanges (the Intel XScaleŽ nodes) in order to bypass side roads (motes). Intel researchers have demonstrated that removing the gateway bottleneck enables data to move more rapidly across the network and results in greater reliability and less energy usage.
Industrial and Commercial Applications Being Explored at Intel
Intel continues to advance sensor network technology and is now exploring commercial and industrial applications of the technology, through a new group called the Intel Research Sensor Network Operation (SNO). In addition to continuing Intel's R&D efforts, the new operation is simultaneously developing new sensor technology (including the new Intel Mote 2 and software) and overseeing the many field deployments of sensor networks. Two key deployments, described below, involve testing industrial applications of sensor network technology-one in an Intel fabrication plant, another in an oil tanker on the North Sea. Such industrial applications could potentially save companies millions of dollars.
Detecting Equipment Vibration at an Intel Fabrication Plant
Intel is conducting a trial deployment of a wireless sensor network to monitor the health of semiconductor fabrication equipment in one of its plants in Oregon. Specifically, the network senses the vibration signature of water purification equipment, providing data for preventive maintenance operations. In the same way that a car's engine sounds "right" when it is well tuned, heavy equipment has a characteristic vibration signature in normal operation.
Intel currently uses manual monitoring to predict failures and schedule maintenance or replacement to avoid costly manufacturing downtime. Deploying wireless sensor networks, which can be installed inexpensively and provide more frequent and more reliable data, could reduce equipment downtime and eliminate costly manual equipment monitoring.
Preventive Maintenance on an Oil Tanker in the North Sea: The BP Experiment
In a joint research effort by Intel and BP, a crude oil tanker based in the Shetland Islands in Northern Scotland was used as a test site for deploying wireless sensor networks to continuously monitor machinery vibration.
Intel recently collaborated with BP, one of the world's largest petroleum and petrochemicals companies, to test the use of sensor networks to support preventive maintenance on board an oil tanker in the North Sea. BP wanted to determine whether a sensor network could operate in a shipboard environment, where it would have to withstand temperature extremes, substantial vibration and significant rf (radio frequency) noise in certain parts of the ship.
An interior view of the ship's main engine room.
A sensor network was installed onboard the ship and operated successfully for over four months. During this trial deployment, the system gathered data reliably and recovered from errors when they occurred. The project was recognized by InfoWorld as one of the top 100 IT projects in 2004, an award given to "innovative new projects that highlight the resourcefulness of the IT community." BP is now exploring the use of sensor network technology throughout the company, in shipping, manufacturing and refining operations.
Through these deployments and others, Intel is demonstrating the feasibility and benefits of sensor network technology.
Intel Sensor Network Platform Kit to be Released Later this Year
The Intel Research Sensor Network Operation will release a development kit later in 2005. The kit will be available to researchers, OEMs, solution providers and others wanting to develop real world applications utilizing Intel sensor network technologies. The kit will leverage the 32-bit Intel Mote 2 platform, which represents the next generation of mote technology. It combines the Intel PXA271 processor, an 802.15.4 radio, and a rich set of I/O interfaces to enable the integration of a variety of sensors.
By incorporating Intel XScaleŽ technology, Intel's sensor network platform allows data analysis to be pushed into the network, reducing the communication overhead and increasing energy efficiency (and hence battery lifetime). The powerful capabilities of this sensor network platform will enable a range of demanding new applications.
To address these performance issues, Intel has pioneered techniques and protocols to leverage heterogeneity in sensor networks. Our approach allows a sensor network to automatically leverage resources in the environment, such as a corporate network infrastructure, a wall power socket, or a node with a high-performance processor, at a subset of nodes to increase network lifetime and data fidelity.
For example, a sensor network might utilize a corporate 802.11 network (or an ad hoc 802.11 network). High-end nodes, such as Intel XScaleŽ -based nodes, are connected by the wireless network and overlaid on a sensor network. The structure is analogous to a highway overlaid on a roadway system. Sensors can enter and exit the 802.11 "highway" at multiple interchanges (the Intel XScaleŽ nodes) in order to bypass side roads (motes). Intel researchers have demonstrated that removing the gateway bottleneck enables data to move more rapidly across the network and results in greater reliability and less energy usage.
