Siemens Press Picture/ PolyC Company RFID tags can differentiate individual product items to which they are attached

Ubiquitous sensor networks

A challenge for standardization

Often, when you enter a modern office building, the glass doors open for you, and lights may come on automatically as you go into a dark room. This “magic” is achieved by motion sensors. When entering a building of the future, you might be automatically identified and welcomed with a personal greeting. This could be done by using not only motion sensors, but also a radio-frequency identification (RFID) tag embedded in your name badge that interacts with a database containing your profile.

These three elements — sensors, tags and communication and processing capacity — make up future networks that have been given several names. Some people call them “invisible”, “pervasive” or “ubiquitous” computing, while others refer to “ambient intelligence” or the “Internet of things”. In this article, the term used is “ubiquitous sensor networks” (USN).

Applications

The revolutionary aspect of USN is their ability to connect almost any object with another, through detecting, storing, processing and integrating information gathered from sensors. This forms a network that is “aware” of its context. It can tell, for example, whether an object is moving or stationary; hot or cold.

Sensor technologies have enormous potential, as they could facilitate new applications and services in a wide range of fields. Most applications fall into one of three broad categories:

• Detection of intruders, for example, or when temperatures pass a threshold. And sensors on robots could save lives and limbs by detecting landmines in former conflict zones.
• Tracking, of items in supply chains, for instance, (including cattle and meat) and of vehicles in intelligent transport systems.
• Monitoring, such as of a patient’s blood pressure, of dangerous environments, of the structural status of bridges, or of the movement of wild animals.

  The main components of a USN are: Sensor network: Comprising sensors, linked through a wired or wireless connection, and an independent power source (e.g. battery, solar power). Access Network: Intermediary or “sink nodes” collecting information from a group of sensors and facilitating communication with a control centre or with external entities. Network infrastructure: likely to be based on a next-generation network (NGN). Middleware: Software for the collection and processing of large volumes of data. Applications platform: A technology platform to allow USN to be used in a particular industrial sector or situation.

Depending on the sensor type, the links between them may be provided by either wired or wireless communication. Domains in which USN are used include civil engineering, education, health care, agriculture and environmental monitoring. They also have much to offer regions that are vulnerable to natural disasters, where extensive warning systems are needed to prevent loss of life and property. Receiving quick, accurate data is vitally important when responding to natural disasters. High-resolution remote sensing data, for example, are especially useful for documenting hazards, and for determining where to locate facilities to respond to disasters, store emergency supplies and plan reconstruction.

Monitoring the activity of two dangerous volcanoes in Ecuador, for example, was the aim of a joint project by vulcanologists and computer scientists from Harvard University and three other institutions in the United States. They deployed a wireless sensor network at the Tungurahua and Reventador volcanoes to collect seismic and low-frequency acoustic signals that were sent to a remote location (Figure 1). The project was among the first to use tiny, low-power wireless sensor nodes, which are cheaper, lighter and smaller than previous systems.

Just 16 battery-powered sensor nodes were installed on the upper slopes of Reventador (see photograph below, right). The collected data were sent over a long-distance radio link to an observatory. Over three weeks, the network captured 230 volcanic events, producing useful data for scientific research and disaster prevention. The low cost, size, and power requirements of wireless sensor networks give them an advantage over previous instrumentation.

Figure 1 — Monitoring a volcano
Source: Harvard University, School of Engineering and Applied Sciences.

Wireless sensor networks have also found commercial applications in the construction industry. For example, in San Francisco, United States, a wireless sensor network is embedded into the Golden Gate Bridge to continuously monitor stress loads. In bad weather or earth tremors, engineers receive alerts and can take action to keep the bridge safe.

Targeted services could also benefit from USN. For example, shops and restaurants could send messages to the mobile phone of a potential customer in the vicinity with matching tastes. USN could also provide directions to a mobile device used by a person with disabilities, so that he or she could follow an accessible route.

Standardization is needed

Standardization is essential for the effective diffusion of any technology. However, because they are highly diverse, USN present a standardization challenge that is unusually complex.

Some sensor nodes are so tiny (so-called “smart dust”) that they involve nanotechnology, while the need to support mobility could mean working with a range of wireless standards — from second-generation (2G) to third-generation (3G) and WiMAX — as well as other technologies such as near-field communications. Furthermore, because a USN may provide a platform for a wide range of applications (many of which have unique requirements), there is also a need to standardize common elements that can be shared by applications. One of the most important issues is the development of protocols for sensor networks, as well as interoperation with backbone network infrastructure such as next-generation networks (NGN).

Stockxpert. Developing countries could gain by manufacturing RFID microchips (above) and tags (below), which are increasingly in demand for USN

A number of standards development organizations (SDO) are working on USN, which can either be based on the Internet Protocol (IP) or on other protocols. ITU’s Telecommunication Standardization Sector (ITU–T) has established a platform to help coordinate the work of SDO: the Joint Coordination Activity on Network aspects of Identification systems, (JCA-NID), including RFID and USN. Current efforts include the development of a standards roadmap and a harmonized terms and definitions document relevant to USN, which aims to enhance the common understanding of technical challenges at an international level (for more information visit www.itu.int/ITU-T/jca/nid).

Within ITU–T, standardization of USN is being carried out in various Study Groups under the Next-Generation Network Global Standards Initiative (NGN-GSI). At an NGN-GSI meeting held in Seoul in January 2008, for example, the Republic of Korea’s Electronics and Telecommunications Research Institute (ETRI) submitted a proposal to Study Group 16 for a new study question on USN applications and services. The proposal foresees a work programme of new and amended Recommendations for completion by 2010. This could form part of a larger programme of work on “Beyond NGN”.

Meanwhile, ZigBee, for example, provides a suite of communication protocols. Released in 2004, it is an implementation of the IEEE 802.15.4 standard for wireless personal area networks (WPAN). Alternative technologies include ultra-wide band (UWB), Bluetooth, and WiBree.

Benefits for developing countries

The falling prices of sensor units (below USD 100) and RFID tags (below 5 US cents) are increasing the number of potential applications for USN. In developing countries, USN could be applied in situations where network engineers face particular challenges, such as unreliable power supplies. Because they can be operated with batteries or solar power, USN can be deployed in diverse environments.

Although most of the research and standardization work on USN is taking place in the developed world, it can be said that developing countries will benefit most from the technology. For example, the manufacturing of RFID chips and sensors as components of USN is likely to soon become a commodity business. China has emerged as a leading manufacturer of RFID chips, and developing countries that have a software sector (such as India, the Philippines and Viet Nam) could benefit from contracts to create customized USN middleware. On the demand-side, it is also likely that developing countries will be major beneficiaries, especially in the field of environmental monitoring. They are most at risk from natural disasters related to climate change, with particular vulnerabilities among least developed countries and small island developing States. In other applications, such as landmine clearance or agricultural management, in the long term, developing countries might become the main users of USN.