You might know it. And because you do not understand it, it does not trouble you. Each time you connect with anybody or a website online, your network issues your device address to the site or the person you are…
Wireless sensor networks: Architectures and Applications
Since their creation, wireless Sensors networks have experienced an ever growing success in scientific and industrial communities. Thanks to its various advantages, this technology has been able to establish itself as a key player in current network architectures.
The terrestrial media indeed offers unique properties, which can be summarized in three points: the ease of deployment, the ubiquity of information and the reduced cost of installation.
During its evolution, the wireless paradigm has seen the birth of various derived architectures, such as: cellular networks, local wireless networks and others. Over the past decade, a new architecture has emerged: wireless sensor networks.
This type of network results from a merger of two parts of the modern computing: embedded systems and wireless communications. A wireless sensor network (RCSF), or “Wireless Sensor Network” (WSN), is composed of a set of on-board processing units, called “motes”, communicating via wireless links.
The general goal of a WSN is to collect a set of parameters from the environment surrounding the motes, such as the temperature or pressure of the atmosphere, in order to route them to treatment points. RCSFs are often considered to be the successors of ad hoc networks.
In fact, the RCSFs share with MANETs (Mobile Ad hoc NET works) several properties in common, such as the absence of infrastructure and wireless communications. But’ one of the key differences between the two architectures is the area of application.
Unlike the MANET networks, which have not been able to achieve real success, the RCSFs have been able to attract a growing number of industrialists, given their realism and their concrete contribution. Indeed, the need for continuous monitoring of a given environment is quite common in various activities of society.
Industrial processes, military tracking applications, habitat monitoring, as well as precision farming are just a few examples of a vast and varied range of possible applications for continuous monitoring offered by the RCSF. Thanks to this potential rich in applications, the RCSFs have known how to differentiate themselves from their MANET origin and attract large companies around the world, such as IBM, Sun, Intel and Philips.
Unfortunately, RCSFs are not perfect! Because of their low cost and their deployment in sometimes hostile areas, the motes are quite fragile and vulnerable to various forms of failure: breakage, low energy, etc. These problems make the RCSFs of innately fragile systems, which must be considered as a normal property of the network.
Applications of RCSF
RCSFs can have many applications (see following figure). Among them, we cite:
- Discoveries of natural disasters: You can create an autonomous network by dispersing the nodes in nature. Sensors can thus report events such as forest fires, storms or floods. This allows a much quicker and more efficient response by the emergency services.
- Intrusion detection: By placing sensors at various strategic points, one can thus prevent burglaries or the passage of game on a railway track (for example) without having to resort to expensive video surveillance devices.
- Business applications: One could imagine having to store foodstuffs requiring a certain humidity and a certain temperature (min or max). In these applications, the network must be able to collect this different information and alert in real time if the critical thresholds are exceeded.
- Pollution control: Sensors could be dispersed over an industrial location to detect and control gas or chemical leaks. These applications would give the alert in record time and be able to follow the evolution of the disaster.
- Agriculture: Knots can be incorporated into the soil. We can then question the network of sensors on the state of the field (determine for example the driest areas in order to water them first). One can also imagine equipping herds of cattle with sensors to know at all times, their position which would avoid breeders from having to use sheepdogs.
- Medical surveillance: By implanting mini video sensors under the skin, one can receive real-time images of a part of the body without any surgery for approximately 24 hours. We can thus monitor the progression of a disease or the reconstruction of a muscle.
- Building control: Sensors can be included on the walls of the dams to calculate the pressure exerted in real time. It is therefore possible to regulate the water level if the limits are reached. One can also imagine including sensors between the sandbags forming a makeshift dike. Early detection of water infiltration can be used to strengthen the dam accordingly. This technique can also be used for other constructions such as bridges, railways, mountain roads, buildings and other engineering structures.