Image Processing Reference
In-Depth Information
TABLE .
Current Sensor Networks Companies List
Company Name
Headquarters Location
HTTP Address
Ambient Systems
The Netherlands
http://www.ambient-systems.net
CrossBow
San Jose, California
http://www.xbow.com
Dust Networks
Berkeley, California
http://dust-inc.com
Ember
Boston, Massachussets
http://www.ember.com
Millennial Net
Cambridge, Massachussets
http://www.millennialnet.com
SYS Technologies
San Diego, California
http://www.systechnologies.com
MeshNetics
Dresden, Germany
http://www.meshnetics.com
Sensicast
Needham, Massachussets
http://www.sensicast.com
Tendril Networks
Boulder, Colorado
http://www.tendrilnetworks.com
ArchRock
San Fracisco, California
http://www.archrock.com
EnOcean
Oberhaching, Germany
http://www.enocean.com
3.4 Design Challenges
When designing a wireless sensor network one faces on one hand the simplicity of the underlying
hardware,and,ontheotherhand,therequirementsthathavetobemet.Tosatisfythem,new
strategies and new sets of protocols have to be developed [-]. In the following, we will address
the main challenges that are present in the wireless sensor network field. The research directions
involved and the open questions that still need to be answered will be presented as well.
To begin with, a high-level description of the current goals for the sensor networks can be
synthesized as
•
Long life
—hesensornodeshouldbeableto“live”aslongaspossibleusingitsownbat-
teries. This constraint can be translated to a power consumption less than µW. The
condition arises from the assumption that the sensor nodes will be deployed in a harsh
environment where maintenance is either impossible or has a prohibitively high price.
It makes sense to maximize the battery lifetime (unless the sensor nodes use some form
of energy scavenging). he targeted lifetime of a node powered by two AA batteries is a
couple of years. his goal can be achieved only by applying a strict energy policy which
will make use of power-saving modes and dynamic voltage-scaling techniques.
•
Small size
—he size of the device should be less than mm
.hisconstraingavethe
sensor nodes the name of “smart dust,” a name which gives a very intuitive idea about
the final design. Recently, the processor and the radio were integrated in a chip having a
size of approximately mm
. What is left is the antenna, the sensors themselves, and the
battery. Advances are required in each of these three fields to be able to meet this design
constraint.
•
Inexpensive
—he third high-level design constraint is about the price of these devices.
To encourage large scale deployment, this technology must be really cheap, meaning that
the targeted prices in the range of a couple of cents.
3.4.1 Locally Available Resources
Wireless sensor networks may consist of thousands of devices working together. Their small size
comes also with the disadvantage of very limited resources availability (limited processing power,
low-rate unreliable wireless communication, small memory footprint, and low energy). This raises
the issue of designing a new set of protocols across the whole system.
Energy is of special importance and can by far be considered the most important design constraint.
The sensor nodes are mainly powered by batteries. In most of the scenarios, due to the environment
where they are deployed, it is impossible to have a human change their batteries. In some designs,
energy scavenging techniques can also be employed. Still, the amount of energy available to the nodes
can be considered limited and this is why the nodes have to employ energy-efficient algorithms to
maximize their lifetime.
