Algorithm for low energy consumption via WAN.
In order to connect smart units with limited energy resources
via TCP/IP, Neava has developed an algorithm for the efficient
control of the power in wireless LANs (IEEE 802.4.15) and a minimal
use of system resources without creating adverse effects.
Adaptive Quality Control in IEEE 802.15.4
The last decade has seen an explosive growth in the domain for
smart objects. Smart objects are small computers, often equipped
with a sensor and a wireless communication device. Today smart
objects are embedded in everything from everyday objects like car
keys to industrial machinery.
A vision for the future is to connect these smart objects into an
'Internet of things' where they can exchange information with each
other and make better decisions based on that information. This
would open up a whole new dimension of applications for smart
objects.
Smart objects usually have limited resources when it comes to
computing power, memory and energy. The IP protocol has for a long
time been deemed unsuitable for these reasons, but has in the last
years been proved to work on smart objects as well. This is good
news, because IP has shown to be a scalable, robust and long-lived
communication protocol that supports a wide range of applications,
devices and underlying protocols.
There are however obstacles in realizing the 'Internet of
things' and one of these obstacles is conserving energy. Since
smart objects usually have a limited power supply, e.g. batteries,
and communicate wirelessly, communication has to be as energy
efficient as possible. One way to accomplish this is to control a
smart object's transmission power dynamically and optimally. Not
only does this save energy on transmissions, but can avoid costly
retransmissions caused by interference and reduce other
interference related problems like overhearing.
In this Master Thesis we develop an adaptive transmission power
control algorithm for TCP/IP and the underlying low-rate protocol
standard IEEE 802.15.4. Using a closed-loop control system based on
extended TCP acknowledgments called Quality Acknowledgments (QACKs)
we show that it is possible to control transmission power
effectively using only minimal system resources, overhead and
without causing ill effects.