Energy efficient
wireless communication network design is an important and challenging problem.
It is important because mobile units operate on batteries with energy supply.
It is challenging because there are many different issues that must be dealt
with when designing a low energy wireless communication system (such as
amplifier design, coding, and modulation design), and these issues are coupled
with one another.
Furthermore, the
design and operation of each component of a wireless communication system
present trade-offs between performance and energy consumption.
Therefore, the
challenge is to exploit the coupling among the various components of a wireless
communication system, and understand trade-offs between performance and energy
consumption in each individual component, in order to come up with an overall
integrated system design that has optimal performance and achieves low energy
(power).
The key observation is that constraining the energy of a node imposes
a coupling among the design layers that cannot be ignored in performing system
optimization. In addition, the coupling between layers requires simulation in
order to accurately determine the performance. The purpose of this power is to
present a methodology for the design, simulation and optimization of wireless
communication networks for maximum performance with an energy constraint.
Before we
proceed, we illustrate, through simple examples, a couple of issues that need
to be addressed. To highlight the trade-offs between performance and energy
consumption at individual components, consider the design and operation of an
amplifier. The amplifier boosts the power of the desired signal so that the
antenna can radiate sufficient power for reliable communications. However,
typical power amplifiers have maximum efficiency in converting DC power into RF
power when the amplifier is driven into saturation. In this region of
operation, the amplifier voltage transfer function is nonlinear.
Because of this
non linearity, the amplifier generates unwanted signals (so called
intermodulation products) in the band of the desired signal and in adjacent
bands. When the amplifier drive level is reduced significantly (large back off)
the amplifier voltage transfer characteristic becomes approximately linear. In
this case it does not generate intermodulation products. However, with large
back off the amplifier is not able to efficiently convert DC power into RF
power. Thus, there is considerable wasting of power at low drive levels, but at
high drive levels more interfering signal are generated.
To highlight the
coupling among the design of individual components of a wireless system,
consider packet routing in a wireless network that contain no base station
(i.e. an ad hoc network). For simplicity consider a network with nodes A, B and
C shown in figure. If Node A wants to transmit a message to Node C, it has two
options. Transmit with power sufficient to reach Node C in a single
transmission, or transmit first from A to B with smaller power, and then B to
C. since the received signal power typically decays with distance as d4, there
is significantly smaller power loss due to propagation in the second option
because d^4ac>d^4ab+d^4bc.however even though Node A transmits with smaller
output power, it does not necessarily proportionally decreases the amount of
actually consumed because of the amplifier's effect discussed above.
Furthermore,
besides the energy required for packet transmission, there are energy
requirements for packet reception and information decoding. The probability of
packet error reception that is achieved depends on energy allocated to the
receiver. Consequently, there is a coupling among amplifier design, coding and
modulation design, and decoding design as well as routing protocol.
0 Comments