Image Processing Reference
In-Depth Information
balance energy consumption. However, this protocol assumes negligible energy consumption in the
idle state, and every node should always be active. hus, if idle power consumption is high, energy
dissipation in the idle state will dominate the total energy consumption. However, the protocol does
nothaveparticularrequirementsfortheMAClayer,sotheauthorssuggesttheuseofalow-power
MAC that puts itself to sleep when no activity is required.
7.7 Energy-Aware QoS-Enabled Routing Protocols
Energy-aware QoS-enabled routing protocols address the inherent conflict between the limited
resourcesofeachsensornode,particularlyintermsofenergy,andtheneedtoachievethedesired
QoS such as end-to-end real-time performance. In the following parts of this section some examples
of EAR protocols able to provide some QoS for real-time data are described.
7.7.1 Energy-Aware QoS Routing Protocol for Wireless Sensor Networks
In [Akk] an energy-aware QoS routing protocol for WSNs which can efficiently handle best-effort
traffic is presented. The protocol finds QoS paths for real-time data with given end-to-end delay
requirements. An analytic estimation of the end-to-end delay of a path is performed based on queuing
theory. he protocol uses the estimated delays in order to select the delay-constrained path with the
least cost. Here the cost of a link is a function of the distance between source and destination, energy,
expected time, and error rate. In this way it is possible to find a path that meets delay requirements
while optimizing energy and error rate. This protocol manages both real-time and non-real-time
traffic, trying to balance low delay for real-time traffic. with high throughput for non-real-time traffic.
Each node is assumed to have two different queues, for real-time and non-real-time traffic, and a
scheduler that determines the order in which data has to be sent. This order is based on a value
r
,
which represents the amount of bandwidth to be used for real-time traffic, while -
r
is the ratio for
non-real-time traffic. So an algorithm to calculate a proper
r
-value is proposed, which reduces the
complexity of the problem by assuming an equal network-wide
r
-value for each link, so that a simple
optimization problem can be formulated. he network-wide
r
-value guarantees a given service rate
for real-time and non-real-time data on each link.
Simulation results in [Akk] show that the protocol performs well with respect to both QoS
metrics (i.e., throughput and average delay) and an energy-based metric (i.e., the average lifetime of a
node). he results also show the impact of real-time data rate, buffer size, and packet drop probability
on the performance of the protocol.
7.7.2 Real-Time Power-Aware Routing (RPAR) Protocol
In[Chi]theRPARprotocolforWSNsispresented.hegoalofthisprotocolistoobtaingoodsot
real-time performance and minimize the deadline miss ratio, while also providing increased energy
efficiency.
RPAR is based on empirical assessments of the impact of transmission power on the delivery
velocity of each packet, defined as the total distance the packet travels divided by its end-to-end
delay.
In [Chi] it is shown that the impact of transmission power and one-hop distance on delivery
velocity is signiicant. his is because the increased transmission power improves the quality of the
wireless link, so the number of transmissions needed to deliver a packet is reduced. Moreover, the
measurements performed show that the delivery velocity increases as the one-hop distance increases
within a range, while it descends sharply when the one-hop distance exceeds the range due to link
quality degradation. A higher transmission power increases such a drop-off range and allows for
