wilamowski-b-m-irwin-j-d-industrial-communication-systems-2011

48-4 Industrial Communication Systems
Legacy support for
contention-free
services, optional
Support for
prioritized QoS
HCF
Support for
parameterized QoS
MAC
PCF
EDCA
HCCA
Support for
contention services
DCF
FIGURE 48.1
MAC architecture with HCF.
The DCF is totally distributed and can be used within both ad hoc and infrastructure network configurations.
Conversely, the PCF is centralized and can be used only in infrastructure network configurations.
In the DCF, each station senses the shared channel and transmits when it finds a free channel
according to a carrier-sense mechanism. In the PCF, a coordinator station polls the other nodes, thus
enabling them to transmit in a collision-free way. The Enhanced Distributed Channel Access (EDCA)
mechanism is based on DCF medium access, except for the ability to assign different priorities to various
kinds of traffic flows. HCF-Controlled Channel Access (HCCA) defines a parameterized QoS support.
It also relies on a polling procedure with an underlying time-division multiple access (TDMA)
principle but is more flexible when compared to the legacy of the PCF.
Currently, the most widely used channel access mechanisms are the DCF and the EDCA, while the
PCF and the HCCA have received little attention so far, especially where commercial products are concerned,
because the available chipsets still lack support for these mechanisms. In the following, the four
different access methods are outlined.
48.4.1 Distributed Coordination Function
The IEEE 802.11 DCF operating mode is based on Carrier-Sense Multiple Access with a Collision
Avoidance (CSMA/CA) protocol and on a random backoff time following a busy medium condition. The
random backoff time is intended to reduce the collision probability between multiple stations accessing
a shared medium at the point where collisions would most likely occur. The highest probability of a
collision exists just after the medium becomes idle following a busy medium because multiple stations
could have been waiting for the medium to become available again. This is the situation in which the
random backoff procedure comes into action to resolve medium contention conflicts.
Before starting transmission, a node listens to the channel for a time called, a Distributed Interframe
Space (DIFS), to assess whether the channel is idle or not. If the channel is idle, each node generates
a random backoff interval in order to reduce the probability of collisions with other nodes trying to
access the sensed idle channel at the same time. Each node decreases its backoff counter as long as the
wireless channel is sensed to be idle. If the counter has not reached zero, and the channel becomes
busy again, the backoff counter is frozen and reloaded as soon as the channel is sensed to be idle again
for a DIFS.
When the backoff interval is over, and if the channel is still idle, the transmission starts. The random
backoff interval, expressed as a number of time slots, is generated in the set {0,CW – 1}, where CW denotes
the contention window size. The initial value of the contention window is CWmin. In the case of an unsuccessful
transmission (due to collisions or losses), the CW is doubled up to a maximum value CWmax.
© 2011 by Taylor and Francis Group, LLC