Showing posts with label communication. Show all posts
Showing posts with label communication. Show all posts
Friday, September 8, 2017
Summary of “Broadcast Communication in Vehicular Ad Hoc Network Safety Applications
Summary of “Broadcast Communication in Vehicular Ad Hoc Network Safety Applications
IEEE 802.11 is an standard for wireless LANs that specifies the characteristics of physical and MAC layer. Availability and capability of IEEE 802.11 makes it the first choice for car-to-car communication in vehicular ad hoc networks (VANETs). Also, the result of different studies shows that contention-based solutions like IEEE 802.11 DCF outperforms contention-free solutions [no reference yet].
In IEEE 802.11 standard, RTS/CTS handshaking eliminates the collision but this mechanism can not be used in broadcast communication because in this case there are multiple nodes would need to answer to the same RTS message that results in collision. Moreover, the same problem appears in the case of ACK message. Consequently, not only it can not eliminate the hidden nodes, but collision detection is also impossible. Therefore, it results in low reliability in broadcast communication comparison to unicast communication.
In mobile ad hoc networks (MANETs), research community mainly focused on multi-hop communication and not one-hop broadcast messages. But oliveira et al. [1] Shows that DCF performances significantly drops when broadcast traffic occupies more than 50% of the total traffic. This situation can not be found in MANETs but safety applications in VANETs have the same characteristic and also they demand high level of reliability.
In the case that 802.11 does not have capability of collision detection, contention window (CW) will never be increased for broadcast messages and always uses the minimum CW (CWmin). This increases the importance of having a dynamic CWmin based on the network density (number of contending neighbors).
There are lots of research in ad hoc networks that they tried to estimate the number of contending nodes but it is very difficult to obtain a good estimation. Also, there are exist another approach that is trying to calculate CW based on the number of idle and busy (due to collision) slots. However, all of these studies assume that IEEE 802.11 can detect collision with ACK messages that is not valid in broadcast communication.
If CWmin (or CW because CWmax does not used in this case) will be selected very small it causes high collision probability because it increases probability of that multiple nodes select the same time slot to transmit. Also, if it will be selected very large, it causes high beacon drop probability because each beacon has limited time to be received in destination and selecting a large CW increases the total delivery delay (it must be received before arriving next beacon). In consequence, CW must be calculated precisely based on the number of nodes in neighborhood.
References:
- R. Oliveira, L. Bernardo, P. Pinto The Influence of Broadcast Traffic on IEEE 802.11 DCF Networks Computer Communications, vol. 32, no. 2, pp. 439-452, February 2009
download file now
Labels:
“broadcast,
ad,
applications,
communication,
hoc,
in,
network,
of,
safety,
summary,
vehicular
Subscribe to:
Posts (Atom)