Information Technology Reference
In-Depth Information
only on the link quality indicator (LQI). The au-
thors compared the performance of CTP using two
different frequencies: it is shown that CTP using
the frequency channel with less interference - and
hence higher LQI - provides better performance
than that using the frequency channel with more
interference. This shows that a solution that can
select a frequency channel with higher LQI is
preferable. Similar to CodeBlue, the prioritization
of different vital signs is not considered, which
may not satisfy the QoS requirements of different
types of data.
follow a common channel hopping sequence for
communications.
Unlike RICH-DP, Multi-channel MAC
(MMAC) (So & Vaidya, 2004) uses a default
channel for traffic indication and incorporates an
energy-saving mechanism. The multi-channel hid-
den terminal problem is solved by using temporal
synchronization. Time is divided into fixed-time
intervals using beacons and a small window at
the start of each interval to indicate traffic and
negotiate channels.
Like RICH-DP, Slotted Seeded Channel Hop-
ping (SSCH) protocol (Bahl, Chandra, & Dunagan,
2004) employs a channel-hopping scheme. But,
unlike in RICH-DP, channel hopping is not only
used for control but also for data transmissions.
Scheduling packets are employed to arrange
hopping schedule so that communications do not
interfere with each other, while synchronization
techniques are employed to assign traffic to dif-
ferent channels.
Multi-channel MAC (McMAC) (So, Walrand,
& Mo, 2007) is proposed to avoid control chan-
nel congestion so that it can use a large number
of channels efficiently. Each node-specific MAC
address is used as a seed to generate hopping
sequence. Consequently, the receiver's hopping
pattern can be predicted and different node pairs
are able to rendezvous at the same time on mul-
tiple channels and communicate with each other.
Compared to SSCH, the hopping pattern is chosen
at random and careful pairwise scheduling is not
needed. Also, unlike in SSCH, network-wide
synchronization is not required.
Most of the above MAC protocol solutions are
designed to avoid contention or collision without
considering channel fading. Although Opportu-
nistic Multiradio MAC (OMMAC) (Chen, Zhai,
& Fang, 2008) accounts for channel fading, it is
evaluated only by simulations and it is not clear
how well it would perform in a real environment.
These protocols do not take the actual channel
quality measurements into account and, hence,
cannot avoid using channels that may be experi-
Multi-Channel MAC Protocols
Multi-channel MAC protocols, several of which
are summarized and compared in (Wang, Abol-
hasan, Safaei, & Franklin, 2007; Wang, Zhou, &
Qin, 2008; Mo, So, & Walrand, 2008), have been
proposed to increase throughput and to reduce
signal interference. Depending on the number
of transceivers in use, the multi-channel MAC
protocols are divided into two classes: those with
only one transceiver and those with multiple (≥
2) transceivers. Performance of the second class
is generally better than that of the first class due
to the ability of simultaneously receiving and
transmitting packets, and the ability of receiv-
ing multiple packets; however, this is achieved
at the price of higher hardware complexity and
cost. Conversely, our solution is designed to be
practical and to run smoothly on existing low-cost
ZigBee devices, which have only one transceiver.
Therefore, our MAC approach falls in the first
class; for this reason, in the following we focus
on multi-channel MAC protocols using only one
transceiver.
Receiver-Initiated Channel-Hopping with
Dual Polling (RICH-DP) (Tzamaloukas &
Garcia-Luna-Aceves, 2001) is a receiver-initiated
collision-avoidance protocol that does not require
carrier sensing or the unique code assignment for
collision-free reception. All nodes in a network
Search WWH ::
Custom Search