re-submit

Paper Review: MACAW: A Media Access Protocol for Wireless LAN’s

The paper introduces a new media access protocol, MACAW, which is a
modified variation of the MACA protocol for use in wireless LAN’s.
It begins by stating that the simulation results only apply to the
particular radio technology in their lab. However, they hope to
explore basic performance and design issues inherent in wireless media
access protocols.

The paper begins by detailing the lab setup as well as a list of
simplifying assumptions.
The reader must be careful not to interpret the results of the paper
incorrectly. These results are not intended to show that MACAW can be
applied universally to wireless networks. The authors have been
careful to state that the results apply only to their lab. However,
the paper does introduce a host of issues inherent in wireless media
access protocols and presents several feasible solutions.

The MACAW protocol basically extends the MACA protocol with two basic
goals: high network utilization and fair access to the medium.
Changes, however, have been made to improve performance and fairness.
The backoff algorithm is modified to ensure fairness (equal
probability of winning contention for all nodes) and to prevent large
oscillations (no rapid change in backoff parameter value). The
algorithm is also modified to ensure that fairness is granted by
stream, rather than by station. An ACK has been included to speed up
recovery by handling it in the link-layer, rather than transport. A
DS control packet is included to give other stations a better idea of
when the next contention period will take place. An RRTS packet is
included to solve a specific situation leading to contention problems.
  It is used for stations that don’t receive CTS packets after sending
an RTS. The receiver sends a RRTS to contend for the channel on
behalf of the sender.

One weak point in the simulations used for this paper is that they
seem to be ‘fitted’ for each particular test case. At times 2
stations are used and for others as many as 6 are used. Even if it
may not be the case, the impression this gives is that the results may
not be true for all cases.

Aside from this, the paper does contribute some interesting new ideas.
  They show that it is indeed possible to solve some of the problems,
such as the hidden station, that plague wireless media access
protocols without excessive overhead or significantly affecting
throughput. Since the paper sets out with modest goals, they seem to
provide enough evidence to support their claims. They address several
issues and their explanations are clear and concise. The paper is
organized well and was relatively easy to follow. The use of diagrams
added to the quality of the paper. Overall it is well-done.

-- Nadeem Abji
Received on Thu Sep 21 2006 - 09:50:24 EDT