Paper Review : MACAW: A Media Access Protocol for Wireless LAN’s
Reviewer : Seh Leng Lim
The paper attempts to present a solution for solving an ongoing problem of controlling access to the shared media of a wireless network. This problem is very important because of the proliferation of mobile computing devices such as portables, palmtops and personal digital assistants which are all likely to attach to the network via wireless connectivity.
One common wireless multiple access algorithm is the CSMA protocol whereby each station senses for carrier before transmitting. However, this fails in the hidden terminal scenario where collisions occur in the receiver instead of the transmitter.
Another alternative to the CSMA for packet radio is the MACA protocol which can overcome problems associated with the hidden terminal scenario.
The main contribution of the paper is its improvements to the original MACA algorithm for use in wireless LAN. The enhanced algorithm is aptly termed “MACAW”.
The 3 key main ideas are :
(a) an enhanced backoff algorithm to ensure fairness in bandwidth allocation and network efficiency
(b) the RTS-CTS-DS-DATA-ACK message exchange protocol
(c) the RRTS message
The original backoff algorithm in MACA does not produce adequate fairness in simple one-cell configuration. The MACAW backoff algorithm makes use of a backoff copying scheme to ensure fairness in allocation as well as a gentler backoff adjustment scheme to achieve efficiency. Since backoff counter is an indication of the level of congestion in the vicinity of a specific receiver node, MACAW proposes maintaining backoff counter independently for each receiver. Using per-receiver backoff counter is particularly useful in multi-hop environments, since congestion level at different receivers can be very different.
As opposed to using TCP, recovery at the link-layer is much faster. Therefore, the basic RTS-CTS-DATA exchange protocol in MACA is amended in MACAW to include an acknowledgement packet, ACK, that is returned from the receiver to the sender immediately upon completion of data reception. The original backoff algorithm in MACA also does not allow an exposed terminal C to transmit even though it is in range of B which is sending to A (which is out of its range). MACAW overcomes this by requiring stations to transmit a short synchronizing Data-Sending (DS) packet before the DATA packet. All this results in the RTS-CTS-DS-DATA-ACK message exchange protocol which is implemented in MACAW.
In addition MACAW provides the Request-for-Request-to-Send (RRTS) packet to overcome the unfair allocation problems in the two cell configuration example in Figure 6 where two pads are in range of their respective base stations and also in range of each other, and the base stations are sending data to their respective pads.
I think that the paper has some signification contribution to the media access control problem of the wireless network. I will give the paper a rating of 4. This is because it has convincing but limited experimental data to support the claims that MACAW does in fact help to ensure fairness in bandwidth allocation as well as achieve network efficiency. I see that the work of this paper still has to be further enhanced as the authors rightly pointed out. There are still problems unsolved by the MACAW solution such as how to use it in a multi-cast setting, as well as the two cell configuration problem in Figure 7 where base station B1 is sending data to pad P1, and pad P2 is sending data to base station B2.
From this paper, system researchers and builders can better appreciate the problems and limitations associated with reliable data transfer in a wireless LAN.