Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Save to My Library
Look up keyword
Like this
7Activity
0 of .
Results for:
No results containing your search query
P. 1
Maca

Maca

Ratings: (0)|Views: 985 |Likes:
Published by ipas

More info:

Published by: ipas on Nov 10, 2008
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

04/20/2013

pdf

text

original

 
1
MACA - A New Channel Access Method for Packet Radio
Phil Karn, KA9Q ABSTRACT 
naThe existing Carrier Sense Multiple Access (CSMA) method widely used imateur packet radio on shared simplex packet radio channels frequently suffers from-lthe well-known "hidden terminal problem" and the less well known but related probem of the "exposed terminal." This paper proposes a new scheme, Multiple Accesscwith Collision Avoidance (MACA), that could greatly relieve these problems. MACAan also be easily extended to provide automatic transmitter power control. This couldincrease the carrying capacity of a channel substantially.
n1. Introductio
In the classic hidden terminal situation,astation Y can hear both stations X and Z, but Xnd Z cannot hear each other. X and Z areotherefore unable to avoid colliding with eachther at Y. (See figure 1.)In the exposed terminal case (figure 2), a.Ewell-sited station X can hear far away station Yven though X is too far from Y to interfereldwith its traffic to other nearby stations, X wilefer to it unnecessarily, thus wasting an oppor-ttunity to reuse the channel locally. Sometimeshere can be so much traffic in the remote areasithat the well-sited station seldom transmits. This a common problem with hilltop digipeaters.saThis paper suggests a new channel acceslgorithm for amateur packet radio use that caneAminimize both problems. This method, Multiplccess with Collision Avoidance (MACA), wasAinspired by the CSMA/CA method (used by thepple Localtalk network for somewhat differentsreasons) and by the "prioritized ACK" schemeuggested by Eric Gustafson, N7CL, for AX.25.aIt is not only an elegant solution to the hiddennd exposed terminal problems, but with thetnecessary hardware support it can be extendedo do automatic per-packet transmitter power"control. This could substantially increase thecarrying capacity" of a simplex packet radioa
  
channel used for local communications in
  
MACA is an acronym, not a Spanish word.
1
densely populated area, thus satisfying both thernFCC mandate to use "the minimum poweecessary to carry out the desired communica-ations" (Part 97.313) and to "contribute to thedvancement of the radio art" (Part 97.1 (b)).
2. How CSMA/CA Works
CSMA/CA as used by Localtalk works asafollows. When a station wants to send data tonother, it first sends a short Request To Sendrr(RTS) packet to the destination. The receiveesponds with a Clear to Send (CTS) packet.qOn receipt of the CTS, the sender sends itsueued data packet(s). If the sender does notRreceive a CTS after a timeout, it resends itsTS and waits a little longer for a reply. This)ithree-step process (not counting retransmissionss called a
dialogue
. Since a dialogue involves-transmissions by both stations, I will avoid conusion by referring to the station that sends theesRTS and data packets as the
initiator 
, and thtation that sends the CTS as the
responder 
.aThe RTS packet tells a responder that datollows. This gives the responder a chance toyeprepare, e.g., by allocating buffer space or bntering a "spin loop" on a programmed-I/Ouinterface. This is the main reason Localtalses the CSMA/CA dialogue. The Zilog 8530nbHDLC chip used in the Apple Macintosh cauffer the 3-byte Localtalk RTS packet in itstnFIFO, but without a DMA path to memory ieeds the CPU to transfer data to memory as itarrives. The CPU responds to the arrival of an
 
tRTS packet by returning a CTS and entering a- 2 -ight read loop, waiting for the data to arrive.
2
n(A timeout prevents a system lockup if the dataever arrives.)As is standard for CSMA schemes,nCSMA/CA requires stations to stay off the chan-el when another transmission is already in pro-togress. CSMA/CA also requires any station thaverhears an RTS or CTS packet directed else-twhere to inhibit its transmitter for a specifiedime. This helps reduce the probability of a col-.Tlision with a subsequent CTS or data packethis is the CA or
Collision Avoidance
part of rpCSMA/CA. However, collisions are not a majoroblem on Localtalk; the network is physicallyarsmall, carrier sensing is fairly rapid, the datate is relatively low, and (if the network is.Pproperly built) there are no hidden terminalslain CSMA would work well, but there was lit-Rtle extra cost to the CA feature (given that theTS/CTS dialogue was already needed for other
3
reasons) so it was included.
. Turning CSMA/CA into MACA
nsHidden and exposed terminals abound oimplex packet radio channels, and this makestothem very different from Localtalk and mosther types of local area networks. When hid-taden terminals exist, lack of carrier doesn’lways mean it’s OK to transmit. Conversely,rwhen exposed terminals exist, presence of car-ier doesn’t always mean that it’s bad totltransmit. In other words, the data carrier detecine from your modem is often useless. So I’llomake a radical proposal: let’s ignore DCD! Inther words, lets get rid of the CS incCSMA/CA. (It’s too hard to build good DCDircuits anyway...)Instead we’ll extend the CA part of whatwe’ll call MA/CA (or just plain MACA). Theey to collision avoidance is the effect that RTSntand CTS packets have on the other stations ohe channel. When a station overhears an RTSntaddressed to another station, it inhibits its owransmitter long enough for the addressed stationn
  
to respond with a CTS. When a statio
  
It would be nice if we could use this feature onp
2
acket radio with our programmed-I/O HDLC interfaces,i(e.g., DRSI PCPA, Paccomm PC-100). Unfortunatelyf our RTS/CTS packets carry full source and destina-Ftion call signs, they would not fit into the 3-byte 8530IFOs. So high speed operation will still require eitherDMA or a dedicated I/O processor.
overhears a CTS addressed to another station, itoinhibits its own transmitter long enough for thether station to send its data. The transmitter isshinhibited for the proper time even if nothing ieard in response to an RTS or CTS packet.ZcFigure 3 shows an example. Stationannot hear X’s transmissions to Y, but it
can
aChear Y’s CTS packets to X. If Z overhearsTS packet from Y to X, it will know not toXtransmit until after Y has received its data from.But how does Z know how long to waithafter overhearing Y’s CTS? That’s easy. Weave X, the initiator of the dialogue, include inosits RTS packet the amount of data it plans tend, and we have Y, the responder, echo thatoinformation in its CTS packet. Now everyoneverhearing Y’s CTS knows just how long totmwait to avoid clobbering a data packet that iight not even hear.As long as the link between each pair of sstations in the network is reciprocal (i.e., all thetations have comparable transmitter powers andpreceiver noise levels), the receipt of a CTSacket by a station not party to a dialogue tellsiit that if it were to transmit, it would probablynterfere with the reception of data by thesiresponder (the sender of the CTS). MACA thunhibits transmission when ordinary CSMAsrwould permit it (and allow a collision), thuelieving the hidden terminal problem. (Colli-tlsions are not
totally
avoided; more on this poinater.)Conversely, if a station hears no responsettto an overheard RTS, then it may assume thahe intended recipient of the RTS is either downe4or out of range. An example is shown in figur. Station X is within range of Y, but not Z.pWhen Y sends traffic to Z, X will hear Y’s RTSackets but not Z’s CTS responses. X mayitherefore transmit on the channel without fear onterfering with Y’s data transmissions to Z,Meven though it can hear them. In this caseACA allows a transmission to proceed when,tordinary CSMA would prevent it unnecessarilyhus relieving the exposed terminal problem.a(Because modems have a capture effect, hearingCTS doesn’t
always
mean that you’d cause atycollision if you transmit, so the problem isnet completely solved. More on this point later.)
 
4. Metaphors for MACA
- 3 -MACA is not really a novel idea; itt jmerely formalizes a procedure many people (noust radio amateurs) instinctively use in personalsconversation. A typical cocktail party has manyimultaneous conversations. The average guesthseldom waits for total silence in the room beforee speaks, but if someone asks him to pauseewbecause he is trying to hear someone else, hill usually do so. The MACA RTS packet ispanalogous to Bob saying "Hey, Tom!" and CTSacket is analogous to Tom responding withpt"Yeah, Bob?". This causes most people to stoalking if they are close to Tom (except, of -pcourse, for Bob). The same thing (should) hapen in manual amateur radio operation when-oever a station finishes a transmission with "gonly" (or "KN" on CW or RTTY).soThe Prioritized ACK scheme also involveverheard packets that inhibit other stations for-bspecified periods of time. In this case, the inhiiting packet is a data packet and the protectedynstation is sending an acknowledgement that maot be audible at the inhibited stations. Full(protection against collisions is not provideddata packets can still collide) but the perfor-rmance improvement due to the lower ACK lossate is reported to be substantial.naMore formally, MACA can also be sees a single-channel, time-multiplexed form of BBusy Tone Multiple Access (BTMA). InTMA, receivers transmit "busy tones" oneasecondary channels whenever their receivers arctive. This warns the other stations in rangernthat they should not transmit even if they heaothing on the data channel. On the other hand,tstations not hearing busy tones are free toransmit even if there is already a signal on theadata channel. Indeed, stations need not pay anyttention at all to the data channel when decid-.Aing to transmit; only the busy channel matterss long as the propagation characteristics aretidentical between the main and secondary (busyone) channels, BTMA is effective. Unfor--qtunately, the need to use widely separated freuencies to avoid self-interference tends to.Bmake the link characteristics less symmetricalTMA also obviously increases the hardwareucomplexity and spectrum requirements of eachser station. On the other hand, because MACAdduses the same channel for the "busy tone" anata, paths between pairs of stations are muchmore likely to be symmetrical.
5. Collisions in MACA
Unlike BTMA, however, collisions.Tbetween RTS packets can still occur in MACAhese are minimized with a randomizedtuexponential back-off strategy similar to thased in regular CSMA. Since there is no carrierrsensing in MACA, each station simply adds aandom amount to the minimum interval eachnRstation is required to wait after overhearing aTS or CTS packet. As in regular CSMA, this-tstrategy minimizes the chance that several staions will all jump on the channel at the instantlwit becomes free. The extra random intervaould be an integral multiple of the "slot time",Rand in MACA the slot time is the duration of anTS packet. If two RTS packets collidecnonetheless, each station waits a randomlyhosen interval and tries again, doubling the.Eaverage interval on each successive attemptventually one of them will "win" (i.e., transmitttfirst) and the CTS from its responder will inhibihe "losing" station until the winning station cancomplete its dialogue.Even though collisions can occur betweeneoRTS packets, MACA still has the advantagver CSMA as long as the RTS packets areslsignificantly smaller than the data packets. Aong as this is true, collisions between RTS-spackets are much less "costly" than the colliions that would otherwise occur between dataspackets. The savings in collision time also payor the overhead of the RTS and CTS packets.ApAs mentioned earlier, the basic MACrotocol only reduces the chances of collisionstinvolving data packets; it does not guaranteehat they will never occur. This is because a-tCTS packet requires a certain minimum signalo-noise ratio at a station for it to be understoodlmand obeyed. Even if the station powers are welatched, a pair of stations might have justienough of a path between them to allow them tonterfere with each other’s reception of weatsignals, but not enough of a path to allow themo hear each other’s CTS packets. Although thesaseriousness of this problem is unknown, it doeppear that the power-controlled version of 
6
MACA discussed later would greatly reduce it.
. Bypassing the MACA Dialogue
etIf the data packets are of comparable sizo the RTS packets, the overhead of thescRTS/CTS dialogue may be excessive. In thiase, a station may choose to bypass the normal

Activity (7)

You've already reviewed this. Edit your review.
1 hundred reads
1 thousand reads
Anomer Delta liked this
Mitsuki Kumiko liked this
spinaz liked this
agopad liked this
sumittokle liked this

You're Reading a Free Preview

Download
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->