[ns] Still problem with mac 802.11

Alper Kamil Demir demir at kou.edu.tr
Wed Apr 20 10:35:30 PDT 2005


>SIFS is designed to allow the cards to process (i.e., to understand) the 
>previous packet (RTS, CTS or DATA) and change the mode from Rx to Tx: 
>both take a finite amount of time. 

This is how SIFS is calculated. In general, 4 types of IFS are used to provide access priorities (SIFS, PIFS, DIFS, EIFS). SIFS is used to seperate transmissions belonging to a single dialog i.e. Fragment-Ack, and there is always at most one single station to Tx at this given time, hence having priority over all stations. 

>RTS/CTS in single-hop networks reduces the time-waste should there be a 
>collision due to the finite propagation time of radio waves (+CCATime+
>Rx-Tx mode switch time).

It is exponential backoff that solves contention problem, hence reduces collisions. RTS/CTS and some other mechanisms not used in 802.11 improves this by solving specific problems mentioned below.

Alper K. Demir

----- Original Message -----
From: "Alper Kamil Demir" <demir at kou.edu.tr>
To: "glitch74" <glitch74 at libero.it>, "ns-users" <ns-users at ISI.EDU>
Subject: RE: [ns] Still problem with mac 802.11
Date: Wed, 20 Apr 2005 18:41:33 +0300

> 
> 
> IEEE 802.11 uses CSMA/CA algorithm to mediate the access to the 
> shared medium. data (DATA), access control (RTS, CTS), and 
> reliability control (ACK) are different types of frames used in 
> CSMA/CA, consequently in 802.11. RTS/CTS is one of access 
> mechanisms used in 802.11. There is also a basic access mechanism. 
> Mainly, access mechanisms are aimed to solve problems occuring 
> because of limited range of radio communications and phisically 
> distributed nature of radio nodes such as hidden terminal, exposed 
> terminal, etc... On the other hand, different types of IFS (inter 
> frame space) time control (such as DIFS, SIFS, etc..) are used to 
> provide a mechanism for priority access. Albeit 802.11 is not 
> intended for service differentation (802.11e is used for this), 
> however, all these mechanisms also provides a means of bandwith 
> utilization and simple quailty of service.
> RTS/CTS does not necessarily mean that we are reserving the shared 
> medium. However, it may be considered as a very simple impilicit 
> reservation mechanism if properly applied. You can both reserve and 
> share resources. They are not mutually exclusive. If it is very 
> implicit then you may need to contend. I am not sure which 
> contention window you are refering to. Hope this helps.
> 
> Alper K. Demir
> 
> 
> 
> 
> From: glitch74
> Sent: Wed 4/20/2005 4:59
> To: ns-users
> Subject: [ns] Still problem with mac 802.11
> 
> 
> Hi ns-users!
> I've still problem with mac 802.11 implemented in ns2.
> 
> I can't understand why mac 802.11 implementation use a rts/cts/data/ack
> pattern.
> In 802.11 packets would not need rts/cts/data/ack to be sent! Isn't it?
> If a packet waits a difs+backoff to be sent, why it is needed the pattern
> above?
> Using rts/cts means that we're doing reservation of the shared medium.
> But if we reserve it....it isn't shared!!! We not need to contend it, infact
> i've observed that the value of the contention window is always 31 (the lower
> value) ns2 never doubles it, so it never reaches the value 1032 (upper level)
> Thank's a lot.
> Please HEEEEEEEELP ME!!!!!!!!!!!!!!!! I'm going crazy!
> Bye
> 
> 
> 
> 
> ____________________________________________________________
> 6X velocizzare la tua navigazione a 56k? 6X Web Accelerator di Libero!
> Scaricalo su INTERNET GRATIS 6X http://www.libero.it

-- 
_______________________________________________
NEW! Lycos Dating Search. The only place to search multiple dating sites at once.
http://datingsearch.lycos.com


More information about the Ns-users mailing list