WARP Project Forums - Wireless Open-Access Research Platform

You are not logged in.

#1 2015-Jul-28 04:21:45

crimechb
Member
Registered: 2010-Sep-01
Posts: 205

About No Mac part of 802.11 Ref design

Dear Sir,

When we modify low-level TokenMAC  to no mac and test the throughput.
There is no result.

We have some questions below:
1. When we using the NO Mac which it will send the Beacon Frame or not,or it will send other control packet?
2. T_SLOT, T_SIFS, Does them has any function?
3. T_PHY_RX_START_DLY, TX_PHY_DLY_100NSEC, How to calculate them?

Last edited by crimechb (2015-Jul-28 04:29:46)

Offline

 

#2 2015-Jul-28 08:38:03

welsh
Administrator
From: Mango Communications
Registered: 2013-May-15
Posts: 612

Re: About No Mac part of 802.11 Ref design

Could you please elaborate on your test setup and the results you are seeing?  The NoMac project is part of the current reference design release (v1.3) and should not require any modification to use.  I do not understand how you would modify TokenMac to NoMac when TokenMac was never part of a software release. 

1. When we using the NO Mac which it will send the Beacon Frame or not,or it will send other control packet?

If you look at the 802.11 Reference design architecture, you can see that you can independently change the behavior of CPU-Low without affecting the behavior of CPU-High.  This allows us to change low-level MAC behavior without affecting how high-level MAC behavior.  If you are using the AP high-level MAC behavior with the NoMac low-level MAC behavior, the node will behave just like an AP, but will have none of the DCF behavior (i.e. carrier sensing, retransmissions, etc) that the DCF low-level MAC provides.  Therefore, you will still see control packets such as Beacon Frames when you use NoMac, just like you would with DCFMac.

2. T_SLOT, T_SIFS, Does them has any function?

If you search through the code (the SDK "Search --> File ..." works really well), you can see the definition of defined values like T_SLOT.  In general, there are comments about what these values are.  The values for T_SLOT and T_SIFS are defined by the 802.11 standard and are used accordingly.   

3. T_PHY_RX_START_DLY, TX_PHY_DLY_100NSEC, How to calculate them?

As you can see from the comments, values like TX_PHY_DLY_100NSEC were empirically measured since they are dependent on my hardware components in the RX signal chain and are difficult to precisely calculate from all the data sheets.  Looking through the code, you can see how they are used to adjust timing so that the behavior of the node conforms to the 802.11 standard.

Offline

 

Board footer