[Ns-developers] Adding Wifi Phy models

[Nicola Baldo]

Hi,

sorry for the late reply. Some comments below...

Mathieu Lacage wrote:
> On Sun, 2009-03-22 at 17:14 +0100, Nicola Baldo wrote:
> 
>> Such a hardware is for instance Calradio:
>> http://calradio.calit2.net/calradio1a.htm
>>
>> For those who are not familiar with it, Calradio is an experimental 
>> platform with a 802.11b PHY and a fully customizable MAC layer written 
>> in C code which runs on a DSP. The PHY in particular is based on two 
>> components: the baseband module (the Intersil HFA3863 chipset I 
>> mentioned earlier) and the RF module (based on the Maxim MAX2820 chipset).
> 
> Not that I don't care about this kind of hardware but, really, the goal
> of the ns-3 PHY module is to modelize real hardware :) (more on this
> below)
> 
>> In Calradio, the setting of the PHY mode belongs to the baseband module, 
>> whereas the setting of the frequency to be used belongs to the RF 
>> module. Both these settings are performed by writing to appropriate 
>> configuration registers when desired. As a consequence, per-packet 
>> frequency selection can be implemented by setting a new frequency just 
>> prior to transmission.
> 
> I would be curious to know how much time it takes to make the RF
> baseband actually perform the frequency switch. I suspect that this
> delay is sufficiently long that frequency switches on a per-packet basis
> are not practical and I suspect that this is also the case in real
> hardware and I also suspect that the cost of building hardware which can
> do this for 802.11 is sufficiently high that it would prohibit any real
> deployment.
> 
> But, well, the real issue here is that the goal of the current codebase
> is to modelize real 802.11 systems where per-packet frequency switches
> are simply not feasible so, I would tend to happily ignore this
> possibility for now.
> 


The switch latency of Calradio is limited by the channel switch latency
of the RF chipset (average 150 us, maximum 200 us, according to the
MAX2820 datasheet). The additional latencies associated with having the 
MAC requesting the channel switch operation are negligible (less than 4
microseconds). I think that implementing a per-packet channel switch
strategy with these latencies is feasible, even if of course a faster
switching would be desirable. Actually, the MAX2820 chipset is several
years old, so I wonder if faster switching RF chipsets have become
available in the market in the meanwhile.

Overall, Calradio is not using any exotic and expensive hardware.
Basically it is using the same PHY hardware as the old PRISM-based wifi
devices. The difference is that the MAC is implemented as a fully
customizable C code runnig on a DSP, rather than being implemented as a
proprietary firmware. I would say that any solution which can be
implemented on Calradio could be easily manufactured and sold for a
price similar to that of currently available commercial devices.


Regards,

Nicola