RTS/CTS and contention window about open-plc-utils

[razzlefratz]

Christina,

Thanks for asking. There is no concept of RTS/CTS on the chipsets you mention. Neither one has a serial interface. Both have an MII interface. The QCA7000 has a serial interface that supports UART and can be set up to observer RTS/CTS protocol, if desired. Before we go too much further on this, are we sure about the chipset model numbers?

[christinavl]

Dear Charlie,

Thank you very much for the quick reply. As far as I know QCA7000 is HomePlug GreenPhy. I do not have such kind of devices. Basically, I am a researcher and I am studying the MAC protocol of high data-rate devices, so I would like to modify some parameters such as the contention window, but I know that the firmware is encrypted. I would also like to test the network with the RTS/CTS mechanism which is specified by the IEEE 1901 standard, but I guess it is not possible if this concept does not exist. Concerning the chipset model numbers, this is what the command int6k -a returns, so I guess they are correct. Thank you for your help.

[razzlefratz]

The QCA7000 is different from other Qualcomm PLC chipsets. It is a
serial-to-powerline bridge that supports either UART or SPI interfaces.
RTS/CTS is supported in UART mode. You are correct that the QCA7000 is a
Green PHY chipset but its is a low data rate device by design at 10mb. The
QCA7500, in contrast, is 1gb. I ill be happy to steer you in the right
direction if I understand what you want to do. Regards, Charlie Maier.

On Thu, Nov 20, 2014 at 7:37 AM, christinavl [email protected]
wrote:

Dear Charlie,

Thank you very much for the quick reply. As far as I know QCA7000 is
HomePlug GreenPhy. I do not have such kind of devices. Basically, I am a
researcher and I am studying the MAC protocol of high data-rate devices, so
I would like to modify some parameters such as the contention window, but I
know that the firmware is encrypted. I would also like to test the network
with the RTS/CTS mechanism which is specified by the IEEE 1901 standard,
but I guess it is not possible if this concept does not exist. Concerning
the chipset model numbers, this is what the command int6k -a returns, so I
guess they are correct. Thank you for your help.

—
Reply to this email directly or view it on GitHub
#39 (comment).

[christinavl]

I am studying the MAC layer of high data-rate PLC (HomePlug AV-AV2). I have managed to find some configurations (that is, some values for the contention windows and the deferral counters, which are parameters of the MAC layer) that boost the performance of the MAC. Thus, I would like to modify these parameters on the PLC chips to test some metrics such as throughput. I guess that the values for the contention windows and the deferral counters can be found and modified via the firmware. However, this firmware is encrypted, so no tests can be run with modified parameters. The only thing I could find is the .pib file, but I am not sure I could use this file to modify such parameters. It would be useful to know if these parameters can be modified somehow. Thanks and regards, Christina

[razzlefratz]

Hi Christina,

The sorts of parameters you are talking about cannot be modified via the
PIB. As you guessed, they are programmed into the firmware and the firmware
implements features that distinguish our powerline products from those of
our competitors. It is Qualcomm intellectual property and so I must keep
you guessing, but thanks for enquiring.

On Mon, Nov 24, 2014 at 7:39 AM, christinavl [email protected]
wrote:

I am studying the MAC layer of high data-rate PLC (HomePlug AV-AV2). I
have managed to find some configurations (that is, some values for the
contention windows and the deferral counters, which are parameters of the
MAC layer) that boost the performance of the MAC. Thus, I would like to
modify these parameters on the PLC chips to test some metrics such as
throughput. I guess that the values for the contention windows and the
deferral counters can be found and modified via the firmware. However, this
firmware is encrypted, so no tests can be run with modified parameters. The
only thing I could find is the .pib file, but I am not sure I could use
this file to modify such parameters. It would be useful to know if these
parameters can be modified somehow. Thanks and regards, Christina

—
Reply to this email directly or view it on GitHub
#39 (comment).

[razzlefratz]

Hi Christina,

I will be closing this issue soon, if that is OK with you.

[christinavl]

Yes, sure. Thanks.