Re: [Discuss-gnuradio] An Old Idea for GSoC 2018

Hi Sakthivel,

great to hear that you are interested in contributing to GNU Radio! I'm sure GNU Radio would benefit from a well-written gr-mimo module.

Practical difficulties are very likely to appear during the design and the implementation, how problematic that is for the feasibility of the project, however, mainly depends on you. :) Synchronization and channel estimation will be one issue you have to deal with, and if you use multiple USRPs for the trasnmitter or the receiver, you will have to deal with the inter-device synchronization, too. The latter could, e.g., be alleviated by simply using USRP B210s and restricting the system to 2x2 antennas as you wrote yourself.

At your stage, I think it's most important to get a clear idea of what you would like to implement. As this is not one of our ideas from the list, the acceptance of the project does not only depend on your proposal and the idea itself, we would also have to find a mentor for that and this is much easier with a clearly scoped proposal.

So make a plan with deliverables you think you can actually realize in the given time and come back to this list so we can discuss it further.

If anybody here on this list thinks he'd want to mentor such a project, feel free to chime in!

Cheers,
Felix

On 02/27/2018 02:53 PM, Sakthivel Velumani wrote:

Hi all,

I am a graduate student at Technische Universitat Darmstadt, Germany, majoring in signal processing for wireless communication. I am originally form India and did my bachelors there. I haven't heard about GNU Radio until a few months back and I regret for that, as I had more time in my undergrad and now graduate study is quite intense.

But still I managed to play with GR for few months now and also trying to build a block for carrier sync for OQPSK signals.

I saw MIMO transceiver idea in the GSoCOldIdeas page and I found it interesting and it suits me. I have done projects on solving problems in MIMO networks but the works were purely theoretical and simulation based. So I am eager to try out MIMO stuffs with hardware. As mentioned in the idea, I too feel that it would be nice for beginners to enjoy the effect of MIMO practically for a better understanding of the subject. So a standard gr-mimo module with a simple MRC decoder and a multi user beamforming encoder would do the job.

I further saw some discussion in the mailing list from 2014 regarding the same idea, and there were concerns (by Micheal Dickens) about the feasibility and practical difficulties in implementation as 3 months would not be sufficient. I am a bit confused as I think 3 months would be sufficient to implement the stuffs mentioned in the idea, or may be I am missing something here. I even saw Alick Zhoa and some others saying that they had already implemented 2x2 MIMO system in GR without OFDM but might have got obsolete. 

However it has been few years since then and I hope someone could give more insight on implementing a practical MIMO transceiver and difficulties in it.

So my concerns are

• I would be glad if someone could explain the possible difficulties in implementing a practical 2x2 MIMO system in GNU Radio and with USRPs
• Considering the difficulties, is it doable in a span of 3 months?
• If someone had already worked on similar stuffs, their help would be appreciated.
• Will the standard MIMO module (gr-mimo) be really beneficial to the project? If not, I would rather work on it at my leisure than for GSoC as the idea would be less likely to be accepted.

Regarding the hardware, my university laboratory has several USRPs and I can use them.

Best regards,

Sakthivel Velumani

_______________________________________________  Discuss-gnuradio mailing list  Discuss-gnuradio@gnu.org  https://lists.gnu.org/mailman/listinfo/discuss-gnuradio  

--   Karlsruhe Institute of Technology (KIT)  Communications Engineering Lab (CEL)    Felix Wunsch, M.Sc.  Research Associate    Kaiserstraße 12  Building 05.01  76131 Karlsruhe    Phone: +49 721 608-46276  Fax: +49 721 608-46071  E-Mail: Felix.Wunsch@kit.edu    www.cel.kit.edu    KIT -- University of the State of Baden-Württemberg and  National Laboratory of the Helmholtz Association