Re: [Idnet] Intelligence-Defined Network Architecture and Call for Interests

[김민석 <[e-mail]>]

Hi Brian,


As you mentioned by the prior email, anticipating network DDos attacks is really trendy issue to solve by ML techniques.

We also make some efforts how to avoid fagile nodes by a trustworthy communication, that means quantifying trustworthiness of node with normalization of various requirements such as security function, bandwidth and etc.

We are freshly approaching in routing layer with confidence using our own requirements, TPD(Trust Policy Distribution) and TD(Trust Degree). These requirements are considered to be solved by Reinforcement Learning (RL) that is one of the ML algorithms. RL is useful to control some of network policy about specific actions and states with reinforced and purnished rewards (+/-), but the problem is too slow to acquire satisified performance. Other ways to say it, anormaly dectection and regression analysis might be both efficient approaching methods to solve the issues Dave mentioned.


Best Regards,



Min-Suk Kim

Senior Researcher / Ph.D.
Intelligent IoE Network Research Section,
Electronics and Telecommunications Research Institute (ETRI)
e-mail          :  [e-mail]<mailto:[e-mail]>
http://www.etri.re.kr/











________________________________
보낸 사람 : "Brian Njenga" <[e-mail]>
보낸 날짜 : 2017-03-29 03:38:49 ( +09:00 )
받는 사람 : Jérôme François <[e-mail]>, Oscar Mauricio Caicedo Rendon <[e-mail]>, Sheng Jiang <[e-mail]>
참조 : [e-mail] <[e-mail]>
제목 : Re: [Idnet] Intelligence-Defined Network Architecture and Call for Interests



Got it, thanks! Great response. What other research questions do people have?



On Tue, 28 Mar 2017, 21:26 Jérôme François, <[e-mail]<mailto:[e-mail]>> wrote:



Hi,



class="gmail_msg">



class="gmail_msg">
Le 28/03/2017 à 20:06, Brian Njenga a écrit :



class="gmail_msg">

I'm glad that the discussions​ on how to make AI/ML useful in the design of the internet's architecture are continuing, albeit under a different name. I'd like to participate remotely as well.

I have a question that someone more thoughtful​ than I could answer: Are there research questions on how to anticipate network-threatening DDOS attacks, such as those coming from the Mirai botnet family, using ML? Even going as far as designing and standardising an efficient, secure network protocol for IoT devices. This is a complicated issue, which involves emerging markets, so I'm interested in useful ideas from any angle.

Regarding DDoS, I think we have to consider two angles even when considering protecting the network. The flooding-like attacks (1) and the most sophisticated attack that can highly degrade service performance even with few packets (2) (and of course this also concerns network service, even more than before with network softwarization which puts network functions into VM, being thus more exposed than before in my opinion).



class="gmail_msg">



class="gmail_msg">For type 1, there are usually observable signs of a future attack (without ML) in the hours or days before with an increasing load. However even if you observe it, it is very hard to fully characterize it and so prevent it to really happen. For type 2, you basically need to predict what should be the load induced by a packet or flow when considering the targeted service. Assuming you have no complete knwoledge of the service, you can try to use some learning and regression techniques to do it but assuming also that the trafic is encrypted, the problem becomes more difficult.



class="gmail_msg">



class="gmail_msg">jerome



class="gmail_msg">



class="gmail_msg">

Thanks, and I'm glad to meet you all.

Best regards,



class="gmail_msg">Brian Muhia.



class="gmail_msg">
On Tue, 28 Mar 2017, 20:02 Oscar Mauricio Caicedo Rendon, <[e-mail]<mailto:[e-mail]>> wrote:



class="gmail_msg">
Hi, all,



class="gmail_msg">



class="gmail_msg">
If there is a meeting, I would like to participate remotely.



class="gmail_msg">



class="gmail_msg">
Best regards,



class="gmail_msg">



class="gmail_msg">
Oscar



class="gmail_msg">



class="gmail_msg">
On Tue, Mar 28, 2017 at 11:29 AM, Sheng Jiang <[e-mail]<mailto:[e-mail]>> wrote:



class="gmail_msg">
Hi, all,



class="gmail_msg">



class="gmail_msg">Although there are many understanding for Intelligence-Defined Network, we are actually using this IDN as a term reference to the SDN-beyond architecture that we presented in IETF97, see the below link. A reference model is presented in page 3, while potential standardization works is presented in page 9.



class="gmail_msg">



class="gmail_msg">https://www.ietf.org/proceedings/97/slides/slides-97-nmlrg-intelligence-defined-network-01.pdf



class="gmail_msg">



class="gmail_msg">Although it might be a little bit too early for AI/ML in network giving the recent story of the concluded proposed NMLRG, we still would like to call for interests in IDN. Anybody (on site in Chicago this week) are interested in this or even wider topics regarding to AI/ML in network, please contact me on [e-mail]<mailto:[e-mail]> . Then we may have an informal meeting to discuss some common interests and potential future activities (not any activities in IETF, but also other STO or experimental trails, etc.)  on Thursday morning.



class="gmail_msg">



class="gmail_msg">FYI, we have already working on a Work Item, called IDN in the ETSI NGP (Next Generation Protocol) ISG, links below.



class="gmail_msg">



class="gmail_msg">https://portal.etsi.org/tb.aspx?tbid=844&SubTB=844



class="gmail_msg">https://portal.etsi.org/webapp/WorkProgram/Report_WorkItem.asp?WKI_ID=51011



class="gmail_msg">



class="gmail_msg">Meanwhile, please do use this mail list as a forum to discuss any topics that may applying AI/ML into network area.



class="gmail_msg">



class="gmail_msg">Best regards,



class="gmail_msg">



class="gmail_msg">Sheng



class="gmail_msg">_______________________________________________



class="gmail_msg">IDNET mailing list



class="gmail_msg">[e-mail]<mailto:[e-mail]>



class="gmail_msg">https://www.ietf.org/mailman/listinfo/idnet



class="gmail_msg">



class="gmail_msg">



class="gmail_msg" clear="all">



class="gmail_msg">--



class="gmail_msg">
Oscar Mauricio Caicedo Rendón
PhD Computer Science - Federal University of Rio Grande do Sul
Full Profesor - University of Cauca



class="gmail_msg">
________________________________
Universidad del Cauca: Comprometidos con la calidad.
_______________________________________________



class="gmail_msg">IDNET mailing list



class="gmail_msg">[e-mail]<mailto:[e-mail]>



class="gmail_msg">https://www.ietf.org/mailman/listinfo/idnet



class="gmail_msg">
--



class="gmail_msg">
Some say he really tries to learn efficiently.



class="gmail_msg">




class="gmail_msg">

_______________________________________________
IDNET mailing list
[e-mail]<mailto:[e-mail]>
https://www.ietf.org/mailman/listinfo/idnet




class="gmail_msg">
--



Some say he really tries to learn efficiently.