Today a great number of possibilities exist for gathering information about an object, in order to decide which attack vector is optimal in any given instance. A significant part of the information comes from social networks, wireless networks, cellular networks, and the way the network is used. Collection of the information is mostly based on monitoring the network traffic between users, and between users and network servers. Most of the tools existing today for executing these actions are based on learning machines, analysis of the information and analysis of the traffic layers.


The problem:

Recently these tools have experienced problems and new challenges, due to the process quickly accelerating with the encouragement of giant companies like Google, turning the Internet traffic into encrypted [data] (https, ssl, ssh). This process almost completely nullifies the ability to perform information analysis, and limits the analysis of traffic layers. This necessitates new solutions from the field of learning machines.


The solution:

In light of this, Dr. Ofir Pele and Dr. Amit Dvir teamed up together with CyberBit Company under the Info-Media Consortium, in order to develop learning systems that will enable improved selection of the attack vector on an object, by means of analysis of the encrypted traffic without cracking it. The system will focus on identifying elements related to the potential attack target, such as:

- The type of device

- The type of operating system

- The type of browser being used

Congestion, problems, disconnections in the network

Type of usage – which actions were performed (did I send or receive mail? Did I upload pictures or just text?)

It is important to understand that in order to create an optimal attack vector, we must know what weaknesses exist with the object (weaknesses in the operating system, device, browser, etc.).


Research goals:

A learning system for characterizing and classifying encrypted traffic that will enable the execution of optimization for the attack vector, by means of identifying the type of device, operating system and other parameters.


The research team:

Dr. Ofir Pele – Chief Researcher

Dr. Amit Dvir – Chief Researcher

Yehonatan Tzion – Master’s degree student

Yehonatan Milshtein – Master’s degree student

Maor Behomie – Undergraduate student

Yosi Amichai – Undergraduate student

Itai Kirschenbaum - Undergraduate student

Yonah Kenon - Undergraduate student

Eviatar Grestal - Undergraduate student

Eliran Logasi - Undergraduate student