Analysis of encrypted traffic without decryption

System for traffic analysis without decryption. This method is simply called "machine learning." It turned out that if a very large volume of various traffic was applied to the input of a special classifier, the system with a very high degree of probability can detect the actions of malicious code inside encrypted traffic.



Network threats have changed and become smarter. Recently, the concept of attack and defense has changed. The number of events on the network has increased significantly. Attacks have become more sophisticated, and hackers have a wider range of attacks.


According to Cisco statistics, cybercriminals over the past year have 3 times increased the number of malicious programs that they use for their activities, or rather, to hide it - encryption. It is known from theory that the β€œcorrect” encryption algorithm cannot be hacked. In order to understand what is hidden inside the encrypted traffic, you must either decrypt it knowing the key, or try to decrypt it with various tricks, or by hacking it in the forehead, or using some vulnerabilities in cryptographic protocols.


Picture of the network threats of our time
Picture of the network threats of our time


Machine learning


! , .


(Machine Learning) β€” , , . «» . – -. «». ( 2- ). , - ( ) . , - ( ) () , , (). - , , , , . , , .


(if-else). . «» . .


. .


. 1 2012 – 10 , , 2 2011 .


. 20 . Β«- - Β». . : , . , . - Β« ?Β». – . – 1 , – 0. ( ). ( ), .


, , . 2- . – , – .


Binomial and normal probability distributions


, Β« ?Β». 2 . 1 – ( ). 2 – , ( ).


, . 2 – .


– . , – .


– ( , ). , .


. .



– . – ( ) Β« Β». , , . - , ( ), .


β€” BrainPort.


2009 , Β« Β», – BrainPort. BrainPort : , , . , .


BrainPort lollipop with glasses and camera
BrainPort


BrainPort at work
BrainPort



. , . – . , .


Encrypted Traffic Analytics


Encrypted Traffic Analytics Stealthwatch. Stealthwatch β€” Cisco , .


Stealthwatch Enterprise Flow Rate License, Flow Collector, Management Console Flow Sensor.


Cisco Stealthwatch Interface
Cisco Stealthwatch


, . ( ), «» . – WannaCry, Tor .


Visualization of the growth of traffic encryption in the network


Encryption in Macroeconomics


Encrypted Traffic Analytics (ETA) . , .


. - . TLS (transport layer security β€” ). , . , .


Encrypted Traffic Analytics (ETA) Principle
Encrypted Traffic Analytics (ETA)


. Cisco 2 – «» . , , , . , .


– , . – , , , .. , . , .


, – – , () .


– , 99% .


Steps for checking traffic in ETA
ETA


– . , ( ). -, ( ). -, ( , ).


Outdated Man-in-the-Middle Concept
Man-in-the-Middle


, .


4 : TCP/IP – , DNS – , TLS – , SPLT (SpaceWire Physical Layer Tester) – .


Protocols working with ETA
, ETA


. TCP/IP ( , ..), DNS , «» . TLS «» (fingerprint) (cert). – . , : , .. , .


, . , . , ( ). , . - .onion , . , .


ETA Work Result
ETA


, , , .


ETA


. -,
, .
, – . -,
, ,
, , .


Sensors and Stealthwatch
Stealthwatch


– , – . , , – ISR = Cisco (Cisco Integrated Services Router); ASR = Cisco (Cisco Aggregation Services Router); CSR = Cisco (Cisco Cloud Services Router); WLC = Cisco (Cisco Wireless LAN Controller); IE = Ethernet- Cisco (Cisco Industrial Ethernet); ASA = Cisco (Cisco Adaptive Security Appliance); FTD = Cisco Firepower Threat Defense; WSA = - (Web Security Appliance); ISE = (Identity Services Engine)


Comprehensive monitoring based on any telemetry data


Β«CiscoΒ» . , …


. Stealthwatch . , . 100 , , , , , , .. . . , , , ( 2). . . , , , , . . , Stealthwatch , , .


Anomaly Detection Using Behavior Modeling


:


Branch network deployment option (simplified)
()


Branch network deployment option


, . , , . .



, , , , . , , , .


, , , Stealthwatch. , «», « » , , , . , , .


, , β€” . , .


, Stealthwatch. Stealthwatch, , .


Multi-Level Machine Learning Capabilities


1 .


99 % . , , , , . , . β€” . . , , . 2 3 . .


. . 70 , Stealthwatch , , (DNS) - . 70 , , . , . .


. , . , - - . , , .


2 .


, , . , 90 %. :


  • - ( )
  • ,
  • Β« Β».

. , , .


. 100 . , , . , , . , .


. , . , , . , . . - .


3 .


β€” , , , , . , , , , Β« Β».


. 99 100 % , () . , . , , , . , , ,
, . , .


Multilevel training of the neural communication system based on ETA
ETA



, . , . , . Β« Β», . , , Stealthwatch , .


.


World map showing 460 million IP addresses
, 460 IP-


.


, ?


, . , , 2 .


1. . Cisco. , . , D-Link, MikroTik .. – . , .


2. . , - , , - , . , ( ), . , .


2 , . , .


All Articles