The TOR Data Communication System

Sunil Agarwal1CG12CS100

Guide: Mr. Sreekanth V

Outline

Problem History TOR Advantage over old onion routing Design Components of TOR How TOR works Cells

Outline

Circuit creation Additional functionality Hidden services Features and advantages Disadvantages Attacks and defense

Problem

Internet surveillance like traffic analysis reveals users privacy.

Encryption does not work, since packet headers still reveal a great deal about users.

Anonymity is needed. Solution: a distributed, anonymous network

History

Chaum’s Mixes(Mix Network)First described by David Chaum in 1981Routing protocol that create hard-to-trace

communications by using a chain of proxy servers known as mixes, which take messages from multiple senders, shuffle them, and send them back out in random order to the next destination.

History

Disadvantage of Chaum’s mixOverhead of using public and private key encryption

and decryptionLightly implemented and testedHigh delaySusceptible to end-to-end timing attack

History

Onion RoutingA technique for anonymous communication over a computer

networkMessages are encapsulated in layers of encryption, similar to

layers of an onionEach intermediary node knows only the location of the

immediately preceding and following nodesdeveloped in the mid-1990s at the U.S. Naval Research

Laboratory by employees Paul Syverson, Michael G. Reed, and David Goldschlag to protect U.S. intelligence communications online

further developed by the Defense Advanced Research Projects Agency (DARPA) and patented by the Navy in 1998.

What is TOR

TOR is a distributed anonymous communication service using an overlay network that allows people and groups to improve their privacy and security on the Internet.

Individuals use TOR to keep websites from tracking them, or to connect to those internet services blocked by their local Internet providers.

TOR's hidden services let users publish web sites and other services without needing to reveal the location of the site.

Advantage of TOR over Onion Routing No mixing, padding, or traffic shaping Many TCP streams can share one circuit Directory servers End-to-end integrity checking Perfect forward secrecy Congestion control

Design

Overlay network on the user level Onion Routers (OR) route traffic Onion Proxy (OP) fetches directories and

creates virtual circuits on the network on behalf of users.

Uses TCP with TLS All data is sent in fixed size (bytes) cells Each onion router maintains a long-term

identity key and a short-term onion key

Components of TOR

Client: the user of the TOR network Server: the target TCP applications such as web servers TOR (onion) router: the special proxy relays the application data Directory server: servers holding TOR router information

How TOR WorksAlice Bob

OR2

OR1

M√M

A circuit is built incrementally one hop by one hop Onion-like encryption

Alice negotiates an AES key with each router Messages are divided into equal sized cells Each router knows only its predecessor and successor Only the Exit router (OR3) can see the message, however it does

not know where the message is from

M

OR3

MC1

C2

C3

C1 C2

C2 C3

C3 Port

Cells All data is sent in fixed size (512 bytes) cells There are 2 types of cell: Control and Relay Control cell commands:

Padding, create, destroy. Relay cell commands:

Begin, data, connected, teardown, extend, etc. Having constant packet size camouflages the

type of data being exchanged

Cells

CircID CMD DATA

509 bytes12

CircID CMDDigest498 bytes12

RelayStreamID Len DATA

22 1 1

Controlcell

Relaycell

Commands in UseAlice CarolBob Website(link is TLS-encrypted) (link is TLS-encrypted) (unencrypted)

Create c1, E(gx1)

Created c1, gy1, H(K1)

Relay c1{Extend, Carol, E(gx2)}Create c2, E(gx2)

Created c2, gy2, H(K2)Relay c1{Extended, gy2, H(K2)}

Relay c1{{Begin <website>:80}}Relay c2{Begin <website>:80}

(TCP handshake)Relay c2{Connected}

Relay c1{{Connected}}

Relay c1{{Data, ”HTTP GET…”}}Relay c2{Data, ”HTTP GET…”}

”HTTP GET”

(response)Relay c2{Data, (response)}

Relay c1{{Data, (response)}}

. . . . . .. . .

Legend:E(x) – RSA encryption{X} – AES encryption

cN –a CircIDH –secure hash function

Additional functionality

Integrity checking Only done at the edges of a stream SHA-1 digest of data sent and received First 4 bytes of digest are sent with each message

for verification Rate limiting and fairness

Tor servers use a token bucket approach to enforce a long-term average rate of incoming bytes, while still permitting short-term bursts above the allowed bandwidth.

Congestion Control

OR-to-OR congestion might happen if too many users choose the same OR-to-OR connection.

Circuit Level throttling 2 windows keep track of relay data to be transmitted to

other ORs (packaging window) and data transmitted out of the network (delivery window)

Windows are decremented after forwarding packets and increments on a relay sendme message towards OP with streamID zero.

When a window reaches 0, no messages are forwarded

Congestion Control

Stream-level throttling similar to the circuit-level mechanism. ORs and OPs

use relay sendme cells to implement end-to-end flow control for individual streams across circuits.

Each stream begins with a packaging window (currently 500 cells), and increments the window by a fixed value (50) upon receiving a relay sendme cell.

Rather than always returning a relay sendme cell as soon as enough cells have arrived, the stream-level congestion control also has to check whether data has been successfully flushed onto the TCP stream.

Hidden Service and Rendezvous Points Location-hidden services allow Bob to offer a TCP

service without revealing his IP address. Tor accommodates receiver anonymity by allowing

location hidden services Design goals for location hidden services

Access Control: filtering incoming requests Robustness: maintain a long-term pseudonymous identity Smear-resistance: against socially disapproved acts Application transparency

Location hidden service leverage rendezvous points

Features and Advantage

More the number of users and relays, better the anonymity.

Ease of use through socks proxy Protection against strong and weak attacks No single TOR node is aware of the complete

plan of communication TOR builds anonymous paths for the client

based on a list of bridge nodes

Disadvantages

QoS considerations for dynamically assigning certain traffic to certain categories due to packet encryption

Directory information servers can be blocked Central servers for TOR discovery that lists TOR

servers and guarantees connectivity for TOR servers

Open design, Open Source Code Slow Performance

Attacks and defense

Passive Observing user

traffic patterns End-to-end

timing correlation

End-to-end size correlation

Website fingetprinting

Active Compromise

keys Iterated

compromise Run a onion

proxy DoS non-

observed nodes

Directory Destroy

directory servers

Trick the directory servers into listing a hostile OR

Conclusion This seminar presented the TOR anonymous system

and its corresponding details that have made such a system a success.

Avoiding traffic analysis, and hiding the identities of users, is the aim of any anonymous system. Since most anonymous systems rely on aging encryption technologies for which global adversaries are a capable of compromising, then the integrity of data might be at stake.

One of the key elements that worry anonymous systems researchers is QoS for the bandwidth utilized by peers on the systems and the overall network performance.

Conclusion More research in QoS and a bandwidth-choking

approach is required while concentrating on security and functionality implications.

Future work involves conducting a detailed study and comparison of the various anonymous systems

THANK YOU