Authentication

Authentication

Authentication means principals or data origin can be identified accurately / verification of identity of a person or system.

Link to original

Strong authentication

Cryptographic challenge-response protocols

Cryptographic challenge-response protocols (“strong authentication”): one entity (claimant) “proves” its identity to another entity (verifier) by demonstrating knowledge of a secret known to be associated with that entity without revealing the secret itself to the verifier during protocol.

Link to original

Weak authentication

Weak authentication

Conventional passwords schemes involve time-invariant passwords (“weak authentication”). The basic idea is that the user provides a user ID and password pair which is checked against the known information to authenticate the user.

Link to original

Password Attacks

• Dictionary attacks

  Dictionary attack: passwords found in lists of words can be brute forced. Entropy of passwords may be increased to prevent brute force attack potential.

  Link to original

• Password ageing

  Password ageing is limiting a password’s life span to force it to be changed regularly.

  Link to original

• Passphrases

  Passphrases use easily memorable phrases while retaining entropy.

  Link to original

PINs

Personal Identification Number

Personal identification numbers (PINs) are fixed (time-invariant) passwords.

• Usually used in combination with “something possessed”, such as a card.
• May serve as second level of security for entry into buildings which have independent first level of security (guard, video camera, etc).

Link to original

Multi-factor authentication

Multi-factor authentication

Multi-factor authentication is an authentication method in which a user has to successfully present authentication factors from at least two of the three categories:

• knowledge factors (“things only the user knows”)
• possession factors (“things only the user has”)
• inherence factors (“things only the user is”)

Requiring more than one independent factor increases the difficulty of providing false credentials.

Link to original

One-time passwords

One-time password

A one-time password is where each password is used only once, this prevents passive adversaries from eavesdropping and later attempting impersonation.

Can be implemented through:

• Shared lists of one-time passwords.
• Sequentially updated one-time passwords (derive new password after use).
• One-time password sequences based on one-way function.

Link to original

Zero-knowledge protocols

Zero-knowledge proof

In zero-knowledge proofs we specify a statement that is being proven.

• The statement is revealed to the verifier.
• The verifier (or others) should not learn anything else.
• Everybody can draw conclusions from everything they learned.

Link to original

Fiat-Shamir Identification Protocol

Fiat-Shamir Identification Protocol

There are three principals Prover Peggy, Verifier Victor, and Trusted Third Party Trent. We setup as follows:

• Trent chooses two large prime numbers $p$ and $q$ to calculate $n=p\times q$.
• $n$ is announced to the public, whereas $p$ and $q$ are kept secret.
• Peggy chooses a secret number $s$ between $1$ and $n-1$, and calculates $v=s^2\text{ mod }n$. She keeps $s$ as her private key and registers $v$ as her public key with the third party.

Victor knows $v=s^2\text{ mod }n$ but does not know $s$. Peggy wants to convince Victor that she knows the secret $s$ but Victor should not learn $s$.

We believe that square root modulo $n$ is not easy to compute.

Verification of Peggy by Victor proceeds in 4 steps (this is a round):

1. Peggy chooses a random number $r$ between $1$ and $n-1$. $r$ is called the commitment Peggy calculates the witness $x=r^2\text{ mod }n$ and sends it to Victor.
2. Victor sends the challenge $c$ to Peggy, which is either $0$ or $1$.
3. Peggy calculates the response $y=(r\times s^c)\text{ mod }n$ and sends it to Victor to show that she knows her private key $s$ modulo $n$. She claims to be Peggy.
4. Victor calculates $y^2\text{ mod }n$ and $(x\times v^c)\text{ mod }n$. If these values are congruent, then Peggy either knows the value of $s$ (honest) or she has calculated the value of $y$ is some either ways (dishonest) because in module $n$ arithmetic we have that: $y^2{=}_n(r\times s^c{)}^2{=}_n{r}^2\times s^{2c}{=}_n{r}^2\times (s^2{)}^c{=}_{n}x\times v^c$ The verification is repeated several times with the value of $c$ equal to $0$ or $1$ at random. Peggy must pass the test in each round to be verified: if she fails one single round => we abort the process entirely.

Link to original

Example: trying to cheat

Pamela does not know the secret $s$ but tries to prove its knowledge. Pamela guesses that Victor is going to choose $c=0$ (if she guesses wrong, she loses).

Or for $c=1$:

Victor would like to learn the secret $s$ but we can conclude that Victor learns nothing except the proved statement. (zero-knowledge property of the protocol)

Pamela must find numbers $x$ and $y$ such that $x\times v{=}_n(x\times s^2)$. => Choose $y$ randomly then set $\frac{r^2}{v}\text{ mod }n$.

Other zero-knowledge protocols

Feige-Fiat-Shamir Protocol: uses a vector of private keys, public keys, and challenges. Guillo-Quisquater Protocol: is an extension in which fewer number of rounds can be use to prove the identity of the claimant.

Social engineering

Social engineering

Social engineering is the acquisition of security-sensitive information or unauthorised access privileges by an outside attacker, based upon the abuse of a trust relationship.

Human-based social engineering:

• Social engineer may call a help desk, the engineer will often know names of employees and lingo of the company.
• A social engineer may pretend to be an important user which means their request is less likely to be turned down, and they may threaten to report the employee to their supervisor if they don’t comply.
• The social engineer may have obtained the name of someone in the organisation who has authority to grant access information, and thereby pretend that they sent them.
• A social engineer may pretend to be someone from an infrastructure-support group.

Computer-based social engineering:

• Malware in mail attachments
• Websites which farm credentials
• Popup windows that farm information
• Fake wireless networks that could capture information
• Phishing: attackers use web-based services to launch attacks on devices connected to web to acquire various bits of information by masquerading as someone trustworthy within some electronic communication.
  • SMiShing: phishing over SMS
  • Vishing: phishing over VoIP

Link to original

Malware & Firewalls

Not examined, slides lec 3. 14-34.