Cyber 221 Exam2

What is authentication and its two main goals?

Authentication is verifying that an entity is who it claims to be. Goals: (1) verify identity, (2) ensure trust/integrity of the interaction.

What is identification authentication?

Verifying the identity of a user or entity (e.g., proving who you are when logging in).

What is message authentication?

Verifying the origin and integrity of a message — confirming it came from a specific sender and was not altered.

What is message origin authentication?

Confirming WHO sent the message — verifying the sender's identity.

What is message content authentication?

Confirming WHAT was sent — verifying the message content has not been tampered with.

Why can't identity be proven using name, PSU ID, or SSN?

These can be stolen, guessed, or looked up. Anyone with that info can impersonate you. True authentication requires something only the legitimate entity can demonstrate.

What are the three categories of authentication factors?

(1) Something you know (password/PIN), (2) Something you have (token/phone), (3) Something you are (biometrics).

Explain "Something You Know" as an authentication factor.

A secret known only to the user, like a password or PIN. Easy to implement but vulnerable to theft, guessing, or phishing.

Explain "Something You Have" as an authentication factor.

A physical object the user possesses, like a security token or authenticator app. Harder to steal remotely but can be lost or physically stolen.

Explain "Something You Are" as an authentication factor.

A biometric characteristic unique to the individual, like a fingerprint or retina scan. Hard to forge but raises privacy concerns and can't be changed if compromised.

What is multi-factor authentication (MFA) and why is it more secure?

MFA requires credentials from two or more different factor categories. Compromising one factor is not enough — an attacker must simultaneously compromise multiple factors.

Describe the step-by-step process of one-way user authentication over a network.

(1) Client sends username. (2) Client submits hashed password. (3) Server looks up stored hash. (4) Server hashes submission and compares. (5) Match = access granted; no match = denied.

What role does the client play in authentication?

Collects user credentials, potentially hashes the password locally, and transmits credentials securely to the server.

What role does the server play in authentication?

Stores hashed passwords, compares submitted hash to stored hash, and makes the accept/deny decision.

What is the password overload problem?

Users must manage many passwords across many services, making it cognitively impractical to remember unique, strong passwords for each one.

Why do users tend to reuse passwords?

Memorizing many unique, complex passwords is too difficult. Users reuse the same password or simple variations to reduce cognitive load.

What risks does password reuse create?

If one service is breached, attackers can use the same password on other services (credential stuffing), causing a cascade of account compromises.

What are two major flaws of storing passwords in plaintext?

(1) If the file is stolen, all passwords are immediately exposed. (2) Admins or insiders with file access can see all users' passwords.

What is a hash function and why is it useful for password storage?

A one-way function that produces a fixed-size digest that cannot be reversed. Only the hash is stored — the real password never needs to be saved.

Why does hashing improve security if the password database is stolen?

Hashes are one-way — attackers cannot directly reverse them. They must brute-force or use dictionary attacks, buying time for users to change passwords.

What properties must a cryptographic hash function satisfy?

Pre-image resistance, second pre-image resistance, collision resistance, deterministic output, and the avalanche effect.

What problem do one-time password (OTP) systems solve?

They solve the replay attack problem — since each password is valid only once, intercepting it is useless after it's been used.

How does the S/Key Lamport scheme generate one-time passwords?

It applies a hash function repeatedly N times to a secret seed, creating a chain. Each login uses one value from the chain in reverse order.

What is a hash chain and how is it used in S/Key?

A sequence where each value is the hash of the previous: x₀ → h(x₀) → h²(x₀) → ... → hⁿ(x₀). S/Key uses this chain, consuming one value per login in reverse.

Why are S/Key passwords used in reverse order?

Hash functions are one-way. The server stores hⁿ(x₀) and the user provides hⁿ⁻¹(x₀). The server hashes what the user provides and checks it equals the stored value.

What does the server store in the S/Key password file?

The current hash value (last valid hash used) and the sequence number (how many hashes remain).

Why is stealing the S/Key password file less useful?

The stored value is a hash — attackers cannot reverse it to find the next valid password due to one-way hash properties.

What happens to the stored hash after each successful S/Key login?

It is replaced with the hash the user just submitted (one step earlier in the chain) and the sequence number decrements by 1.

What is the basic idea behind challenge-response authentication?

The server sends a random challenge. The client computes a response using the challenge + their secret. The server verifies the response independently.

Why does the server generate a random challenge?

To ensure the response is unique to each session — a previously captured response cannot be replayed against a new, different challenge.

What information is used to compute the challenge-response?

The random challenge from the server combined with the user's shared secret, typically via: Response = hash(challenge || secret).

Why is a challenge-response effectively a one-time password?

The response depends on a unique random challenge each session, so the same secret produces a different response every time. A captured response cannot be reused.

How does challenge-response protect against replay attacks?

Each session uses a fresh random challenge. A previously captured response is tied to an old challenge and will fail against the new one.

What problem does public key cryptography solve?

It solves the key distribution problem — two parties can communicate securely without having shared a secret key in advance.

What is a public key?

A key freely shared with anyone. Used by others to encrypt messages for you, or to verify your digital signature.

What is a private key?

A key kept completely secret by the owner. Used to decrypt messages encrypted with your public key, or to create digital signatures.

Why can public keys be shared but private keys must stay secret?

Messages encrypted with a public key can only be decrypted with the matching private key. Without the private key, ciphertext is unreadable — even knowing the public key is useless for decryption.

How does Alice send a confidential message to Bob using public key crypto?

Alice obtains Bob's public key, encrypts the message with it, and sends the ciphertext. Only Bob can decrypt it with his private key.

Which key encrypts a message in public key cryptography?

The recipient's PUBLIC key is used to encrypt.

Which key decrypts a message in public key cryptography?

The recipient's PRIVATE key is used to decrypt.

What is the purpose of a digital signature?

To provide authentication, integrity, and non-repudiation for digital messages — verifying the sender and ensuring the content hasn't been altered.

How does a digital signature prove the sender's identity?

The sender signs with their private key. Only the private key holder could create that signature. Anyone with the sender's public key can verify it.

Why can't symmetric encryption provide non-repudiation?

Both parties share the same key, so either could have created the message. There's no way to prove which party originated it.

What three properties do digital signatures provide?

(1) Authentication — confirms the signer's identity. (2) Integrity — confirms the message wasn't altered. (3) Non-repudiation — sender cannot deny signing.

How can a message provide both authentication AND confidentiality?

Apply two layers: (1) sign with sender's private key for authentication, then (2) encrypt with receiver's public key for confidentiality.

In what order should encryption be applied for both auth and confidentiality?

Sign first (with sender's private key), then encrypt (with receiver's public key). Receiver decrypts first, then verifies signature.

Why is it dangerous to trust a public key someone sends you directly?

Anyone can generate a key pair and claim the public key belongs to someone else, enabling impersonation without a trusted third party to verify the binding.

What is a man-in-the-middle attack in the context of public keys?

An attacker intercepts key exchange, substitutes their own public key for the legitimate party's, reads and re-encrypts all communications without either party knowing.

What is a public key certificate?

A digitally signed document that binds a public key to an identity, signed by a trusted Certificate Authority (CA).

What information is stored inside a digital certificate?

Subject name, subject's public key, issuing CA name, validity period, serial number, and the CA's digital signature.

Why are digital certificates compared to passports?

Like passports, they are issued by a trusted authority, contain identifying info, are hard to forge due to the CA's signature, and can be presented as proof of identity.

What is the role of a Certificate Authority (CA)?

A trusted third party that verifies identities and issues digitally signed certificates binding identities to their public keys.

Why must the CA be trusted?

If the CA issues certificates without proper verification, it enables impersonation. All trust in the PKI system depends on the CA's integrity.

What can happen if a CA is compromised?

Attackers can issue fraudulent certificates, enabling large-scale MITM attacks. The CA must be revoked, invalidating all certificates it ever issued.

How does a system verify a digital certificate?

It uses the CA's public key (pre-loaded as a trusted root) to verify the CA's signature on the certificate. If valid and not expired/revoked, the cert is trusted.

Why are hash functions used in digital signatures?

Signing large messages directly is computationally expensive. Instead, the message is hashed to a small digest, and only the digest is signed — efficient yet still provides integrity.

What is the hierarchical trust model in PKI?

A tree structure where trust flows top-down: root CA signs intermediate CA certs, which sign end-entity certs. Trust is verified by tracing the chain to the trusted root.

What is the role of a root CA?

The ultimate trust anchor in the hierarchy. Its self-signed certificate is pre-loaded into OSes and browsers as inherently trusted. It signs intermediate CA certificates.

What is an intermediate CA and why is it used?

An intermediate CA sits between root and end entities for security isolation — if compromised, only its certs are affected; the offline root CA remains secure.

What is a chain of trust?

The sequence of certificates from an end entity up to a trusted root: EndCert → Intermediate CA → Root CA. Each link is verified by the issuing CA's signature above it.

What problem does Single Sign-On (SSO) solve?

Users must authenticate separately to each service. SSO allows one authentication to grant access to many services, reducing password fatigue and login friction.

What is federated identity?

A system where a user's identity established with one organization is trusted and accepted by other organizations — authenticate once, recognized everywhere.

What are the main entities in an SSO system?

(1) User/principal, (2) Identity Provider (IdP) / SSO Authentication Server, (3) Service Provider (SP).

What role does the SSO authentication server play?

Verifies user credentials once and issues tokens/assertions that service providers accept as proof of identity.

What role does a service provider play in SSO?

Accepts tokens from the SSO server as proof of identity and grants access without requiring its own separate login.

Describe the two main steps of SSO authentication.

(1) Authentication: User logs in to the SSO server, which issues a token. (2) Authorization: User presents token to a service provider, which validates it and grants access.

Why does SSO improve both usability and security?

Usability: one login for all services. Security: fewer passwords mean less reuse; strong auth (MFA) only implemented once at the SSO server.

What are the three major SSO protocols?

(1) SAML, (2) OAuth / OpenID Connect, (3) Kerberos.

What environments typically use SAML?

Enterprise and B2B environments — e.g., employee SSO into corporate apps like Salesforce or Workday.

What environments typically use OAuth / OpenID Connect?

Consumer-facing web/mobile environments — e.g., "Sign in with Google" or "Sign in with Facebook" on third-party apps.

Where is Kerberos most commonly used?

Internal enterprise networks, particularly Windows Active Directory domains, for authenticating users to network services like file shares and printers.

What is Kerberos and what problem does it solve?

A network authentication protocol that authenticates users securely without repeatedly sending passwords over the network, using time-limited tickets instead.

What is the Key Distribution Center (KDC)?

The trusted central Kerberos server that stores all secret keys, issues authentication tickets, and is the foundation of Kerberos trust.

What are the two parts of the KDC?

(1) Authentication Server (AS) — authenticates users and issues TGTs. (2) Ticket Granting Server (TGS) — issues Service Tickets for specific services.

What is a Ticket Granting Ticket (TGT)?

A credential issued by the AS after initial login. It proves to the TGS that the user is already authenticated, without re-entering a password.

What is a Service Ticket?

A ticket issued by the TGS that grants the user access to one specific service, encrypted with that service's secret key.

Why does Kerberos use tickets instead of passwords?

Tickets are time-limited and service-specific — intercepting one has limited value. Passwords are only exposed once at initial login, reducing their risk.

What happens when a user first logs in to Kerberos?

The client sends the username to the AS. The AS returns a TGT (encrypted with KDC's key) and a session key (encrypted with the user's key derived from their password).

What role does the Authentication Server (AS) play in Kerberos?

It verifies the user's identity and issues the TGT and session key, establishing the authenticated session with the KDC.

What information is sent to the client after Kerberos authentication?

(1) A TGT encrypted with the KDC's secret key (opaque to client), and (2) a session key encrypted with the user's own key (decrypted using their password).

How does the client request access to a service in Kerberos?

The client sends the TGT plus an authenticator (timestamp encrypted with session key) and the desired service name to the TGS, which issues a Service Ticket.

What does the Ticket Granting Server (TGS) do?

Decrypts the TGT, extracts the session key, verifies the authenticator, and issues a Service Ticket encrypted with the target service's secret key.

How does a service verify the user in Kerberos?

The client presents the Service Ticket and a new authenticator to the service. The service decrypts the ticket with its own secret key, verifies the authenticator, and confirms identity — no KDC contact needed.

What are biometrics?

Measurable, unique physical or behavioral characteristics of an individual used to automatically identify or authenticate them.

Why are biometrics useful for authentication?

They cannot be forgotten, lost, or easily shared. They provide a strong link to the actual person, making impersonation significantly harder than with passwords.

What is the difference between physiological and behavioral biometrics?

Physiological: based on physical body structure (fingerprints, iris, facial geometry). Behavioral: based on action patterns (typing rhythm, gait, voice, signature dynamics).

What does it mean for a biometric trait to be universal?

Every person in the population possesses that trait — it can be used for everyone, not just some people.

What does it mean for a biometric trait to be unique?

The trait is sufficiently distinct across individuals — no two people share the same biometric value, allowing reliable differentiation.

Why must biometric traits be permanent?

If a trait changes over time, previously enrolled data won't match, making the system unreliable. Stable traits ensure consistent authentication over time.

What does it mean for a biometric trait to be collectible?

It can be measured and captured with practical sensors at acceptable accuracy — not requiring overly expensive, invasive, or inconvenient equipment.

What advantages do biometrics have over passwords?

Cannot be forgotten or lost, cannot be shared or transferred, cannot be brute-forced, provide a direct link to the physical individual, and require no memorization.

Why can biometric systems be expensive or difficult to implement?

They require specialized hardware and sophisticated matching software, must handle natural variation in readings, and need secure storage of biometric templates.

Why is enrollment quality critical in biometric systems?

The enrollment sample is the reference template for all future comparisons. A poor-quality initial scan leads to high false rejection rates and an unreliable system.