Chapter 9i: Public Key Infrastructures

Question 1

Why do we need PKIs?

• Advantage of asymmetric crypto: …
• This allows users to distribute public keys via arbitrary channels, e.g.,
  (1)
  (2)
• Problem: Who ever utilizes another entities public key must be sure that ….

Answer 1

Public keys are not required to be kept secret

1* Alice retrieves Bob’s public key from a mail sent by Bob
2* Bob downloads Alice’s public key from her web site

this public key really belongs to this entity → is authentic → can be trusted!

Question 2

Why do need PKIs?

• Imagine: MitM Eve has exchanged the public keys with keys controlled by her
• Neither Alice nor Bob can spot that the key they have received is not authentic
• As Eve owns the private keys, what can she do now?

Answer 2

• compute signatures in the name of Alice and Bob
• decrypt / re-encrypt confidential messages exchanged between Alice and Bob

Question 3

Why we need PKIs : How to establish “trust” in a public key?

Manual trust establishment: Alice and Bob exchange their keys via some method.

Explain the idea behind a PKI.

Answer 3

Question 4

Certificates: The essence of PKI

Definition:

Semantics of the binding: What does the identifier refer to?
What’s always necessary? What do we need to do if the identifier is a name?

Answer 4

A certificate is a digitally signed binding of an identifier of an entity and the public key of an asymmetric key pair owned by that entity

Question 5

Certificate creation

PKIs are created by issuing certificates between entities. Illustrate.

Answer 5

Question 6

Common forms of PKI

We can now classify PKIs by looking at:

• Who are the issuers?
• Which issuers must be trusted = which TTPs exist?
• How do issuers verify that X and KX-pub belong together, or that X is really X?

Explain the terminology: “issuer”, CA, “Endorser”.

Answer 6

• Depending on the PKI, different words for “issuer” exist
• Often in hierarchical PKIs: “Certification Authority” (CA)
• In non-hierarchical PKIs sometimes: “Endorser”
• These words often hint at the role (power) of the issuers

Question 7

Hierarchical PKIs

Why is this impractical?

Answer 7

• Who decides which global authority is trustworthy for the job?
• What are the agreed verification steps of an identity?
• Hard to imagine any government would rely on an authority outside its legal reach.
• The high load on the CA, might make it easier to trick the CA into mis-issuing a certificate.
• Single point of failure/attack.

Question 8

What’s the role of RAs?

Answer 8

Note:
* Registration Authorities are indeed used today by large CAs
* Example: TUM operates a CA for web and mail certificates,
multiple RAs exist for the different faculties and institutes
* RA for informatics: RBG

Question 9

Hierarchical PKIs: ‘Practical’ solutions to the problem

Many CAs :

Defining CAs as trusted:
* A CA must be trusted by participants in order to be useful
* How should participants decide which CAs to trust?
* “Solution”: ….

Answer 9

• One global CA is infeasible, even with RAs → Let us use many CAs, in different legislations.
• Should we accept them all equally? → No.

operating systems and software like browsers come pre-configured with a set of trusted CAs

Question 10

Webs of Trust may also take many forms:

Answer 10

• Trust metrics to automatically reason about authenticity of bindings between entity and key
• E.g. introduce rules how many delegations are allowed, store explicit trust values, etc.
• CAs may act as ‘special’ participants

Question 11

Currently deployed PKIs:

1. Hierarchical PKI(s) with many CAs
2. Webs of trust

Explain.

Answer 11