Mini Symposium on Information Security

Northeastern University College of Computer and Information Science

November 9, 2004

 


Whitfield Diffie -The Prospects for Secure Computing. 2

Elisa Bertino - Trust Negotiation Systems - Concepts and Issues. 4

Insup Lee – Bridging the gap between specification and implementation. 5

Guevara Noubir – Scalable  Robust and Secure Wireless Networks. 6

Burt Kaliski - Cryptography and Data Security - Long Term Challenges. 6

About these notes  7

 

Whitfield Diffie -The Prospects for Secure Computing

Information security is 100 years old

            Radio changed everything

                        Before radio – central command of navy had no clue where his fleet was

                        After: A single day, soon down to hours or minutes

                        BUT: other people can listen in – no control over the channel

                        Primary information security problem

            Radio uses cipher codes

                        Not terribly well suited to human use – to easy for error propogation

                        In WWI – the Enigma design of electro-mechanical encryption

                        Need table-lookup ability

            WWII – had 10-character/second telegraph

            WWII – 1st secure phone – Sig Sally

                        200bit/second vocoder

                        Uses one-time pad

long-playing records of random bits

                        To replicate this speed, need electronic systems w/ shift registers: can’t use electomechanical

            70s – still character lookup

Public Key – move from multiplication to exponentiation

 

Crown jewel of crypto – Advanced Encryption Standard  (AES)

            Would have been incomprehensible 10 years ago

            Supports long key lengths, even at lowest strength

            Adopted by the US gov’t, and in 2003 for all levels of classified data

                        Cost considerations imply that it likely will come into use

 

Crypto Rump Session – found collisions in SHA-0

Too much WWII mentality for new algorithms

            People are too afraid to their encryption algorithm being broken

            Too many startups based on new systems

            “I think we’re in pretty good shape”

 

Paradigm shift in computer security in the 70s

            With the advent of multiple user systems, have new question of seperation inside the system

            Off-the-shelf systems spread through the research world

                        Not trusted by the standard Intelligence security community

            Nightmare scenario: a compiler is taking notes on a classified program that is compiling

                        à Confinement

                                    Run programs in confined environments so that they can’t hurt you

                                                70s-80s – doesn’t leak your top secrets

                                                Today – “we’re happy if it doesn’t take over your machine”

                                    Rules: read only what you’re supposed to read, can’t downgrade clearances

                                    Problem: covert channels inside system

                                                I.e. can signal other programs by trying get into the run queue

            30 years on – Best System is Sun’s Trusted Solaris

                        BUT – still worse than the original specification

Why will we do better in the next decade than we have in the past 3?

            1) Cryptography

                        We know more about cryptography now, more widely implemented

                        More fundamental than network traffic: can use it to define and verify system states

                                    Use it to secure, authenticate system state

            2)Hardware

                        Guard the keys on the hardware level

                                    Prevents other software from undermining key system

                        No way to get the key if the hardware doesn’t have the command to get key

            3) Trusted computing platforms

                        Ask whether I am running what I think I’m running

                                    Simplest: boot off original media (CD from factory)

                                                BUT – nflexible, can’t patch

                                    More flexible – boot off a boot server

                                                BUT – need to make sure that you’re getting the right image

                                                            Have to check sigs on the

                                                BUT – only use one OS

                                    Most flexible – attestation set

                                                Credible claims that you are running what you actually are

            4) Networking

                        Another least-appreciated computer advances – Client-server computing

                                    Have a sensitive process: put it on a computer by itself

                                    Sensitive information is isolated, just focus on good gate-keeper

                        Want to flexibly allocate processes to processors

                                    Hardware-enforced domains that allocate resources, keep processes isolated

                                                Helps (but doesn’t solve) covert channel

                        Grid computing – computers as networks

                                    Opposed to Von Neuman machine, which uses flexibility between information and instructions

                                                Source of stack execution exploit (Buffer overflow)

                                    Grid – every process uses network I/O

                                                Can now use network policy

            5) Language advances

                        1st generation talked about operating systems

                                    Don’t need to secure all instructions

                        Nature of information is to be inscrutable – i.e. halting problem

                                    BUT – can build decidability into language

                                                i.e. Architecture only allows things to run for a certain time limit à things will always halt

                                    Analog – statement about investigability of code

                                                E.g. We don’t know that you are bad, but you were unaccountable for 3 years, so you could be a spy

                        Solution: only run code that you can understand and inspect

                        Java is the camel’s nose

                                    Java code will behave nicely

                                    BUT – need to inspect that the binary is from a compiled java program

                                                Inspect the binary to prove that it came from a java program – Is this possible???

                        Proof-carrying code – executables come with provable attestations that they came from good code

           

Improvements / Predictions

            Web services are going to work

                        Caveat – some disagree

                        Now – know where my programs are running

                                    One my desk

                                    At ebay, amazon, etc

            Algorithm selling

                        Suppose Fast Fourier Transforms were proprietary – only run by Princeton

                        We have the possibility of going into business to run algorithms inhouse

                                    People can outsource processing

            Utility computing

                        Complements the algorithm startup

                        BUT – how do you manage trust flow?

                                    It will have to look like the sub-contracting procedure

                                                Building security into contracts

                                    Can we learn anything from contract theory?

                                                Need enforcement mechanisms

                                                            Have some real-world examples: secure copying center

                                                            Negotiation and configuration control

Q: Speaker yesterday brought up question of how you now the web connection is valid.  What about algorithm?  How do you guarantee trustworthiness?

A: Need good interface (why is the browser resizable), need authenticaion (SSL) but should use bilateral authentication – scalability

            BUT – he acknowledges that SSL only checks if someone paid verisign

            Need to start with a trusted client

                        I.e. trusted solaris has a part of the screen that no processor can get rid of

            Problem of authenticating processes is practical, not principle of CS

 

Q: How do you control the life-cycle of a document?

A: Problem in the large and the small

            Large: why are some secrets kept for decades, but not others?

                        Interesting top-level secrets leak faster than boring confidential

                        STILL: why don’t secrets leak faster?

                                    Same social network that spreads can also slow them

                        Need a political science/ sociology study of secrets

            Small:  Can’t clean up after ourselves – very little disk-scrubbing

 

C: Browsers are buggy, frustrating

A: Need a keying infrastructure

            Not an adequate commercial motivation – high startup cost

 

Elisa Bertino - Trust Negotiation Systems - Concepts and Issues

Many definitions of trust        

            Kini & Choobineh – belief based on opinions of systems

                        Human element

            Gambetta – quantifiable probability that agent will perform action

                        Independent of monitoring

            Gradison & Sloman

                        Firm belief in the compentency to act dependably, securely and reliably

Trust relations

            Specific for given type of transaction

                        A trusts B for X

            Measurability is desirable

                        At least relative valuation: A trusts B more than C

            Directed

                        A trusts B is not the same as B trusts A

Things that can help with trust management

            Identity services

            Authentication services: access to resources

Delegation, fine-grained access,

            Anonymity services – need some accountability too

            Trust rating and recommendation

            3rd party notarization

            Auditable logs, secure storage, guaranteed message delivery

            BUT – information flows so variably, that you have to control all the flows, or cannot control any of it.

 

Trust negotiation model

            Goal: establish trust between parties in order to exchange sensitive information and services

            Mutual exchange of credentials

            Disclosure policies are ad hoc,

            Alice wants to see credential from ecommerce site about BBB before she releases her credentials

Language requirements

            Well-defined semantics

            Monotonicity – not sure how this matters…

            Credential combination

            Authentication

            Constraints on property values (properties of attributes)

                        Privacy issues if anyone can prove predicate

            Sensitive policies – not sure why I don’t want to spread the policy?  Is this security through obscurity?

System requirements

            Credential ownership

            Credential validity

           

Examples of systems

            Keynote – Blaze and Faigenbaum

            TrustBuilder

            Trust-X – Bertino et al.

Trust-X

            XML framework

            Sequential execution of phases

                        Introductory – no critical information exchanged

                        Policy exchange

                        Credential exchange

                                    Done in an order specified by policies

            Key phase: policy evaluation - bilateral

Privacy issues

            Include P3P as part of the policy

            Often demand more credentials than necessary

            Problem: possibility of deadlock where each side demands to see other credential first

                        Alice wants to see pharmacy’s cert before releasing

                        BUT – pharma will only disclose it to known patients à why?

            Can use privacy-enhanced credentials to avoid deadlock – follow up on how this works

Paranoid ratings:

            Standard

            Suspicious – always ask the other party to prove that they have credential

                        How do you prove that you have cert without revealing sensitive aspects of cert

            Strongly suspicious – demand actual credential

            Trusting – speed thngs up

Need to think about relative value of given pieces of information

            (Italian & University) < (American & University)

            Private concept groups – assumptions about what the other party knows

 

Q: Transitivity in trust management?

A: Could be very useful, but not the primary focus.

 

Insup Lee – Bridging the gap between specification and implementation


Problem –

            Scalability wrt. Size and complexity

            SW has more details

Solutions

            Model-based code generation

            Run-time verification

Model checking is formal, but doesn’t scale well

Funtime verification

            Use a verifier running alongside the program to see if the program has run properly

            Can use feedback to program to correct itself

MaC & runtime verification

… [took care of some administrative work during much of this talk]

Can use this for intrustion detection

            Make sure that the code has not been tampered with

 

 

Guevara Noubir – Scalable  Robust and Secure Wireless Networks

Many benefits of flexible, reconfigurable networks

            Ubiquitous computing

            Disasters – need to rebuild network

            Safety services

Need better technology for this – no integration, no QoS, no seamless adaptivity,

Characteristics of wireless networks

            Limited radio spectrum

            Shared medium – collisions

            Limited energy at nodes, limited computation

            Limited storage

            Dynamic topology with unreliable connectivity

            Need to enforce fairness

BUT – networks are very flexible, nodes can adapt

Attacks

            Multilayer DoS attacks

                        Physical layer

                                    Can partition network, or force packets to chosen paths

                                    Just needs to cause a few bits to be lost to ruin whole packet

                                    Counter: spread spectrum, directional antennas

                                                Higher level: crypto interleavers (cover traffic) jam-free or mobile

                        MAC layer – target control traffic and mechanisms

                                    Backoff allows attacket to spend less energy

                        Network – injection, or disruption of routing info

                                    Inject wrong routing data: disruption or resource consumption

                                                Black hole, routing loop

                                                Blackmailing – ruin a routers reputation

                        Transportation – exploit weakness in congestion control

Secure multi-casting

Ad Hoc vs. wired multicast

            Unreliable links, Higher packet loss

            Cost as a function of hops

Accumulative relaying

            Subset of a general relay problem

            Wave path – use a heuristic to find a lower energy boundary of transmission

`

 

Burt Kaliski - Cryptography and Data Security - Long Term Challenges


Two kinds of solutions: apply current knowledge to alleviate problems, or discover new knowledge that overcomes them

Looking ahead 30 years

1)Algorithms are robust for 30+ years for known attack on classical computers with sufficient keys

                        BUT: Quantum computers would break number theory-based public key crypto, halve effective private key size

                        Point: with few exceptions, no algorithms are proven secure unconditionally

Easy Solutions (with current tech)

            Employ multiple different algorithms

                        I.e. sign something twice

            Deploy secret-key crypto long term

            Use Merkle hash signatures – not number theory based

                        BUT – could still be vulnerable to specific attacks

            Include a link with quantum cryptography

Better solutions

            Develop alternative algorithms for PK’s

            Need a long testing process

            Provable resistance to attacks, either new algs or existing ones

            Quantum crypto that does not require photon transmission

2)No data is safe in the long run

                        Usability competes with security

                        Side channel analysis can expose keys

                                    i.e. number of multiplies in a decryption can reveal key

                        Availability requirements encourage multiple copies of data

                        Storage of keys

            Easy solutions

                        Build implementations that address side-channel attacks

                        Employ architectures with implicit keys

                                    Secret splitting – n shares with k required to reconstruct

                                    Distributed cryptography – k shares are never reconstituted, but used in series

                        “Heal” the effect of compromises

                                    Pro-active – assume that they will be compromised, shares are periodically refreshed

                                                Limits future use of breakdown

                                                Can do it simply from a mathematical perspective – split a 0 to maintain the sum

                                    Forward – keys are updated periodically such that past keys cannot be computed from current ones

                                                Put bounds on compromise – maintains validity of past signatures

            Better

                        Better algorithms

                        Develop new, practical data protection techniques

                        Invent something physics-based

3) Future networks may be less secure

            Mobile or ad hoc creates more attack opportunities

                        Router table issues

                        Selfish nodes

Need an architecture of implicit data built on a foundation of provable algorithms

            From gigabit security to terabit and beyond

 

Q: What about policy issues?

A: Attention to security – that it’s part of safety

            Need R&D, need to think about quantum computing

                        Will the market sustain growth?  What is the marginal benefit of a partial quantum computer?

                        ROI

 

C: Quantum encryption in RF is possible, but the energy is less than ambient temperature

 

 

About these notes

These notes were transcribed by Allan Friedman at the conference.  They represent only a fraction of the papers presented, since the conference consisted of several parallel sessions.  It is possible that I may have misunderstood some of these presentations, or missed an important point.  All italic comments are personal notes, and should not be attributed to the speaker.  It goes without saying that curious readers should read the actual papers; those interested should contact the authors.