Title: NETW 05A: APPLIED WIRELESS SECURITY Data-Link Security Solutions
1NETW 05A APPLIED WIRELESS SECURITY Data-Link
Security Solutions
- By Mohammad Shanehsaz
- Spring 2005
2Objectives
- Static and Dynamic WEP TKIP
- Explain the functionality, strengths, and
weaknesses of WEP and TKIP - Explain appropriate scenarios and applications of
static and dynamic WEP and TKIP - Install and configure static and dynamic WEP
TKIP - Illustrate feasibility of WEP exploitation
- Manage scalable WEP TKIP solutions
3Objectives
- 802.1x and EAP
- Explain the functionality of 802.1x EAP
- Explain dynamic key generation and rotation for
solution scalability - Explain the strengths, weaknesses, and
appropriate applications of 802.1x EAP - Install and configure 802.1x EAP, including
LEAP, EAP-TLS, EAP-TTLS, EAP-MD5, PEAP, - Manage scalable 802.1x and EAP solutions
4802.11 MAC Basics
- Management and control frames are sent in clear
text and unauthenticated - This is the basis for many types of attack
scenarios - For some types of attacks particular vendors have
instituted proprietary solutions - Many of these vulnerabilities will be addressed
by the 802.11i standards
5Categories of Authentication Encryption
- There are three main categories
- Static WEP
- Dynamic WEP
- Proprietary protocols
- There are variations on each type
6Static WEP
- Security solution based on unchanging shared keys
that are preconfigured on all nodes by network
administrator - Protects the wireless link with simple
authentication and data encryption - Not a complete solution, it can be cracked using
common tools such as WEPcrack or Airsnort
7Cracking WEP
- Cracking WEP requires three things
- Large number of captured packets
- Long periods of time to capture those packets
- Fast machine to process the information contained
in the packets to derive the WEP key - It can takes days to crack it, is it worth it ?
8TKIP
- Temporal Key Integrity Protocol is a set of
modifications to the existing WEP algorithm - IEEE 802.11i task group created TKIP
- TKIP is a type of dynamic WEP solution where WEP
keys are rotated on a changeable interval, but
static WEP key is still used as keying material
9WEP Weaknesses Addressed
- TKIP algorithms address the following weaknesses
- Forgery
- Weak-Key attacks
- Collision attacks
- Replay attacks
10Forgery
- TKIP supports per-packet authentication
- Forgery attacks are performed by capturing
encrypted packets, changing some data within
them, and then resending the packets - TKIP uses message-integrity check (MIC) called
Michael to thwart attempts - MICs add significant network overhead
11Weak Key Attacks
- WEP construct a per-packet RC4 key by
concatenating an RC4 base key and 24 bits IV - TKIP uses key-mixing to derive shortlived
encryption keys - TKIP uses 128 bit temporal key combined with the
clients MAC address and large 48 bit IVs to
produce the key for encryption
12Collision and Replay attacks
- TKIP uses 48 bit IVs, which increases the
possible number of IVs, to prevent collision
attacks - TKIP prevents replay attacks by using sequencing
number for generated packets
13Availability
- For those products that are currently Wi-Fi
certified, most can be upgraded to support TKIP,
assuming the vendor has made a firmware upgrade
available- - check the web site for upgrades
14802.1x / EAP
- 802.1x with use of the Extensible Authentication
Protocol implements what is generally referred to
as dynamic WEP - Dynamic Key Generation, Distribution, Rotation
- EAP is a layer 2 authentication protocol
replacing PAP and CHAP - It is appropriate for medium to large enterprise
environment - Basing authentication on individualized user
credentials such as usernames and passwords,
certificates, smartcards and other like methods
15802.1x Standard
- IEEE standard that provides an authentication
framework for 802-based LANs - It was originally used in wired networks and has
since been adapted for wireless networks - It provides port-based access control so that
before the switch or access point will establish
a connection, the user credentials must be
verified - 802.1x standard addresses only access control and
authentication framework and does not address
data privacy, so that the problems with WEP still
exist, EAP eliminates the problems through
dynamic key generation
16802.1x Standard
- There are three terms defined by the IEEE
standard that describe the devices used in 802.1x - Supplicant-a client that is being authenticated
- Authenticator-an access layer device such as AP
or bridge that requires supplicants to be
authenticated in order to pass traffic through it - Authentication server-( typically a RADIUS ) the
device that is doing the authentication of the
supplicant
17802.1x Standard Advantages
- Maturity Interoperability
- User-based identification
- Dynamic Key Management
- Flexible Authentication
18Maturity Interoperability
- The industrys choice to use in WLAN because of
time-proven use in wired network - Supports of mature protocols such as EAP and
RADIUS which are open standards providing max
interoperability in centralized identification
and key management
19User-based Identification
- Basing authentication on actual user not a
particular wireless device, on a scalable
database such as RADIUS or other databases that
RADIUS directly supports (Active Directory, NDS,
LDAP, SQL) - Centralized authentication and management save
time and money
20Dynamic Key Management
- Per-user per-session keys eliminates attacks
based on obtaining the WEP key - Automated key management systems allow keys to be
reissued without an administrators intervention
21Flexible Authentication
- There are several supported authentication
solutions to choose from - Changing the authentication mechanism does not
require any hardware replacement
22EAP Protocol
- Provides an extensible method for PPP server to
authenticate its clients - EAP supports two-and three-factor authentication
(passwords, certificates, biometrics, etc) - EAP was designed to prevent proprietary
authentication solutions from being implemented
which would have had a negative effect on the
interoperability and compatibility between
systems - EAP is within OS of the server or application
software on the client - Windows XP natively supports EAP
23EAP Authentication Types
- There are many EAP types
- EAP-MD5
- EAP-TLS
- LEAP
- EAP-TTLS
- PEAP
24EAP-MD5
- First Authentication type created by RFC2284 for
802.1x - Uses the same challenges handshake protocol as
PPP-based CHAP, except challenges and responses
are sent as EAP messages - It has three weaknesses
- One-way authentication
- Challenge passwords
- No per-session WEP keys
- Rarely used because of its weaknesses
25EAP-MD5 Weaknessesone-way Authentication
- Because only the supplicant gets authenticated,
an impersonator could be added as rogue RADIUS
server to obtain the login credentials of a
legitimate user
26EAP-MD5 WeaknessesChallenge Passwords
- Authentication server challenge the supplicant
with a random string of text - The supplicant hashes the challenge with its
password and send it back - The server validates the response based on its
knowledge of the password - Eavesdropper can obtain both the challenge and
the hash, which he/she can break it with
dictionary attack to obtain users password
27EAP-MD5 Weaknessesno per-session WEP keys
- After authentication, communication is either not
encrypted, or encrypted with a static WEP key - Because of static WEP vulnerability , it allows
eavesdropping on the data
28EAP-TLS (EAP-Transport Level Security )
- Developed by Microsoft and standardized by
Internet Engineering Task Force - It is based on the secure socket layer protocol
used for secure web traffic - It uses both server-side and client-side
certificates for user identification (mutual
authentication) - More appropriate for organizations that have
already deployed a PKI (public key
infrastructure) - Per-session WEP key is set up, and client can be
re-authenticated and re-keyed as often as needed
without inconveniencing the end user
29TLS Authentication
- The TLS process begins with the handshake
process - 1. The SSL client connects to a server and makes
an authentication request - 2. The server sends its digital certificates to
the client - 3. The client verifies the certificates validity
and digital signature - 4. The server requests client-side authentication
- 5. The client sends its digital certificate to
the server - 6. The server verifies the certificates validity
and digital signature - 7. The encryption and message integrity schemes
are negotiated - 8. Application data is sent over encrypted tunnel
via the record protocol
30EAP-TLS Authentication
- The EAP-TLS authentication process is as follows
- 1. The client sends an EAP start message to the
access point - 2. The access point replies with an EAP Request
Identity message - 3. The client sends its network access identifier
(NAI), which is username, to the access point in
an EAP Response message - 4. The access point forwards the NAI,
encapsulated in a RADIUS Access Request message
to the RADIUS server - 5. The RADIUS server responds to the client with
its digital certificate
31EAP-TLS Authentication
- 6. The client validates the RADIUS servers
digital certificate - 7. The client replies to the RADIUS server with
its digital certificate - 8. The RADIUS server validates the clients
credentials against the client digital
certificate - 9. The client and RADIUS server derive encryption
keys - 10. The RADIUS server sends the access point a
RADIUS ACCEPT message, including the clients WEP
key, indicating successful authentication - 11. The access point sends the client an EAP
Success message - 12. The access point sends the broadcast key and
key length to the client, encrypted with the
clients WEP key
32EAP-Cisco Wireless (LEAP)
- Ciscos proprietary Lightweight Extensible
Authentication Protocol was designed to support
802.1x/EAP based authentication - It was developed to support networks with a
variety of OS that may not natively support EAP - LEAP supports mutual authentication between a
client and a RADIUS server - LEAP provides user-based, centralized
authentication as well as per-session WEP keys - Used in Ciscos Aironet products
- Its security level is considered moderate or
strong based on the strength of the passwords
used - See figure 11.12 on page 256 for LEAP Process
33EAP-TTLS (Tunneled Transport Layer Security )
- Was co-developed by Funk Software and Certicom,
supported in Funks Odyssey software - EAP-TTLS requires only an authentication server
certificate - TTLS uses TLS channel to exchange
attribute-value pairs (AVPs) - After authentication server is authenticated
using its digital certificate, an encrypted
tunnel is established between the supplicant and
authentication server to pass the supplicants
authentication credentials - See figure 11.13 for EAP-TTLS Process
34Key security Features of EAP-TTLS
- Almost any kind of supplicant authentication
credentials (passwords, tokens, etc ) can be used
inside the encrypted tunnel - Low overhead since requirement of only
server-side certificate - Many types of authentication algorithms may be
used inside the encrypted tunnel-MS-CHAPv2,
MS-CHAP, CHAP, PAP,EAP-MD5 - Strong protection against eavesdroppers seeking
to perform dictionary attack - Mutual authentication, fast connections while
roaming, and automatic re-keying of encryption
keys
35Protected EAP ( PEAP )
- PEAP was developed by Microsoft, Cisco and RSA
Security to address deficiencies of EAP
(Unprotected user information during the EAP
negotiation, No support for fast reconnections
when roaming, No support for fragmentation and
reassembly) - PEAP was designed to protect EAP communication
between clients and authenticators - It provides support for identity protection by
using TLS to create an encrypted tunnel after
verifying the identity of authentication server
36Protected EAP (PEAP) continue
- After encrypted tunnel is established a second
EAP authorization process occurs inside the
tunnel - The client is authenticated inside the tunnel
using any implemented EAP authorization type
(tokens, passwords,etc) - It has built-in support for packet fragmentation
and reassembly, as well as fast reconnects - See figure 11.15 on page 263 for PEAP process
37PEAP Authentication
- 1. The client sends an EAP start message to the
access point - 2. The access point replies with an EAP Request
Identity message - 3. The client sends its network access identifier
(NAI), which is its username, to the access point
in an EAP Response message - 4. The access point forwards the NAI to the
RADIUS server encapsulated in a RADIUS Access
Request message - 5. The RADIUS server responds to the client with
its digital certificate - 6. The client validates the RADIUS servers
digital certificate
38PEAP Authentication
- 7. The client and server negotiate and create an
encrypted tunnel - 8. This tunnel provides a secure data path for
client authentication - 9. Using the TLS Record protocol, a new EAP
authentication is initiated by the RADIUS server - 10. The exchange includes the transactions
specific to the EAP type used for client
authentication - 11. The RADIUS server sends the access point a
RADIUS ACCEPT message, including the clients WEP
key, indicating successful authentication
39EAP-TTLS vs PEAP
- Both were designed to use older authentication
methods while maintaining the strong
cryptographic foundation of TLS - Both have similar structure
- Both are two-stage protocols that establish
security in stage one and then exchange
authentication in stage two - Stage one establish a TLS tunnel and
authenticates the authentication server to the
client with a certificate
40EAP-TTLS vs PEAP
- Microsoft and Cisco both support PEAP Ciscos
Aironet Client Utility (ACU) and Windows XP with
service pack1 - There are two types of PEAP supported by
Microsoft PEAP-EAP-MS-CHAPv2 and PEAP-EAP-TLS - PEAP-EAP-TLS, server and client side certificates
are required - PEAP-EAP-MS-CHAPv2, server certificates and
client passwords are required
41EAP Considerations
- The factors to include when deciding
- Mutual Authentication
- Dynamic Key Generation, Rotation, and
Distribution - Cost and Management Overhead
- Acceptance, Standardization, and Support
- Availability and Implementation
42Proprietary Protocols
- These protocols are used because
- Added security through per packet authentication
- Added security through use of leading-edge
encryption algorithms not yet supported by
standards - Added security due to the entire communications
process between client and server being strongly
encrypted - Compression to increase throughput over the
half-duplex medium
43Proprietary Protocols
- Enterprise Encryption Gateways use proprietary
protocols in order to achieve stronger security
and increased throughput, but the main
disadvantage here is vendor interoperability