Internet Draft Florent Bersani File: draft-bersani-eap-psk-01.txt France Telecom R&D Expires: August 2004 February 2004 The EAP PSK Protocol Status of this Memo This document is an Internet-Draft and is in full conformance with all provisions of Section 10 of RFC2026. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html. Copyright Notice Copyright (C) The Internet Society (2003). All Rights Reserved. Abstract This document specifies an Extensible Authentication Protocol (EAP) method for authentication and session key derivation using a pre- shared key (PSK). This PSK is used by a unique underlying cryptographic primitive, a block cipher, which is instantiated with AES-128. EAP-PSK performs mutual authentication and session key derivation. It provides identity protection and shall provide fast reconnect in future versions. It provides a secure communication channel within EAP in case the authentication is successful. This secure channel can be used to allow for instance protected result indications. EAP-PSK is designed to be itself extensible. EAP-PSK is intended to be easy to deploy and well-suited for authentication over insecure networks such as IEEE 802.11. Bersani Expires ­ August 2004 [Page 1] INTERNET-DRAFT EAP PSK February 2004 2004 Table of Contents 1. Introduction...................................................4 1.1 Design goals for EAP-PSK....................................4 1.2 Caveats.....................................................4 1.3 Specification of requirements...............................5 1.4 Terminology.................................................5 1.5 Conventions.................................................7 1.6 Related work................................................7 2. Protocol overview..............................................8 2.1 Cryptographic design of EAP-PSK.............................8 2.1.1 The authentication...................................8 2.1.2 The key derivation...................................9 2.1.3 The protected channel...............................10 2.2 EAP-PSK key hierarchy......................................11 2.2.1 The PSK.............................................11 2.2.2 The TEK.............................................11 2.2.3 The MSK.............................................12 2.2.4 The EMSK............................................12 2.2.5 The IV..............................................12 2.3 EAP-PSK message flow.......................................12 2.3.1 EAP-PSK basic message flow..........................12 2.3.2 EAP-PSK advanced message flows......................13 2.4 Retry behavior.............................................14 2.5 Fragmentation..............................................14 3. EAP-PSK state machines........................................15 3.1 EAP-PSK peer state machine.................................15 3.1.1 The Init state......................................15 3.1.2 The ID_sent state...................................16 3.1.3 The MAC_sent state..................................17 3.1.4 The P_Channel state.................................17 3.1.5 The Failure state...................................18 3.1.6 The Success state...................................18 3.2 EAP-PSK server state machine...............................18 3.2.1 The Init state......................................19 3.2.2 The ID_req state....................................19 3.2.3 The MAC_req state...................................20 3.2.4 The P_channel state.................................21 3.2.5 The Failure state...................................21 3.2.6 The Success state...................................21 4. EAP-PSK message format........................................22 4.1 Table of the Type field of the different attributes........22 4.2 Format of the different attributes.........................22 4.2.1 AT_IDREQ............................................22 4.2.2 AT_IDRES............................................22 4.2.3 AT_MAC..............................................23 4.2.4 AT_PCHANNEL.........................................24 4.2.5 AT_RAND.............................................25 4.2.6 AT_STATUS...........................................25 Bersani Expires ­ August 2004 [Page 2] INTERNET-DRAFT EAP PSK February 2004 2004 5. IANA considerations...........................................26 6. Security considerations.......................................26 6.1 Identity protection........................................26 6.2 Mutual authentication......................................27 6.3 Key derivation.............................................27 6.4 Dictionary attacks.........................................28 6.5 Protected channel..........................................28 6.6 Negotiation attacks........................................28 6.7 Fast reconnect.............................................28 6.8 Man-in-the-middle attacks..................................28 6.9 Generating random numbers..................................29 7. Security claims...............................................29 8. Intellectual Property Right Notice............................29 9. Acknowledgements..............................................30 10. References...................................................30 10.1 Normative.................................................30 10.2 Informative...............................................31 11. Authors' Addresses...........................................33 12. Full Copyright Statement.....................................33 Annex A: Work still to be done on this document..................33 1. Editorial...................................................33 2. Security....................................................34 3. Technical...................................................34 Annex B: Guidance for PSK generation from a password.............35 Bersani Expires ­ August 2004 [Page 3] INTERNET-DRAFT EAP PSK February 2004 2004 1. Introduction 1.1 Design goals for EAP-PSK The Extensible Authentication Protocol, [EAP], provides a standard mechanism for support of additional authentication methods within [PPP]. EAP is also used within IEEE 802 networks through the [IEEE 802.1X] framework. EAP supports many authentication mechanisms usually called EAP methods. This document specifies an EAP method that uses a pre-shared key (PSK). Design goals for this method were: 1. Simplicity: It should be easy to implement and to deploy without any pre-existing infrastructure. 2. Wide applicability: It should be possible to use this method to authenticate over any network. In particular, it should be suitable for [IEEE 802.11] wireless LANs and comply to [IEEE 802REQ] 3. Security: It should be conservative in its cryptographic design and enjoy security proofs 4. Extensibility: It should be possible to add to this method the required extensions as their need appears 5. Patent-avoidance: It should be free of any Intellectual Property Right claims Thus, EAP-PSK relies on a single cryptographic primitive, [AES], and performs mutual authentication and session key derivation. It also provides identity protection and shall provide fast reconnect in future versions. It provides a secure communication channel within EAP in case the authentication is successful. This secure channel can be used to allow for instance protected result indications. It uses a Type-Length-Value design to ensure that it will be easy to extend. This design is quite analogous to the attribute format used in [EAP-SIM] or the TLV suggested in [EAP-TLV]. Some of these TLVs may be encapsulated in a protected channel (which is itself implemented as a dedicated TLV), quite similarly to the ideas presented in [PEAP-TLV]. 1.2 Caveats Since PSK are of frequent use in security protocols, because a PSK simply means a cryptographic key in the symmetric cryptographic setting (see e.g. [HAC] for an introduction to symmetric vs. asymmetric cryptography), attention should be paid not to confuse EAP-PSK with any other protocols that may also refer to a PSK, for instance [WPA] when used in its PSK mode. Bersani Expires ­ August 2004 [Page 4] INTERNET-DRAFT EAP PSK February 2004 2004 EAP-PSK’s PSK should also not be confused with the PSKs possibly used by other protocols relying on PSKs: EAP-PSK’s PSK should be cryptographically separated from any other PSK or else the security of EAP-PSK might be voided. It is a rule of the thumb in cryptography to use different keys for different applications. The generation of the PSK used by EAP-PSK is outside of the scope of this document. The PSK SHOULD be generated by a good source of randomness (see [RFC 1750]). In particular, a PSK should not be confused with a password. However, in case one wants to generate the PSK from a password, although this is strongly discouraged, guidance to do so is provided in annex B. The definition of the repository of the PSK used by EAP-PSK is also outside of the scope of this document. In particular, nothing prevents from storing EAP-PSK's PSK on a tamper-resistant device such as a smart card rather than having it memorized or written down on a sheet of paper. Indeed such a storage on a smart card might allow to choose stronger credentials and to avoid their duplication (that is to say, for instance, preventing a user from easily sharing its PSK with somebody else). 1.3 Specification of requirements The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [RFC 2219]. 1.4 Terminology TBC This document frequently uses the following terms: AES Advanced Encryption Standard, a block cipher, please refer to [AES] for more details Backend Authentication Server A backend authentication server is an entity that provides an authentication service to an authenticator. When used, this server typically executes EAP Methods for the Authenticator. [This terminology is also used in [IEEE 802.1X.].] EAP Authenticator (or simply Authenticator) The end of the EAP link initiating the EAP authentication methods. [Note: This terminology is also used in Bersani Expires ­ August 2004 [Page 5] INTERNET-DRAFT EAP PSK February 2004 2004 [IEEE 802.1X.], and has the same meaning in this document]. EAP Peer (or simply Peer) The end of the EAP Link that responds to the authenticator. [Note: In [IEEE 802.1X], this end is known as the Supplicant.] EAP Server (or simply Server) The entity that terminates the EAP authentication with the peer. In the case where there is no Backend Authentication Server, this term refers to the EAP Authenticator. Where the EAP Authenticator operates in pass-through, it refers to the Backend Authentication Server. MAC Message Authentication Code. Informally, the purpose of a MAC is to facilitate, without the use of any additional mechanisms, assurances regarding both the source of a message and its integrity, please refer to [HAC] for more details MAP Mutual Authentication Protocol. The name of a family of protocols that allow mutual authentication and that are specified in [EAKD] OMAC One Key CBC-MAC, a method to generate a message authentication code, please refer to [OMAC] for more details PSK Pre-shared key. A pre-shared key is a cryptographic key in the symmetric setting that is derived by some out-of-band mechanism. It is merely a sequence of binary digits of given length that should have been chosen at random. PRF Pseudo-random function. Please refer for instance to [EAX] for a precise cryptographic definition of this term PRP Pseudo-random permutation. Please refer for instance to [EAX] for a precise cryptographic definition of this term Temporary Identity A temporary identity, contrary to a permanent identity, is Bersani Expires ­ August 2004 [Page 6] INTERNET-DRAFT EAP PSK February 2004 2004 a NAI that the user uses a pseudonym for one EAP dialog. This features is used for identity protection. It is the server's responsibility to provide the peer with temporary identities in case identity protection is required. 1.5 Conventions All numbers involved in cryptographic calculations are considered in network-byte order. || denotes concatenation of strings. [String] denotes the MAC of String calculated as specified by the context. ** denotes integer exponentiation. “i” denotes the unsigned binary representation on 128 bits of the integer I in network byte order. Therefore this notation only makes sense when i is between 0 and 2**128-1. 1.6 Related work There exist other EAP methods which also somehow rely on PSKs: [EAP-SIM] and [EAP-AKA] are two of them, which do not directly compete with EAP-PSK since they are designed to take advantage of the GSM and UMTS infrastructures. There are therefore not easy to deploy in case one does not dispose of such infrastructures. EAP-MD5, described in [EAP], is another one, which has however been deprecated for security reasons: it is not safe to use it over insecure networks. EAP-OTP, described in [EAP], and other One-Time Password methods such as [EAP-SecurID] may also rely on PSKs but they also do not directly compete with EAP-PSK since they require additional elements (e.g. the One-Time password generator and server) on the client and on the server side. [LEAP] (Lightweight EAP) also known as EAP-Cisco Wireless is quite similar to EAP-PSK. However it is a proprietary protocol that has been shown to bear cryptographic weaknesses (see for instance [LEAPVUL]). [EAP-SKE] is quite similar to EAP-PSK. This method is still work in progress and has put emphasis on network efficiency in roaming situations. Work could be done to merge EAP-PSK and EAP-SKE. Bersani Expires ­ August 2004 [Page 7] INTERNET-DRAFT EAP PSK February 2004 2004 [EAP-Archie] is very similar to EAP-PSK. This method is still work in progress and has very much inspired this work. EAP-PSK makes amendments to the cryptographic parts specified in EAP-Archie and provides the protected channel and the TLV approach as new features. Work could be done to merge EAP-PSK and EAP-Archie. [EAP-SRP] is quite different of EAP-PSK except since it uses both symmetric and asymmetric cryptography and it is subject to IPR claims by Stanford university. 2. Protocol overview 2.1 Cryptographic design of EAP-PSK EAP-PSK rely on a single cryptographic primitive, a block cipher, which is instantiated with AES-128. This instantiation has been chosen because: 1. AES-128 is standardized and its implementations are widely available 2. AES-128 has been carefully reviewed by the community and is believed to be secure For a description of AES-128, please refer to [AES]. However, in case it should be needed, new instantiations of EAP-PSK could easily be proposed as it does not intricately depend on the chosen block cipher. The only parameters of the block cipher that EAP-PSK depends on are its block size (128 bits) and its key size (128 bits). EAP-PSK uses three cryptographic parts: 1. An authentication protocol to mutually authenticate the communicating parties, that follows the design presented in [EAKD] under the name MAP1 instantiated with [OMAC] and [AES] 2. A key derivation protocol to derive keying material according to the EAP Key Management Framework [EKMF], that uses the modified counter mode presented in [SOBMMO] 3. An authenticated encryption protocol with associated data to provide a protected channel for both mutually authenticated parties to communicate securely within the method, that uses the [EAX] mode of operation 2.1.1 The authentication The authentication protocol used by EAP-PSK is the Mutual Authentication Protocol 1. The Mutual Authentication Protocol 1 (MAP1) is described in [EAKD]. Bersani Expires ­ August 2004 [Page 8] INTERNET-DRAFT EAP PSK February 2004 2004 It consists of a one and half round trip exchange: Bob Alice | | | RA | |<---------------------------------------------------------| | | | [B||A||RA||RB] | |--------------------------------------------------------->| | | | [A||RB] | |<---------------------------------------------------------| where: 1. RA and RB are random numbers chosen respectively by Alice and Bob 2. A and B are Alice's and Bob's respective identities EAP-PSK instantiates this protocol with: 1. RA and RB 128 bit random numbers chosen respectively by Alice and Bob 2. A and B are Alice's and Bob’s respective permanent full NAIs. It was chosen to include their permanent full NAIs in the calculation of the MAC (and not the temporary NAIs they may be using in case identity protection is enabled) to avoid getting out of the setting of the security proof of [EAKD]. Choosing to use the permanent full NAIs may possibly endanger identity protection (though there doesn't seem to be any trivial attack taking advantage of this) but since mutual authentication has a security proof and identity protection currently does not, it seems reasonable to preserve the existing security proof. [EAKD] was chosen for historical reasons (since EAP-PSK was very much influenced by EAP-Archie, please refer to section 9) and because this scheme enjoyed a security proof. 3. The MAC algorithm we use is OMAC1 with AES-128 using a 128-bit portion of the PSK called AK (see section 2.2) and producing a tag length of 128 bits. OMAC was chosen because of its handling of arbitrary length messages and its design simplicity (though some care must be taken in its implementation to avoid side channel attacks, please refer to [EAX]). It also enjoys a security proof, which has so far not been found to be flawed and is believed to have been extensively reviewed by the cryptographic community. 2.1.2 The key derivation The key derivation is realized using the modified counter mode. Bersani Expires ­ August 2004 [Page 9] INTERNET-DRAFT EAP PSK February 2004 2004 The modified counter mode is described in [SOBMMO]. This mode was chosen because it seems to be one of the rare simple key derivation schemes that relies on a block cipher and has a proof of its security. EAP-PSK instantiates this modified counter mode with all rotation values (the ri following [SOBMMO] Figure 3 notation) taken equal to zero (no rotations) and the counter values (the ci following [SOBMMO] Figure 3 notation) taken respectively equal to the first t integers (that is ci=i, starting with i=1). The parameter t is taken equal to 9. The underlying block cipher used by this counter mode is AES-128. The input block to the key derivation is taken to be: [B||A||RA|RB] as per section 2.1.1 notation. For what regards the output blocks: 1. The first output block of the key derivation is taken to be the TEK 2. The second to fifth output blocks of the key derivation are taken to be the MSK 3. The sixth to ninth output blocks of the key derivation are taken to be the EMSK 2.1.3 The protected channel To provide a protected channel within EAP-PSK in case of a successful authentication, EAP-PSK uses the EAX mode of operation described in [EAX]. EAX is instantiated with AES-128 as the underlying block cipher keyed with the TEK. EAX is instantiated within EAP-PSK with a tag length of 128 bits. The nonce N used by EAX (following [EAX] Figure 3 notation) is a counter starting with “0” and incremented by the one at each subsequent EAP-PSK message (except retransmissions of course) within one EAP-PSK dialog. Thus, N can be considered a replay counter. The message M used by EAX (following [EAX] Figure 3 notation) consists of the message that one party wishes to send to the other over the protected channel (i.e a concatenation of EAP-PSK attributes). Bersani Expires ­ August 2004 [Page 10] INTERNET-DRAFT EAP PSK February 2004 2004 The header H used by EAX (following [EAX] Figure 3 notation) is currently unused and thus taken to be the empty string. EAX was mainly chosen because it strongly relied on OMAC in its design and OMAC had already been chosen in EAP-PSK for the authentication part. It was also chosen because of the simplicity of its design and the security proof it enjoys. It should however be understood that there are currently many other proposed modes for authenticated encryption with associated data (including IPR free ones, like CCM, CWC or GCM, please refer to [MOSKBC] for more details) and that the complexity and novelty of EAX security proof may be a concern. 2.2 EAP-PSK key hierarchy This section instantiates the EAP Key hierarchy described in [EKMF] for EAP-PSK. 2.2.1 The PSK EAP-PSK uses a long-lived 256-bit secret shared between the EAP Peer and the EAP Server called the PSK. The PSK has an internal structure. It consists of two 128-bit subkeys, respectively called the authentication key (AK) and the key- derivation key (KDK). The protocol uses the AK to mutually authenticate the EAP Peer and the EAP Server. The protocol uses the KDK to derive keying material between the EAP Peer and the EAP Server. It should be emphasized that EAP-PSK assumes that AK and KDK are cryptographically separated and that its security proof rely on this assumption. In case, EAP-PSK's PSK is drawn at random from the set of possible keys, this assumption is verified. EAP-PSK assumes that the PSK is known only to the EAP Peer and EAP Server, and the security properties of the protocol may be compromised if it has wider distribution. The protocol also assumes the EAP Server and EAP Peer identify the correct PSK to use with the other by their respective [NAI]s. 2.2.2 The TEK EAP-PSK allows for TEK derivation from a MAC exchanged during authentication and the KDK. The TEK is a 128 bit key that is used to set a protected channel for both mutually authenticated parties to communicate securely within the method. Bersani Expires ­ August 2004 [Page 11] INTERNET-DRAFT EAP PSK February 2004 2004 2.2.3 The MSK EAP-PSK allows for MSK derivation from a MAC exchanged during authentication and the KDK. As specified in [EKMF], the MSK is 64 bytes long. 2.2.4 The EMSK EAP-PSK allows for EMSK derivation from a MAC exchanged during authentication and the KDK. As specified in [EKMF], the EMSK is 64 bytes long. 2.2.5 The IV EAP-PSK does not derive any IV. 2.3 EAP-PSK message flow 2.3.1 EAP-PSK basic message flow Basically, EAP-PSK is comprised of six messages (i.e. three round trips): 1. The first message is sent by the server to the peer to request its identity 2. The second message is sent by the peer to the server to answer the identity request: the peer indicates its identity (temporary or permanent) to the server 3. The third message is sent by the server to the peer and starts the mutual authentication procedure: it essentially consists of a random value chosen by the server 4. The fourth message is sent by the peer to the server and contains a random value chosen by the peer and an authentication tag calculated by the peer over both random values as well as the peer and server’s permanent full NAIs to prove the identity of the peer to the server 5. The fifth message is sent by the server to the peer and contains an authentication tag calculated over the random value chosen by the peer and the server’s permanent full NAI to prove the identity of the server to the peer. This message may also contain data encapsulated in a protected channel that has just been set up as a result of the authentication procedure. This data at least serves to make sure that the protected channel was correctly set up and to give indications on the probable future decision of the server. It may also be used to give to the peer the next temporary identity it should use. Bersani Expires ­ August 2004 [Page 12] INTERNET-DRAFT EAP PSK February 2004 2004 6. The sixth message is sent by the peer to the server and may also contain data encapsulated in a protected channel that has just been set up as a result of the authentication procedure Peer Server | EAP-PSK/AT_IDREQ | |<---------------------------------------------------------| | | | EAP-PSK/AT_IDRES | |--------------------------------------------------------->| | | | EAP-PSK/AT_Rand | |<---------------------------------------------------------| | | | EAP-PSK/AT_Rand, AT_MAC | |--------------------------------------------------------->| | | | EAP-PSK/AT_MAC, AT_PCHANNEL | |<---------------------------------------------------------| | | | EAP-PSK/AT_PCHANNEL | |--------------------------------------------------------->| | | This basic message flow could be comprised of only three messages, were it not the request/response nature of EAP that prevents the third message to be the last one. We take advantage of this situation by mandating the setup of a protected channel. The basic message flow also include a fresh query by the server of the peer's identity, since the current revision of EAP ([EAPbis]) recommends to do so in its section 5.1. 2.3.2 EAP-PSK advanced message flows EAP-PSK provides different advanced features that may lead to different message flows: 1. Identity protection. This feature might add an additional roundtrip to the basic message flow when used. This additional round trip is inserted after the second message in the basic flow when the peer and the server fall out of synchronization on the pseudonym the peer uses to protect its identity. Hence, when the server does not recognize the identity provided by the peer in response to EAP-PSK/AT_IDREQ, the server sends a message to the peer asking it to disclose its permanent identity. The peer MAY respond to this message if its security policy allows him to do so by sending its permanent identity or by resending its Bersani Expires ­ August 2004 [Page 13] INTERNET-DRAFT EAP PSK February 2004 2004 temporary identity or it MAY not respond to this request. The basic message flow then proceeds. 2. Fast reconnect. TBC 3. Conversation over the protected channel. This feature might add some round trips to the basic message flow when used. These additional round trips are inserted after the sixth message in the basic message flow. They consist in exchanging data between the peer and the server over the protected channel that has been set up thanks to the authentication. This protected data exchange might for instance be of some use if the peer’s account is pre paid and is charged on a per packet or temporal basis: in case the peer wants to top it up, it can do so, e.g. by making a financial transaction with the server. This protected data exchange might also be used to check the identity of the claimed Authenticator that the peer has connected to. In each message exchanged over the protected channel, both parties indicate whether the exchange has to continue, may continue or is done for them and what their current decision is (unconditional success, conditional success or failure). These indications mimic the variables presented in [EAP-SM] 2.4 Retry behavior EAP-PSK complies to the EAP retry behavior described in [EAP], that is: the EAP Server is responsible for retry behavior. This means that if the EAP Server does not receive a reply from the peer, it MUST resend the EAP-Request for which it has not yet received an EAP- Response. However, the EAP Peer MUST NOT resend EAP-Response messages without first being prompted by the EAP Server. As a result, it is possible that an EAP Peer will receive duplicate EAP-Request messages, and may send duplicate EAP-Responses. Both the EAP Peer and the EAP Server should be engineered to handle this possibility. 2.5 Fragmentation EAP-PSK does not support fragmentation or reassembly. Therefore it is to be used either over networks which MTU is enough to convey encapsulated EAP-PSK packets without fragmentation or encapsulated in other protocols which take care of fragmentation and reassembly. However, in case it is needed, fragmentation support might well be provided for communication over the protected channel thanks to, for instance, an AT_Fragment attribute. Bersani Expires ­ August 2004 [Page 14] INTERNET-DRAFT EAP PSK February 2004 2004 3. EAP-PSK state machines This section is TBC The state diagrams given below are simplistic. Their goal is to give a compact description of EAP-PSK. Due to its TLV design (i.e. attribute usage), this document only specifies the attributes that MUST be included in the requests or responses. Though it does not forbid adding other non-mandatory attributes, the peer and the server SHOULD for now ignore any non- mandatory attribute that are not encapsulated in the AT_PCHANNEL attribute. 3.1 EAP-PSK peer state machine +-----------+ +-----------+ +-----------+ | | | |<--+ | | --->| Init |--->| ID_sent | | | Failure |<----+ | | | |---+-->| | | +-----------+ +-----------+ | +-----------+ | | | | | +-----------+ | | | |<--+ | +--------------------| MAC_sent | | | | |-------------------------+ | +-----------+ | | | +--------------------------+-------------------------------+ | | | +-----------+ | +-----------+ | | | | | | +-->| P_Channel |----------+------------>| Success | | | | | +-----------+ +-----------+ 3.1.1 The Init state The peer starts in the Init state: in this state it awaits the first EAP-PSK request of the EAP server. This first request MUST be an EAP- PSK packet containing at least one attribute: the AT_IDREQ attribute. The peer MUST silently discard any EAP-PSK packet that does not contain the AT_IDREQ attribute. If identity protection is a concern, the server SHOULD NOT request the peer's permanent identity. Upon reception of an EAP-PSK packet that contains the AT_IDREQ attribute, depending on the flag set in this attribute, the Bersani Expires ­ August 2004 [Page 15] INTERNET-DRAFT EAP PSK February 2004 2004 identities the peer currently has and its security policy the peer has three options: 1. Provide the requested identity if it is able to do so (e.g. if it hasn't lost its current temporary identity) and if its security policy doesn't prevent him from doing so (e.g. if the server directly request the peer's permanent identity, which the peer is configured never to disclose) 2. Provide its permanent identity if its security policy allows him to do so, whatever the identity requested by the server was 3. Silently discard the received packet if the peer cannot or doesn't want to answer until either the peer or the server fails (for instance due to time out or retry counter or retry counter limit) or it receives a packet it can and wants to answer to. If the peer chooses option #1 or option #2, it sends the corresponding response (i.e. an EAP-PSK packet that MUST contain the AT_IDRES attribute) and its EAP-PSK method state moves to ID_send 3.1.2 The ID_sent state In this state the peer MAY receive two different types of EAP-PSK packets: 1. An EAP-PSK packet that contains an AT_IDREQ attribute 2. An EAP-PSK packet that contains an AT_RAND attribute For now the two situations described above SHOULD be considered to be mutually exclusive (i.e. if the peer receives an EAP-PSK packet that contains both AT_IDREQ and AT_RAND attributes), it SHOULD silently discard it. However, this point, as much of this document is open to discussion and may evolve (see Annex A). In case it receives an EAP-PSK packet that contains an AT_IDREQ attribute, the peer has the same options as when it received an EAP- PSK packet containing and AT_IDREQ attribute while in the Init state. The only difference, is that the peer knows that the identity it has already sent seems not to have been recognized by the server. [Note: the state machine diagram however indicates that in case the peer wants then to take option #3 of the Init state while in the ID_sent state, it also can go directly to the Failure state without waiting for time outs or retry counter limits, this behavior is open to discussion, see annex A]. In this situation the peer either stays in the ID_sent state or moves to the Failure state. In case it receives an EAP-PSK packet that contains an AT_RAND attribute, the peer builds an EAP-PSK packet in response that MUST contain an AT_MAC attribute and an AT_RAND attribute and moves on to the MAC_sent state. To build this AT_MAC, it begins by choosing himself a 128 random value, which will be the value of the AT_RAND attribute that the peer MUST include in its response. The value of Bersani Expires ­ August 2004 [Page 16] INTERNET-DRAFT EAP PSK February 2004 2004 the AT_MAC attribute is then computed according to the following formula (please see also section 2.1.1): [B||A||RA||RB] where: o B stands for the ASCII encoding of the permanent full NAI of the peer o A stands for the ASCII encoding of the permanent full NAI of the server o RA stands for the value of the AT_RAND the peer has just received o RB stands for the random value the peer has just chosen to build the AT_RAND it must include in its response 3.1.3 The MAC_sent state In this state, the peer MUST receive an EAP-PSK packet that contains an AT_MAC attribute and an AT_PCHANNEL attribute. Upon reception of an EAP-PSK packet that contains an AT_MAC attribute and an AT_PCHANNEL attribute, the peer MUST start by processing the AT_MAC attribute. It MUST verify that the value of this attribute is equal to the value given by the following formula (please see also section 2.1.1): [A||RB] where: o A stands for the ASCII encoding of the permanent full NAI of the server o RB stands for the value of the AT_RAND attribute that the peer has just sent to the server In case, this verification succeeds the peer moves to the P_channel state to process the AT_PCHANNEL attribute. In case, it fails the peer moves to the Failure state. 3.1.4 The P_Channel state In this state, the peer only processes EAP-PSK messages that contain an AT_PCHANNEL attribute. Furthermore, the AT_PCHANNEL MUST be the only attribute that the peer processes. While processing an AT_PCHANNEL attribute, the peer MUST first check the replay counter (see section 2.1.3). If the replay counter verifies (that is to say is equal to the value of the last counter sent incremented by one starting with a counter value of 0), the peer then decrypts the value of the attribute AT_PCHANNEL and processes Bersani Expires ­ August 2004 [Page 17] INTERNET-DRAFT EAP PSK February 2004 2004 the attributes encapsulated within. Any EAP-PSK packet that fails to verify the replay counter or decrypt the value of the AT_PCHANNEL MUST be discarded without any further processing. The AT_PCHANNEL attribute MUST contain an AT_STATUS attribute that both indicates (adopting EAP state machine terminology, please refer to [EAP-SM]): o The server's current decision regarding the peer (FAIL, COND_SUCC, UNCOND_SUCC) o The server's current decision regarding the continuation of the method (CONT, MAY_CONT, DONE) If the peer receives a valid AT_PCHANNEL attribute, it must answer with an EAP-PSK packet that also contains a valid AT_PCHANNEL attribute. While in this state the peer, depending on its policy, its current STATUS variables and the STATUS variables it receives from the server, o Stay in this state o Advance to Failure state o Advance to Success state 3.1.5 The Failure state In this state, the peer has decided that the current EAP-PSK dialog has failed and MUST discard any incoming EAP-PSK packet corresponding to that dialog. It waits for time outs, retry counter limits or an EAP-Failure packet. This state is an EAP-PSK state and must not be confused with a possible EAP state (e.g. Failure state in Figure 3 of [EAP-SM]). 3.1.6 The Success state In this state, the peer has decided that the current EAP-PSK dialog has succeeded and MUST discard any incoming EAP-PSK packet corresponding to that dialog. This state is an EAP-PSK state and must not be confused with a possible EAP state (e.g. Success state in Figure 3 of [EAP-SM]). 3.2 EAP-PSK server state machine Bersani Expires ­ August 2004 [Page 18] INTERNET-DRAFT EAP PSK February 2004 2004 +-----------+ +-----------+ +-----------+ | | | |<--+ | | --->| Init |--->| ID_req | | | Failure |<----+ | | | |---+-->| | | +-----------+ +-----------+ | +-----------+ | | | | | +-----------+ | | | |<--+ | +--------------------| MAC_req | | | | |-------------------------+ | +-----------+ | | | +--------------------------+-------------------------------+ | | | +-----------+ | +-----------+ | | | | | | +-->| P_Channel |----------+------------>| Success | | | | | +-----------+ +-----------+ 3.2.1 The Init state The server starts in the Init state: in this state, it builds an EAP- PSK packet that contains an AT_IDREQ attribute. It sets the flags of the AT_IDREQ (see section 4) according to its policy. To provide identity protection, the server SHOULD not set any of these flags in the first request it issues to the peer, that is to say, that the server SHOULD accept temporary identities. After sending this EAP-PSK packet, the server advances to the ID_req state. 3.2.2 The ID_req state In this state, the server waits for the response of the peer. The response of the peer MUST be an EAP-PSK packet that contains an AT_IDRES attribute. If the server does not recognize the identity sent by the peer, it can either: o Send another EAP-PSK packet that contains an AT_IDREQ attribute, with flag settings that MAY differ from the first EAP-PSK packet containing an AT_IDREQ attribute that was issued o Move to Failure state If the server ever issues an EAP-PSK packet with the permanent identity flag of the AT_IDREQ attribute set, it MUST send this attribute again within the EAP-PSK protected channel at the end of Bersani Expires ­ August 2004 [Page 19] INTERNET-DRAFT EAP PSK February 2004 2004 the EAP-PSK dialog to allow the detection of attacks against the identity protection. If the server recognizes the identity sent by the peer, it chooses at random a 128 bit value and sends an EAP-PSK packet that MUST contain an AT_RAND which value is the random number it has just chosen. It then moves to the MAC_req state. 3.2.3 The MAC_req state In this state, the server waits for an EAP-PSK packet that contains an AT_MAC and an AT_RAND attribute. Upon reception of an EAP-PSK packet that contains an AT_MAC attribute and an AT_RAND attribute, the server MUST verify that the value of the AT_MAC attribute is equal to the value given by the following formula (please see also section 2.1.1): [B||A||RA||RB] where: o B stands for the ASCII encoding of the permanent full NAI of the peer o A stands for the ASCII encoding of the permanent full NAI of the server o RA stands for the value of the AT_RAND the server sent before entering the MAC_req state o RB stands for the value of the AT_RAND the server has just received In case, this verification succeeds the server send an EAP-PSK that contains an AT_MAC and an AT_PCHANNEL attribute and moves to the P_channel state. The value of the AT_MAC attribute sent by the server is given by the following formula (please see also section 2.1.1): [A||RB] where: o A stands for the ASCII encoding of the permanent full NAI of the server o RB stands for the value of the AT_RAND attribute that the peer has just sent to the server The AT_PCHANNEL attribute MUST contain an AT_STATUS attribute and is calculated according to the procedure specified in section 2.1.3 and 4. If the server and the peer care about identity protection, it SHOULD also contain an AT_IDRES attribute which value gives the peer its next temporary identity to use with EAP-PSK. Bersani Expires ­ August 2004 [Page 20] INTERNET-DRAFT EAP PSK February 2004 2004 In case, the verification of the AT_MAC sent by the peer fails the server moves to the Failure state. 3.2.4 The P_channel state In this state, the server only processes EAP-PSK messages that contain an AT_PCHANNEL attribute. Furthermore, the AT_PCHANNEL MUST be the only attribute that the server processes. While processing an AT_PCHANNEL attribute, the server MUST first check the replay counter (see section 2.1.3). If the replay counter verifies (that is to say is equal to the value of the last counter sent incremented by one starting with a counter value of 0), the server then decrypts the value of the attribute AT_PCHANNEL and processes the attributes encapsulated within. Any EAP-PSK packet that fails to verify the replay counter or decrypt the value of the AT_PCHANNEL MUST be discarded without any further processing. The AT_PCHANNEL attribute MUST contain an AT_STATUS attribute that both indicates (adopting EAP state machine terminology, please refer to [EAP-SM]): o The server's current decision regarding the peer (FAIL, COND_SUCC, UNCOND_SUCC) o The server's current decision regarding the continuation of the method (CONT, MAY_CONT, DONE) While in this state the server, depending on its policy, its current STATUS variables and the STATUS variables it receives from the peer, o Stay in this state o Advance to Failure state o Advance to Success state 3.2.5 The Failure state In this state, the server has decided that the current EAP-PSK dialog has failed and MUST discard any incoming EAP-PSK packet corresponding to that dialog. The logical next step is the server sending an EAP Failure packet. This state is an EAP-PSK state and must not be confused with a possible EAP state (e.g. Failure state in Figure 4 of [EAP-SM]). 3.2.6 The Success state In this state, the server has decided that the current EAP-PSK dialog has succeeded and MUST discard any incoming EAP-PSK packet corresponding to that dialog. This state indicates to EAP that EAP- PSK has succeeded. This state is an EAP-PSK state and must not be confused with a possible EAP state (e.g. Success state in Figure 4 of [EAP-SM]). Bersani Expires ­ August 2004 [Page 21] INTERNET-DRAFT EAP PSK February 2004 2004 4. EAP-PSK message format 4.1 Table of the Type field of the different attributes TBC Attribute Name Type Field Value AT_IDREQ 1 AT_IDRES 2 AT_RAND 3 AT_MAC 4 AT_PCHANNEL 5 AT_STATUS 6 4.2 Format of the different attributes This section presents the respective formats of the different attributes listed in alphabetical order. Within the different attribute formats, reserved bytes are specified. These reserved bytes are essentially here for word alignment on 32 bit boundaries. They are set to zero when sending and ignored on reception. 4.2.1 AT_IDREQ The format of the AT_IDREQ attribute is shown below. 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | AT_IDREQ | Length = 1 |P| Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ The flag P is set to one when the permanent identity of the peer is requested and to 0 otherwise. The use of the AT_IDREQ is defined in section 3. 4.2.2 AT_IDRES The format of the AT_IDRES attribute is shown below. Bersani Expires ­ August 2004 [Page 22] INTERNET-DRAFT EAP PSK February 2004 2004 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | AT_IDRES | Length | Actual Identity Length | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | : Identity : : . : | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ The use of the AT_IDRES is defined in Section 3. The value field of this attribute begins with 2-byte actual identity length, which specifies the length of the identity in bytes. This field is followed by the subscriber identity of the indicated actual length. The identity does not include any terminating null characters. Because the length of the attribute must be a multiple of 4 bytes, the sender pads the identity with zero bytes when necessary. 4.2.3 AT_MAC The AT_MAC attribute is used for authentication within EAP-PSK. The value field of the AT_MAC attribute contains two reserved bytes followed by a 128 bit keyed message authentication code (MAC). The MAC is calculated over message-specific data. The contents of the message-specific data that are MACed are specified separately for each EAP/PSK message in Section 2.1.1. The format of the AT_MAC attribute is shown below. 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | AT_MAC | Length = 5 | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | | MAC | | | | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ When the AT_MAC attribute is expected to be included in an EAP-PSK message, the recipient MUST process the AT_MAC attribute before Bersani Expires ­ August 2004 [Page 23] INTERNET-DRAFT EAP PSK February 2004 2004 looking at any other attributes (apart from the attributes that may be necessary to the verification of the MAC i.e AT_RANDs). If the message authentication code is absent or invalid, then the recipient MUST ignore all other attributes in the message and operate as specified in Section 3. 4.2.4 AT_PCHANNEL AT_PCHANNEL_is used to transmit information between the peer and server over a protected channel, that is to say a channel that provides confidentiality, data origin authentication and replay protection. The format of the AT_PCHANNEL attribute is shown below. 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | AT_PCHANNEL | Length | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | | Nonce | | | | | |---------------------------------------------------------------| | | | Tag | | | | | |---------------------------------------------------------------| | | | Payload | : : : : | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ The value field of the AT_PCHANNEL attribute consists of one reserved bytes followed by a 4 byte nonce, a 4 byte tag and a variable length payload. The nonce is calculated as specified in section 2.1.3. The tag is 128 bit long. It is calculated as specified in 2.1.3 and in [EAX]. The payload consists in the cipher text resulting from the encryption in the EAX mode of operation of the information that the peer and the Bersani Expires ­ August 2004 [Page 24] INTERNET-DRAFT EAP PSK February 2004 2004 server wish to exchange over the protected channel under Nonce and the derived TEK (see Section 2.1). The information that the peer and the server may exchange over the protected channel consists of a concatenation of EAP-PSK attributes. 4.2.5 AT_RAND The format of the AT_RAND attribute is shown below. 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | AT_RAND | Length = 5 | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | | RAND | | | | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ The value field of the AT_RAND attribute contains two reserved bytes followed by a 16 byte random number generated either by the peer or the server freshly for this EAP-PSK authentication exchange. The random numbers are used as a mean to authenticate and indirectly as a seed value for the new keying material. The server MUST choose a fresh RAND value and send it to the peer at the beginning of the EAP-PSK exchange. The peer MUST also choose a fresh RAND value and send it to the server at the beginning of the EAP-PSK exchange. The randomness of the RAND values are critical for the security of EAP-PSK. 4.2.6 AT_STATUS The format of the AT_STATUS attribute is shown below. 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | AT_STATUS | Length |Dec|Con| Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ There MUST be exactly one AT_STATUS encapsulated in each AT_PCHANNEL. The two bit Dec flag is set to: o 00 if the current decision is FAIL o 01 if the current decision is COND_SUCC o 10 if the current decision is UNCOND_SUCC Bersani Expires ­ August 2004 [Page 25] INTERNET-DRAFT EAP PSK February 2004 2004 The two bit Con flag is set to: o 00 if the current continuation state is CONT o 01 if the current continuation state is MAY_CONT o 10 if the current continuation state is DONE 5. IANA considerations This document introduces one new IANA consideration. It requires IANA to allocate a new EAP Type for EAP-PSK. 6. Security considerations The EAP base protocol [EAPbis] highlights several attacks that are possible against the EAP protocol as there is no inherent security mechanisms provided. This section discusses the claimed security properties of EAP-PSK as well as vulnerabilities and security recommendations. 6.1 Identity protection EAP-PSK includes optional identity privacy support that protects the privacy of the peer identity against passive eavesdropping. The peer and the server SHOULD store the pseudonym in a non-volatile memory so that it can be maintained across reboots. An active attacker that impersonates the server may use the AT-IDREQ attribute to attempt to learn the peer's permanent identity by setting the permanent identity flag. However, it is a matter of policy for the peer to accept to respond to such requests or not: the peer can refuse to send its permanent identity if it believes that the server should be able to recognize its temporary identity. If identity protection is a concern then the peer MUST NOT send its permanent identity. Any other policy allows identity protection compromise. If the peer and server cannot guarantee that the pseudonym will be maintained reliably and identity privacy is required then additional protection from an external security mechanism such as Protected Extensible Authentication Protocol (PEAP) [PEAP] may be used. If an external security mechanism is in use the identity privacy features of EAP-PSK may not be useful. The security considerations of using an external security mechanism with EAP-PSK are beyond the scope of this document. Bersani Expires ­ August 2004 [Page 26] INTERNET-DRAFT EAP PSK February 2004 2004 It should also be kept in mind that the layers below EAP-PSK may disclose elements that can lead to identity protection compromise (e.g. the peer’s [IEEE 802.3] Medium Access Control Address). 6.2 Mutual authentication EAP-PSK provides mutual authentication. The server believes the peer is authentic because it can calculate a valid MAC and the peer believes that the server is authentic because it can calculate a correct MAC. The authentication protocol used in EAP-PSK, MAP1, enjoys a security proof in the provable security paradigm, see [EAKD]. The MAC algorithm used in the instantiation of MAP1 within EAP-PSK, OMAC1, also enjoys a security proof in the provable security paradigm, see [OMAC]. The underlying block cipher used, AES-128, is widely believed to be a secure block cipher. Finally, the key used for mutual authentication, AK, is only used for that purpose, making thus this part cryptographically independent of the other parts. 6.3 Key derivation EAP-PSK supports key derivation. The key hierarchy is specified in Section 2.2. The mechanism used for key derivation is the modified counter mode. The instantiation of the modified counter in EAP-PSK (i.e. the selected ri and ci) comply with the conditions stated in [SOBMMO] so that the security proof in the provable security paradigm of [SOBMMO] holds. The underlying block cipher used, AES-128, is widely believed to be a secure block cipher. The key derivation mechanism uses a dedicated key, the KDK. The input block to the key derivation is taken to be: [B||A||RA|RB]. The reasons that motivated such a choice are: o This value is exactly 128 bit long since the tag length chosen for OMAC is 128 bits Bersani Expires ­ August 2004 [Page 27] INTERNET-DRAFT EAP PSK February 2004 2004 o This value is believed to be fresh for each new EAP-PSK dialog since it depends on RA and RB which are two values chosen at random by the peer and the server through OMAC which is believed to be a PRF o This value is believed not to be under the control either of the peer or the server since it is produced by OMAC which is believed to be a PRF and it takes into account both the inputs of the peer and the server The key derivation scheme is believed to be secure since the modified counter mode is proved to be a PRF if the underlying block cipher is a secure PRP, hence the output blocks are cryptographically separated. 6.4 Dictionary attacks Because EAP-PSK is not a password protocol, it is not vulnerable to dictionary attacks: EAP-PSK’s PSK MUST NOT be derived from a password. Derivation of EAP-PSK’s PSK may lead to dictionary attacks. In case, EAP-PSK’s PSK is however derived from a password, guidance is provided in Annex B how to do so. It is also believed that MAP1 makes dictionary attacks harder, since the first value that is MACed should not be predictable neither for the peer nor the server or anybody else. 6.5 Protected channel EAP-PSK provides a protected channel over which the peer and the server can securely exchange information, in case of a successful authentication. This protected channel provides confidentiality, data origin authentication, replay protection and confirmation of the end of the conversation. 6.6 Negotiation attacks EAP-PSK does not protect from negotiation attacks since it currently does not provide version negotiation as only one version is specified. 6.7 Fast reconnect EAP-PSK shall provide fast reconnect. TBC. 6.8 Man-in-the-middle attacks Bersani Expires ­ August 2004 [Page 28] INTERNET-DRAFT EAP PSK February 2004 2004 Due to the use of symmetric cryptography and the security proofs of its cryptographic components, EAP-PSK is believed not to be vulnerable to man-in-the-middle attacks. There are man-in-the-middle attacks associated with the use of any EAP method within a tunneled protocol such as PEAP, or within a sequence of EAP methods followed by each other (see [MTAP]). This specification does not address these attacks. If EAP-PSK is used with a tunneling protocol or as part of a sequence of methods, there should be cryptographic binding provided between the protocols and EAP-PSK to prevent man-in-the-middle attacks. However the mechanism how the binding is provided is beyond the scope of this document. 6.9 Generating random numbers An EAP-PSK implementation SHOULD use a good source of randomness to generate the random numbers required in the protocol. Please see [RFC 1750] for more information on generating random numbers for security applications. 7. Security claims This section provides the security claims required by [EAPbis]. [a] Intended use. EAP-PSK is intended for use over both physically insecure networks and physically or otherwise secure networks. Applicable media include but are not limited to PPP, IEEE 802 wired networks and IEEE 802.11. [b] Mechanism. EAP-PSK is based on symmetric cryptography and uses a 256 bit Pre-shared Key [c] Security claims. The security properties of the method are discussed in Section 6. [d] Key strength. EAP-PSK supports key derivation with 128-bit effective key strength. [e] Description of key hierarchy. Please see Section 2.2. [f] Indication of vulnerabilities. Vulnerabilities are discussed in Section 6. 8. Intellectual Property Right Notice The author neither has, nor is of aware of, any patents or pending patents relevant to material included in this draft. Bersani Expires ­ August 2004 [Page 29] INTERNET-DRAFT EAP PSK February 2004 2004 9. Acknowledgements This EAP method has been inspired by [EAP-SIM] and [EAP-Archie]. It also considerably reused extracts of these documents. Many thanks to their respective authors and especially: Jesse Walker, Russ Housley Henry Haverinen and Joseph Salowey. Many thanks to Aurélien Magniez, Henri Gilbert, Laurent Butti, Jérôme Razniewski and Olivier Charles for their feedback on this document. 10. References 10.1 Normative [AES] Federal Information Processing Standards (FIPS) Publication 197, " Specification for the Advanced Encryption Standard (AES)", National Institute of Standards and Technology, November 26, 2001. [EAP] Blunk, L. and Vollbrecht, J., "PPP Extensible Authentication Protocol (EAP)", RFC 2284, March 1998. [EAPbis] Blunk, L. et al., "Extensible Authentication Protocol (EAP)", Internet-Draft (work in progress), February 2004, http://ietf.levkowetz.com/drafts/eap/rfc2284bis/ draft-ietf-eap-rfc2284bis-09.txt [EAP-SM] Vollbrecht, J. et al., "State Machines for EAP Peer and Authenticator", Internet-Draft (work in progress), October 2003, draft-ietf-eap-statemachine-01 [EAKD] Bellare, M, and P. Rogaway, "Entity Authentication and Key Distribution", CRYPTO 93, LNCS 773, pp232-249, Springer-Verlag, Berlin, 1994. [EAX] Bellare, M. et al., "The EAX mode of operation", January 2004, http://www.cs.ucsd.edu/users/mihir/papers/eax.pdf [EKMF] Aboba, B. et al., "EAP Key Management Framework", Internet-Draft (work in progress), October 2003, draft- ietf-eap-keying-01.txt [NAI] Aboba, B., and M. Beadles, "The Network Access Identifier", RFC 2486, January 1999. [OMAC] Iwata, T., and Kurosawa., K., “OMAC: One-Key CBC MAC” Fast Software Encryption, FSE 2003, LNCS, Springer- Verlag. Bersani Expires ­ August 2004 [Page 30] INTERNET-DRAFT EAP PSK February 2004 2004 [PPP] Simpson, W., Editor, "The Point-to-Point Protocol (PPP)", STD 51, RFC 1661, July 1994. [RFC1750] Eastlake, D. et al., "Randomness Recommendations for Security", RFC 1750, December 1994. [RFC 2219] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, March 1997. [SOBMMO] Gilbert, H., “The Security of One-Block-to-Many Modes of Operation”, Fast Software Encryption, FSE 2003, LNCS, Springer-Verlag. 10.2 Informative [EAP-AKA] Arkko, J. Haverinen, H., “EAP AKA Authentication”, Internet-Draft (work in progress), October 2003, draft-arkko-pppext-eap-aka-11.txt [EAP-Archie] Walker, J. and Housley, R., “The EAP Archie Protocol”, Internet-Draft (work in progress), June 2003, draft- jwalker-eap-archie-01.txt, [EAP-SecurID] Josefsson, S., “The EAP SecurID(r) Mechanism”, Internet-Draft (work in progress), February 2002, draft-josefsson-eap-securid-01.txt [EAP-SIM] Haverinen, H. Salowey, J., "EAP SIM Authentication", Internet-Draft (work in progress),October 2003, draft- haverinen-pppext-eap-sim-12.txt [EAP-SKE] Salgarelli, L., “EAP SKE authentication and key exchange protocol”, Internet-Draft (work in progress), May 2003, draft-salgarelli-pppext-eap-ske-03.txt [EAP-SRP] Carlson, J. et al., “EAP SRP-SHA1 Authentication Protocol“, Internet-Draft (work in progress), draft- ietf-pppext-eap-srp-04.txt [EAP-TLV] Hiller, T. et al., "A Container Type for the Extensible Authentication Protocol (EAP)", Internet-Draft (work in progress), May 2003, draft-hiller-eap-tlv-01.txt [HAC] Menezes, A. et al., “Handbook of Applied Cryptography”, CRC Press, 1996. Bersani Expires ­ August 2004 [Page 31] INTERNET-DRAFT EAP PSK February 2004 2004 [IEEE 802.1X] IEEE STD 802.1X, Standards for Local and Metropolitan Area Networks: Port Based Access Control, June 14, 2001 [IEEE 802.3] Institute of Electrical and Electronics Engineers, "Information Technology - IEEE Standard for Information technology--Telecommunications and information exchange between systems--Local and metropolitan area networks— Specific requirements--Part 3: Carrier Sense Multiple Access with Collision Detection (CSMA/CD) Access Method and Physical Layer Specifications” [IEEE 802.11] Institute of Electrical and Electronics Engineers, "Information Technology - Telecommunications and Information Exchange between Systems - Local and Metropolitan Area Network - Specific Requirements ­ Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications", IEEE Standard 802.11 [IEEE 802REQ] Stanley, Dorothy et al., “EAP Method Requirements for Wireless LANs”, Internet-Draft (work in progress), January 2004, draft-walker-ieee802-req-00.txt [LEAP] Macnally, C., “Cisco LEAP protocol description”, September 2001 [LEAPVUL] Wright, J., “Weaknesses in LEAP Challenge/Response”, Defcon 2003 [MOSKBC] National Institute of Standards and Technology, "Modes of operation for symmetric key block ciphers", http://www.csrc.nist.gov/CryptoToolkit/modes/ [MTAP] Asokan, N. et al., “Man-in-the-middle in Tunnelled Authentication Protocols”, http://eprint.iacr.org/2002/163 [PEAP] Palekar, A. et al., “Protected EAP Protocol (PEAP)”, Internet-Draft (work in progress), October 2003, draft- josefsson-pppext-eap-tls-eap-07.txt [PEAP-TLV] Salowey, J., "Protected EAP TLV", Internet-Draft (work in progress), June 2003, draft-salowey-eap- protectedtlv-02.txt [PKCS5] RSA laboratories, “PKCS #5 v2.0: Password-Based Cryptography Standard” [PWD] National Institute of Standards and Technology (NIST). Bersani Expires ­ August 2004 [Page 32] INTERNET-DRAFT EAP PSK February 2004 2004 “FIPS PUB 112: Password Usage”. May 30, 1985. [WPA] Wi-Fi Alliance, “Wi-Fi Protected Access”, version 2.0, April 2003 11. Authors' Addresses Florent Bersani florent.bersani@francetelecom.com France Telecom R&D 38, rue du General Leclerc 92794 Issy Les Moulineaux Cedex 9 France 12. Full Copyright Statement Copyright (C) The Internet Society (2003). All Rights Reserved. This document and translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or assist in its implementation may be prepared, copied, published and distributed, in whole or in part, without restriction of any kind, provided that the above copyright notice and this paragraph are included on all such copies and derivative works. However, this document itself may not be modified in any way, such as by removing the copyright notice or references to the Internet Society or other Internet organizations, except as needed for the purpose of developing Internet standards in which case the procedures for copyrights defined in the Internet Standards process must be followed, or as required to translate it into languages other than English. The limited permissions granted above are perpetual and will not be revoked by the Internet Society or its successors or assignees. This document and the information contained herein is provided on an "AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE." Annex A: Work still to be done on this document 1. Editorial 1. Clarify the conventions used for the cryptographic calculations. 2. Make this draft more self-contained. 3. Provide test vectors Bersani Expires ­ August 2004 [Page 33] INTERNET-DRAFT EAP PSK February 2004 2004 4. Complete the cross-references in the text (page, section, reference numbers) 5. Harmonize the terminology (e. g. octets or bytes) 6. Complete the terminology section (e.g. with temporary identity) 7. Number the figures 2. Security 1. Study alternative ways to produce the input blocks to the key derivation procedure. 2. Validate the mechanism to prevent abrupt ending of the conversation on the protected channel. 3. Discuss use of other cryptographic algorithms 4. Discuss the choice of AES-128 and OMAC1 and a tag length of 128 bits (e.g. why don't we use a truncated tag) 5. Should we use the authentication capacity of associated data of EAX that is currently left aside? 6. Discuss identity protection 7. Specify the how long the same protected channel may be used without security compromise (a pretty long time I bet but it should be explicitly evaluated and stated) 8. Specify how long the same protected channel may be used without security compromise (a pretty long time I bet but it should be explicitly evaluated and stated) 9. Study DOS attacks resistance 3. Technical 1. Discuss possibility to enhance network efficiency for instance by including the AT_IDREQ and AT_RAND attribute in the first packet. 2. Should the peer be allowed to move directly from the Init state to the failure state to avoid lengthy time outs or retries? 3. Specify how fast reconnect should be implemented 4. Introduce version negotiation 5. Is it desirable to have all attributes aligned on 32 bit boundaries? 6. Harmonize with other standards or draft standards (e. g. EAP and EAP Key management framework) 7. Should the AT_STATUS remain a separate attribute from AT_PCHANNEL? 8. Should an intermediary state be included in the state machine that would handle the entry in the P_Channel state? 9. Specify all that remain TBC, e.g. the different notification messages. 10. Make sure that the interface of this EAP method with the EAP state machine works well 11. Define more attributes Bersani Expires ­ August 2004 [Page 34] INTERNET-DRAFT EAP PSK February 2004 2004 Annex B: Guidance for PSK generation from a password It is formally discouraged to use a password to generate a PSK, since this commonly lead to exhaustive search or dictionary attacks that would not otherwise be possible. However, we provide guidance on how to generate the PSK from a password. Guidance on how passwords should be selected is provided in [PWD]. The technique presented herein is drawn from [PKCS5]. It is intended to mitigate the risks associated with password usage in cryptography, typically dictionary attacks. If the binary representation in ASCII of the password is strictly fewer than 128 bit long (which by the way means that the chosen password is probably weak because it is too short) then its is padded to 128 bits with zeroes. If the binary representation in ASCII of the password is strictly more than 128 bit long, then it is hashed down to exactly 128 bit using the Matyas-Meyer-Oseas hash (please refer to [HAC] for a description of this hash) with IV=0x0123456789ABCDEFFEDCBA9876543210 (this value has been arbitrarily selected). We now assume that we have a 128 bit number derived from the initial password (that can be the password itself if its binary representation in ASCII is exactly 128 bit long). We shall call this number P128. EAP-PSK’s PSK is derived thanks to PBKDF2 instantiated with (following the notations in [PKCS5]): 1. P128 as P 2. The first 96 bits of the binary ASCII representation of the xor of the peer’s and the server's NAI as Salt. 3. 5000 as c 4. 48 as dkLen Bersani Expires ­ August 2004 [Page 35]