Current mechanisms for authenticating communication between devices that share no prior context are inconvenient for ordinary users, without the assistance of a trusted authority. We present and analyze Seeing-Is-Believing, a system that utilizes 2D barcodes and cameraphones to implement a visual channel for authentication and demonstrative identification of devices. We apply this visual channel to several problems in computer security, including authenticated key exchange between devices that share no prior context, establishment of a trusted path for configuration of a TCG-compliant computing platform, and secure device configuration in the context of a smart home. 1.

Abstract. There has been much interest in password-authenticated keyexchange protocols which remain secure even when users choose passwords from a very small space of possible passwords (say, a dictionary of English words). Under this assumption, one must be careful to design protocols which cannot be broken using off-line dictionary attacks in which an adversary enumerates all possible passwords in an attempt to determine the correct one. Many heuristic protocols have been proposed to solve this important problem. Only recently have formal validations of security (namely, proofs in the idealized random oracle and ideal cipher models) been given for specific constructions [3, 10, 22]. Very recently, a construction based on general assumptions, secure in the standard model with human-memorable passwords, has been proposed by Goldreich and Lindell [17]. Their protocol requires no public parameters; unfortunately, it requires techniques from general multi-party computation which make it impractical. Thus, [17] only proves that solutions are possible “in principal”. The main question left open by their work was finding an efficient solution to this fundamental problem. We show an efficient, 3-round, password-authenticated key exchange protocol with human-memorable passwords which is provably secure under the Decisional Diffie-Hellman assumption, yet requires only (roughly) 8 times more computation than “standard ” Diffie-Hellman key exchange [14] (which provides no authentication at all). We assume public parameters available to all parties. We stress that we work in the standard model only, and do not require a “random oracle ” assumption.

Authentication of communication channels between devices that lack any previous association is an challenging problem. It has been considered in many contexts and in various flavors, most recently, by McCune et al., where human-assisted device authentication is achieved through the use of photo cameras (present in some cellphones) and 2-dimensional barcodes. Their proposed Seeing-is-Believing system allows users with devices equipped with cameras to use the visual channel for authentication of unfamiliar devices, so as to defeat man-inthe-middle attacks. In this paper, we investigate an alternative and complementary approach—the use of the audio channel for humanassisted authentication of previously un-associated devices. Our motivation is three-fold: (1) many personal devices are not equipped with cameras or scanners, (2) some human users are visually impaired (hence, cannot be in the authentication pipeline of a vision-based system), and (3) some usage scenarios preclude either taking a sufficiently clear picture and/or the use of barcodes. We develop and evaluate a system we call Loud-and-Clear (L&C) authentication, which, like Seeing-is-Believing, places little demand on the human user. The L&C system is based on the use of a text-to-speech engine to read an auditoriallyrobust, grammatically-correct pass-phrase derived from an authentication string that is to be used by peer devices. In particular, by coupling the auditory reading of the one-way hash of an authentication string on one device with the display of of this text on another device, we demonstrate that L&C is suitable for secure device pairing (e.g., key exchange) and similar tasks. We also describe several use cases, as well as provide some performance data for a prototype implementation and a discussion of the security properties of L&C. 1

We present a simple technique by which a device that performs private key operations (signatures or decryptions) in networked applications, and whose local private key is activated with a password or PIN, can be immunized to offline dictionary attacks in case the device is captured. Our techniques do not assume tamper resistance of the device, but rather exploit the networked nature of the device, in that the device’s private key operations are pe formed using a simple interaction with a remote sewer: This sewer; however; is untrusted-its compromise does not reduce the securiv of the device’s private key unless the device is also captured-and need not have a prior relationship with the device. We further extend this approach with support for key disabling, by which the rightj‘ul owner of a stolen device can disable the device’s private key even if the attacker already knows the user’s password. 1.

We propose and realize a definition of security for password-based key exchange within the framework of universal composability (UC), thus providing security guarantees under arbitrary composition with other protocols. In addition, our definition captures some aspects of the problem that were not adequately addressed by most prior notions. For instance, our definition does not assume any underlying probability distribution on passwords, nor does it assume independence between passwords chosen by different parties. We also formulate a definition of password-based secure channels, and show how to realize such channels given any passwordbased key exchange protocol. The password-based key exchange protocol shown here is in the common reference string model and relies on standard number-theoretic assumptions. The components of our protocol can be instantiated to give a relatively efficient solution which is conceivably usable in practice. We also show that it is impossible to satisfy our definition in the “plain ” model (e.g., without

In this paper we give a detailed formal description of the PAK password-authenticated key exchange protocol and some variants, and provide provide complete proofs of security which we believe are more straight-forward than the original proofs. We also show a new general method (called the Z-method) for making these protocols resilient to server-compromise, so as to not allow an attacker that obtains password verification data from a server to then impersonate a user. When this method is applied to PAK, we call the resulting protocol PAKZ. Finally, we discuss the current state-of-the-art in password-authenticated key exchange, with respect to both theory and practice.

SSL is the de-facto standard today for securing end-to-end transport on the Internet. While the protocol itself seems rather secure, there are a number of risks that lurk in its use, e.g., in web banking. However, the adoption of password-based key-exchange protocols can overcome some of these problems. We propose the integration of such a protocol (DH-EKE) in the TLS protocol, the standardization of SSL by IETF. The resulting protocol provides secure mutual authentication and key establishment over an insecure channel. It does not have to resort to a PKI or keys and certicates stored on the users computer. Additionally, its integration in TLS is as minimal and non-intrusive as possible.

Abstract. Password-based encrypted key exchange are protocols that are designed to provide pair of users communicating over an unreliable channel with a secure session key even when the secret key or password shared between two users is drawn from a small set of values. In this paper, we present two simple password-based encrypted key exchange protocols based on that of Bellovin and Merritt. While one protocol is more suitable to scenarios in which the password is shared across several servers, the other enjoys better security properties. Both protocols are as efficient, if not better, as any of the existing encrypted key exchange protocols in the literature, and yet they only require a single random oracle instance. The proof of security for both protocols is in the random oracle model and based on hardness of the computational Diffie-Hellman problem. However, some of the techniques that we use are quite different from the usual ones and make use of new variants of the Diffie-Hellman problem, which are of independent interest. We also provide concrete relations between the new variants and the standard Diffie-Hellman problem. Keywords. Password, encrypted key exchange, Diffie-Hellman assumptions. 1

Abstract. In most password-authenticated key exchange systems there is a single server storing password verification data. To provide some resilience against server compromise, this data typically takes the form of a one-way function of the password (and possibly a salt, or other public values), rather than the password itself. However, if the server is compromised, this password verification data can be used to perform an offline dictionary attack on the user’s password. In this paper we propose an efficient password-authenticated key exchange system involving a set of servers, in which a certain threshold of servers must participate in the authentication of a user, and in which the compromise of any fewer than that threshold of servers does not allow an attacker to perform an offline dictionary attack. We prove our system is secure in the random oracle model under the Decision Diffie-Hellman assumption against an attacker that may eavesdrop on, insert, delete, or modify messages between the user and servers, and that compromises fewer than that threshold of servers. 1

Goldreich and Lindell (CRYPTO `01) recently presented the first protocol for password-authenticated key exchange in the standard model (with no common reference string or set-up assumptions other than the shared password). However, their protocol uses several heavy tools and has a complicated analysis. We present a simplification of the Goldreich-Lindell (GL) protocol and analysis for the special case when the dictionary is of the form D = {0, 1}d i.e., the password is a short string chosen uniformly at random (in the spirit of an ATM PIN number). The security bound achieved by our protocol is somewhat worse than the GL protocol. Roughly speaking, our protocol guarantees that the adversary can ibreakj the scheme with probability at most O(poly(n)/|D|)\Omega (1), whereas the GL protocol guarantees a bound of O(1/|D|). We also