SUMMARYIn 2009 and 2011, Chen et al. and Lasc et al. proposed two separate authentication schemes for mobile satellite communication systems. Unfortunately, their schemes are unable to protect security in the event of smart card loss. In this paper, we propose a novel version that resists common malicious attacks and improves both the schemes of Chen et al. and Lasc et al. The security of our scheme is based on the discrete logarithm problem and one‐way hash function. A nonce mechanism is also applied to prevent replay attack. Furthermore, our scheme is more efficient than related schemes and thus more suitable for being implemented in satellite communication systems. Copyright © 2012 John Wiley & Sons, Ltd.

No abstract

Three-Party encrypted key exchange (3PEKE) protocol is an important cryptographic mechanism, enabling two clients to communicate with each other through a trusted server. Specifically, it is often applied to provide confidential communications between two participants over an insecure network. In 2007, Lu and Cao proposed a simple and efficient version. Unfortunately, we find that there exists a design weakness in their mechanism. We therefore design a novel mechanism using smart cards. The smart-card based scheme (SC-3PEKE) not only confirms the security essentials of general 3PEKE methods, but also outperforms related works in terms of authentication overheads with the adoption of smart cards.

In single-server architecture, one service is maintained by one server. If a user wants to employ multiple services from different servers, he/she needs to register with these servers and to memorize numerous pairs of identities and passwords corresponding to each server. In order to improve user convenience, many authentication schemes have been provided for multi-server environment with the property of single registration. In 2013, Li et al. provided an efficient multi-server authentication scheme, which they contended that it could resist several attacks. Nevertheless, we find that their scheme is sensitive to the forgery attack and has a design flaw. This paper presents a more secure dynamic identity-based multi-server authentication scheme in order to solve the problem in the scheme by Li et al. Analyses show that the proposed scheme can preclude several attacks and support the revocation of anonymity to handle the malicious behavior of a legal user. Furthermore, our proposed scheme has a lower computation and communication costs, which make it is more suitable for practical applications. many pairs of identities and passwords corresponding to each server. It is very inconvenient for users. Therefore, several multi-server authentication schemes have been designed [11][12][13][14][15][16][17][18][19][20][21][22][23][24]. In a multi-server system, users only need to register with the registration center one time; then, they can access all of the services from different servers in this system. Since 2001, Li et al. [11], Lin et al.[12], Juang [13], and Chang and Lee [14] have designed different multi-server authentication mechanisms based on neural networks, discrete logarithms, or symmetric cryptosystems. However, due to high computation cost, these schemes are inefficient. Moreover, in these schemes, users' identities are static and publicly transmitted in plaintext form, which allows an attacker to potentially track a specific user. Accordingly, user privacy and location need to be protected. To achieve user anonymity, Liao and Wang [15] provided a dynamic identitybased authentication scheme for multi-server architecture. Unfortunately, Hsiang and Shih [16] found that their scheme has some security flaws and fails to provide proper mutual authentication between users and servers; consequently, they provided an improvement to enhance security. However, Yeh et al. [18] found that their method is insecure against the replay attack, the server spoofing attack, behavior denial problem, and session key disclosure; Lee et al. [19] found that their method suffers from the masquerade attack, and lack of mutual authentication; whereas Sood et al. [17] found that their method is vulnerable to the stolen smart card attack, and the impersonation attack, and has an incorrect password change procedure. In order to remedy these weaknesses, Lee et al. and Sood et al. presented improvements. In 2012, Chuang and Tseng [21] thought that most of previous schemes [15,16] cannot prevent the impersonation attack and the server sp...

No abstract