[Paper Review] Cryptanalysis of Multi-Server Authenticated Key Agreement Scheme Based on Trust Computing Using Smart Cards and Biometrics

Dheerendra Mishra|arXiv (Cornell University)|Jan 20, 2014

Advanced Authentication Protocols Security7 references3 citations

TL;DR

This paper cryptanalyzes a multi-server authenticated key agreement scheme using smart cards and biometrics, demonstrating it is vulnerable to stolen smart card attacks, leading to user impersonation, server spoofing, and man-in-the-middle attacks. The scheme also fails to provide forward secrecy, undermining long-term session security despite its lightweight design based on hash functions.

ABSTRACT

Advancement in communication technology provides a scalable platform for various services where a remote user can access the server from anywhere without moving from its place. It has provided a unique opportunity for online services, such that the user need not physically present at the service center. These services adopt authentication and key agreement protocols to ensure authorized and secure access to resources. Most of the authentication schemes support single server environment where the user has to register with each server. If a user wishes to access multiple application servers, he requires to register with each of the servers. Although multi-server authentication schemes introduced a scalable platform such that a user can interact with any server using single registration. Recently, Chuang and Chen proposed an efficient multi-server authenticated key agreement scheme based on smart cards along with password and biometrics. This is a lightweight authentication scheme which requires the computation of only hash function. In this article, we present a brief review of Chuang and Chen's scheme. We analyze Chuang and Chen's scheme and identify that their scheme does not resist stolen smart card attack which causes the user's impersonation attack, server spoofing attack and man-in-the middle attack. Additionally, we show that their scheme has a weak key agreement protocol, which does not ensure forward secrecy.

Motivation & Objective

• To analyze the security of Chuang and Chen’s multi-server authenticated key agreement scheme that combines smart cards, passwords, and biometrics.
• To identify critical vulnerabilities in the scheme’s resistance to stolen smart card attacks.
• To evaluate the scheme’s resilience against server spoofing and man-in-the-middle attacks.
• To assess whether the scheme achieves forward secrecy, a critical property for long-term session security.
• To provide a formal critique of the scheme’s design flaws in a real-world threat model involving powerful eavesdropping adversaries.

Proposed method

• Reverse-engineered the login and authentication phases of Chuang and Chen’s scheme using the same terminology and protocol flow.
• Simulated an adversary with access to a stolen smart card to extract stored parameters including $D_i$, $C_i$, $B_i$, and $ID_i$.
• Constructed a man-in-the-middle attack by intercepting and modifying messages between the user’s smart card and the server.
• Demonstrated server spoofing by forging responses using the stolen identity and biometric parameters.
• Evaluated session key derivation using $SK_{ij} = h(N_1 || N_2)$ and showed that the long-term secret $A_i$ could be used to recompute past session keys.
• Verified the correctness of all attack vectors through formal message flow analysis and equation validation using the scheme’s cryptographic primitives.

Experimental results

Research questions

• RQ1 Can an adversary impersonate a legitimate user after compromising a smart card in Chuang and Chen’s scheme?
• RQ2 Is the scheme secure against server spoofing when an adversary controls the communication channel?
• RQ3 Can a man-in-the-middle attacker establish independent, undetected sessions with both the user and server?
• RQ4 Does the scheme provide forward secrecy, such that past session keys remain secure even if long-term secrets are exposed?
• RQ5 What are the specific cryptographic weaknesses that allow these attacks to succeed with only hash function operations?

Key findings

• The scheme is vulnerable to a stolen smart card attack, allowing an adversary to impersonate a legitimate user by extracting $D_i$, $C_i$, $B_i$, and $ID_i$.
• An adversary can successfully perform a server spoofing attack by forging messages using the stolen user identity and biometric parameters.
• The scheme is susceptible to a man-in-the-middle attack, where the adversary establishes separate, authenticated sessions with both the user and server, maintaining the illusion of direct communication.
• The session key $SK_{ij} = h(N_1 || N_2)$ can be computed by an adversary using only the compromised long-term secret $A_i = h(ID_i || x)$, breaking forward secrecy.
• The scheme’s reliance on a single hash function and lack of mutual authentication mechanisms render it insecure under active, powerful adversaries.
• Despite its claim of efficiency and lightweight design, the scheme fails to meet fundamental security requirements for real-world deployment in multi-server environments.