Mobile Payment Protocol with Deniably Authenticated Property

Yunzhuo Liu, Wen Huang, Ming Zhuo, Shijie Zhou, Mengshi LiSensors 2023 · OpenReview

TL;DR

This paper introduces a novel mobile payment protocol designed for resource-constrained mobile devices. It leverages deniable authenticated encryption to allow users to verify transaction participants while preventing them from proving others' involvement, thus enhancing privacy for sensitive purchases.

What Problem Does This Paper Address?

Mobile payment systems are rapidly gaining traction, offering convenience for everyday transactions. However, this widespread adoption brings significant privacy concerns. A key issue is the potential disclosure of personal information related to transactions, especially for sensitive purchases like specific medications. Current systems might inadvertently reveal such details, leading to potential misuse or embarrassment. Furthermore, mobile devices have limited computational power, which restricts the complexity of security protocols that can be implemented. This paper addresses the challenge of creating a secure and private mobile payment protocol that is both efficient enough for mobile devices and provides a strong guarantee of user privacy, specifically by allowing participants to confirm each other's identities without leaving irrefutable proof of their involvement.

Methods

The core of this paper is the development of a mobile payment protocol that utilizes a deniably authenticated encryption scheme. This scheme is specifically optimized for mobile devices with limited computing resources. The protocol is built upon four key algorithms: Setup, KeyGen, DA-Encrypt, and DA-Decrypt.

  • Setup: Generates system parameters.
  • KeyGen: Creates public and private key pairs for senders and receivers.
  • DA-Encrypt: Encrypts a message using the sender's private key and the receiver's public key, ensuring confidentiality and integrity.
  • DA-Decrypt: Decrypts a ciphertext, verifying the sender's identity and ensuring the message hasn't been tampered with.

The "deniable" aspect is crucial: while participants can confirm each other's identities during a transaction, they cannot later produce irrefutable evidence to prove to a third party that a specific individual was involved. This is achieved by ensuring that the encryption process is designed such that the sender can deny their participation if needed, without compromising the ability of the receiver to verify their identity during the transaction. The protocol is structured into three phases: setup, key generation, and the payment transaction itself.

Datasets

Not discussed in the paper.

Key Insights and Ideas

  • Deniable Authentication for Mobile Payments: The paper highlights the critical need for deniable authentication in mobile payment scenarios, especially when users might be purchasing sensitive items. This goes beyond standard authentication by allowing for privacy even when transaction details could be sensitive.
  • Optimized for Resource-Constrained Devices: A significant contribution is the focus on optimizing the deniable authenticated encryption scheme for mobile devices. This addresses a practical limitation that often prevents more complex cryptographic solutions from being deployed on smartphones and similar devices.
  • Balancing Privacy and Verifiability: The protocol successfully balances the need for users to confirm transaction partners with the desire to prevent irrefutable proof of participation. This is a nuanced approach to privacy that acknowledges the complexities of real-world transactions.
  • Multi-Property Protocol: The proposed protocol aims to satisfy multiple security and privacy goals simultaneously, including confidentiality, integrity, deniable authentication, traceability, non-repudiation, and lightweight overhead.

Key Results

The paper presents a theoretical analysis and an experimental evaluation of the proposed protocol.

  • Security Properties: The protocol is claimed to achieve confidentiality, integrity, deniable authentication, traceability, and non-repudiation. The deniable authentication property is specifically demonstrated by showing that a merchant cannot forge a client's legal ciphertext without the client's private key, thus allowing the client to deny participation.
  • Lightweight Performance: Experimental results, visualized in Figures 3 and 4, show the running time for encryption and decryption operations across various key lengths. The graphs indicate that the protocol exhibits relatively low computational overhead, making it suitable for mobile devices. For instance, the encryption time for a 512-bit key length is shown to be well within practical limits for mobile applications.

Limitations and Future Work

  • Trusted PSP: The protocol assumes the Payment Service Provider (PSP) is a trusted entity. This is a common assumption in many payment protocols, but it represents a potential single point of failure or trust.
  • Experimental Environment Specificity: The performance results are based on a specific experimental setup (Ubuntu 12.04, Intel Core i5 2.6GHz, 8GB RAM, MIRACL library). While indicative, performance might vary on different hardware and software configurations.
  • Future Work: The authors suggest that further research could explore more advanced cryptographic techniques or consider scenarios with untrusted PSPs. The paper also implies that the protocol's efficiency could be further optimized.

Summary updated

Abstract

Mobile payment services have been widely applied in our daily life, where users can conduct transactions in a convenient way. However, critical privacy concerns have arisen. Specifically, a risk of participating in a transaction is the disclosure of personal privacy. This might occur if, for example, the user pays for some special medicine, such as AIDS medicine or contraceptives. In this paper, we propose a mobile payment protocol that is suitable for mobile devices only with limited computing resources. In particular, the user in a transaction can confirm the identity of others in the same transaction while the user cannot show convincing evidence to prove that others also take part in the same transactions. We implement the proposed protocol and test its computation overhead. The experiment results corroborate that the proposed protocol is suitable for mobile devices with limited computing resources.
BibTeX
@inproceedings{liu2023mobile,
  author = {Yunzhuo Liu and Wen Huang and Ming Zhuo and Shijie Zhou and Mengshi Li},
  title = {{Mobile Payment Protocol with Deniably Authenticated Property}},
  booktitle = {Sensors},
  year = {2023},
  url = {https://openreview.net/forum?id=afYRNO5hwS},
}
Figure