Enhance Patient Medication Safety With A RFID-based Authentication Scheme

Youjun Xu, Jialiang He, Jian Wang, Dongxing WangInternational Journal of u- and e- Service Science and Technology 2014 · DOI

Abstract

Medication Safety is an important issue for patients. Automated patient medication systems using RFID technology are used to reduce medication errors and improve patient safety. In this paper, we show the weaknesses of Sonam Devgan Kaul et al's authentication scheme. In order to enhance medication safety for patients, we propose a new lightweight RFID authentication scheme based on dynamic ID. This scheme only requires $O(1)$ work to identify and authenticate a tag in the backend server and is particularly suitable for low-cost RFID systems because only one-way hash functions, XOR operations, and concatenation operations are needed in the tag part, making it practical, secure, and efficient for the healthcare domain.

Introduction

In healthcare settings, Radio Frequency Identification (RFID) can be a lifesaver by ensuring the right patient receives the right drug at the right time. However, many existing RFID security protocols suffer from two major flaws: the "Scalability Wall" and desynchronization.

To identify a tag securely, many servers must perform a linear search ($O(n)$), checking the incoming signal against every single tag in the database. As a hospital's inventory grows, the server becomes a bottleneck. Furthermore, if a message is intercepted or a power failure occurs during an update, the tag might update its secret key while the server does not, leading to desynchronization. This paper addresses these weaknesses, specifically targeting a 2013 scheme by Kaul et al. that required expensive internal clocks and forced computationally heavy $O(n)$ searches.

The Proposed Protocol

The proposed dynamic ID-based mutual authentication protocol involves three parties: the Tag (on the drug/patient), the Reader (held by the nurse), and the Backend Server.

The proposed dynamic ID-based mutual authentication protocol.

Key Technical Features

  • Lightweight Operations: The tag only performs one-way hash functions ($H$), XOR operations ($\oplus$), and concatenation ($\parallel$). It does not require a clock generator, making it compatible with low-cost passive tags.
  • Dynamic Secrets: After every successful authentication, the tag’s identity ($ID$) and secret key ($k$) are updated. This ensures "untraceability"—an attacker cannot track a specific tag over time because its "name" is always changing.
  • The "Pre-pair" Strategy: To solve the desynchronization problem, the server stores both the current $(ID, k)$ pair and the previous $(ID, k)$ pair. If an authentication attempt fails with current credentials, the server checks the previous ones, allowing the system to "heal" if a previous session was interrupted.

Performance and Security Analysis

The most significant advancement is the shift from $O(n)$ to $O(1)$ complexity. By structured handling of the keys, the server can identify a tag almost instantly, regardless of the total number of tags in the system. Additionally, by moving timestamping responsibility to the Reader and Server, the hardware cost of the tags is significantly lowered.

Performance comparison between the proposed scheme and the Kaul et al. scheme.

As shown in the performance analysis, the proposed scheme reduces total traffic cost from $4l$ to $3l$ (where $l$ is the bit-length of the ID) and achieves constant time complexity ($O(1)$) at the server. The protocol successfully withstands common attacks:

  • Replay Attacks: Prevented by fresh random numbers and timestamps.
  • Desynchronization: Solved by the "pre-pair" database record.
  • Cloning: Prevented because secret parameters are never transmitted in the clear and change every session.

Conclusion and Future Work

While the protocol is mathematically sound and efficient, future work could involve physical implementation to measure actual power consumption and latency on specific RFID hardware like EPCglobal Class-1 Gen-2 tags. Additionally, formal verification using methods like ProVerif or BAN logic would provide a mathematical proof of the protocol's security properties. Despite these areas for further study, the proposed scheme offers a practical and secure solution for enhancing medication safety in the healthcare domain.

Summary updated

Abstract

Medication Safety is an important issue for patients.Automated patient medication system using RFID technology is used to reduce the medication errors and improve the patient safety.In this paper, we show the weaknesses of Sonam Devgan Kaul et al's authentication scheme.In order to enhance medication safety for patients, we propose a new lightweight RFID authentication scheme based on dynamic ID.This scheme only requires O(1) work to identify and authenticate a tag in the backend server and is particularly suitable for the low-cost RFID systems because only one-way hash function, XOR operation and concatenation operation are needed in tag part, so it is practical, secure and efficient for health care domain.
BibTeX
@article{xu2014enhance,
  author = {Youjun Xu and Jialiang He and Jian Wang and Dongxing Wang},
  title = {{Enhance Patient Medication Safety With A RFID-based Authentication Scheme}},
  journal = {International Journal of u- and e- Service Science and Technology},
  year = {2014},
  doi = {10.14257/ijunesst.2014.7.4.9},
}
Figure