A RFID Authentication Protocol for Epidemic Prevention and Epidemic Emergency Management Systems

The outbreak of the novel coronavirus has exposed many problems in the auxiliary information system for epidemic prevention and control, which needs to be resolved by using methods such as the antitampering of logistics data and the management and control of epidemic materials. This article discusses the introduction of emerging technologies such as Radio Frequency Identification (RFID), which support privacy protection into the auxiliary information system for epidemic prevention and control. Recently, this paper found that Khwaja et al.'s protocol (RAPUS protocol) is susceptible to database impersonation attacks and reader impersonation attacks. Therefore, this article proposes the enhanced protocol, which not only perfectly solves the problems of the abovementioned protocols but also comprehensively compares multiple protocols. The enhanced protocol has higher efficiency and security. The security of the proposed protocol (RAPUS + protocol) is analyzed by GNY logic and the AVISPA model. The designed scheme can help realize the safety and traceability of epidemic prevention materials and improve the automation and decision-making efficiency of the epidemic prevention.


Introduction
Medical Internet of ings (IoMT) needs higher demand for security than the Internet of ings (IoT) when decreasing medical expense and enhancing the medical workpiece ratio. Khwaja et al.'s protocol [1] uses symmetric cryptography to design a robust authentication protocol. In the enforcement of IoMT, RFID is required to create an identity authentication system as a key technology, which can effectually verify patients [2]. It is helpful to design a safe and efficient RFID for protecting user privacy, improving the safety of the IoMT system, and boosting the efficiency of medical staff to inspect and manage patients.
In medical scenes that need to efficiently verify a good deal of tags in a brief period, the employ of traditional per-tag protocols is inefficient and may put off the cure of patients. To solve the problem that uses the solution of homogeneous linear equations as a key to make the authentication data of the tag encipherment to further decrease the expense, Nikkhah et al. [3] proposed the LAPCHS protocol on account of the cloud healthcare system and analyzed and proved the safety of the protocol against various attacks by heuristic security analysis. Sarier et al. [4] proposed a new biometric based on nontransferable certificate scheme, which maintains the privacy and efficiency of biometric identification and can be easily integrated into the current BBIM system based on efficient brand and PS certificate.
Due to the low cost, high stability, and excellent characteristics of noncontact automatic identification of RFID system, it has been used in many fields, such as waste electronic product processing, farm management, and supply chain management [5][6][7]. In addition, in the field of healthcare, RFID technology is also widely used, such as drug management and vital signs detection [8,9]. During the epidemic, RFID technology also participated in the establishment of the auxiliary information system for epidemic prevention, with the purpose of solving problems such as the management of epidemic material data and the management of epidemic information. Although the RFID system has many advantages in epidemic prevention and control, it faces the risks of security and privacy due to the usage of bottomed wireless media that enables (T, RE)'s mutual communication.
In order to design the secure RFID authentication solution, many protocols have been put forward to guarantee the security of RFID, but they have all been proved to have problems. In 2019, Khwaja et al. [1] proposed the lightweight authentication protocol. e authors claim that their protocol meets the necessary security requirements and solves most security issues. is paper finds that the protocol [1] is susceptible to database spoofing attacks, reader spoofing attacks, and asynchronous attacks. erefore, this article proposes the new protocol on this basis, which can not only solve the problems of the abovementioned protocols but also can make the comprehensive comparison of multiple protocols, which has higher efficiency and security. e security of RAPUS + protocol is analyzed formally and informally through GNY logic, security verification tool AVISPA model. rough the above scheme, it is expected to improve the automation and decision-making efficiency of the epidemic prevention auxiliary information system.
is article analyzes the protocol [1] and improves realistic and lightweight certification protocol to guarantee protection against the known attacks: (1) e protocol [1] is susceptible to database impersonation attacks and reader impersonation attacks (2) e improved authentication protocols are proposed to resist all known attacks (3) RAPUS + protocol performs formal and informally security analysis and compares it with relevant existing protocols on security features and performance
In 2014, Cho et al. [10] put forward the hash-based protocol. Later, Safkhani et al. [11] pointed out that the protocol [10] cannot withstand DoS attacks and impersonation attacks. In the same year, the ECC-based RFID authentication protocols [12,13] were proposed to ensure communication security in the medical environment and improve patient safety. However, Farash et al. [14] proved through analysis that the protocols [12,13] cannot ensure forward secrecy. In 2015, Gope et al. [15] proposed the lightweight protocol. But Khwaja et al. [1] pointed out that their protocol is susceptible to collision, DoS attacks, and stolen attacks. In 2018, Fan et al. [16] proposed the ultralightweight LRMI protocol to protect medical privacy. However, Aghili et al. [17] analyzed that the LRMI protocol cannot withstand traceability attacks and simulation attacks and proposed the SecLAP protocol. However, under the analysis of Safkhani et al. [18], it is found that the protocol in [17] is susceptible to traceability attacks and asynchronous attacks. In the same year, Fan et al. [21] put forward the lightweight authentication scheme on the basis of quadratic. Later, Zhu et al. [22] analyzed and proved that their scheme is susceptible to forward secrecy and impersonation attacks. In 2019, Zhou et al. [19] put forward the protocol on the basis of secondary residues. But the protocol [20] cannot withstand asynchronous attacks. Naeem et al. [23] presented an RFID authentication protocol and suggested an improvement to cater to the correctness and scalability issues. Li et al. [24] presented a mutual-healing group key distribution scheme based on the blockchain which can effectively resist various attacks with small overhead on time and storage. Amin et al. [25] provided an effective solution to solve all existing problems regarding key protocol methods to enhance security. e AVISPA simulation results in the solution ensure that active and passive attacks are protected. Lin et al. [26] constructed a novel secure mutual authentication system and proved the security and privacy requirements, including anonymity, traceability, and confidentiality. Shahidinejad et al. [27] introduced a lightweight authentication protocol for IoT devices named Light-Edge using a three-layer scheme.
Section 3 reviews the protocol [1]. Section 4 demonstrates RAPUS + protocol. In Section 5, the security analysis of the RAPUS + protocol is checked. Section 6 compares RAPUS + protocol with existing protocols. At last, Section 7 summarizes the article.

The Analysis of RAPUS Protocol
By examining the limitations of the RAPUS protocol, this section points out that it is susceptible to database impersonation attacks. Table 2 demonstrates the symbols used in the RAPUS protocol. e doctor and many patients make up every cluster. Patients are moved from one cluster to another. Via the DB Server, the registered patients are authenticated by the doctor and the cluster. e symmetric key K rs is shared between each Doctor and DB. e improved authentication scheme is made up of two parts: the tag registration part and the tag certification part. We have the medical data through the protocol. Besides, we use the decentralization, traceability, and nontampering characters of blockchain technology to guarantee the security of data storage and sharing.
T registrations steps are as follows, as shown in Figure 1.
Step PTR 1. ID Ti is submitted to the DB by each tag.
(1) A random number n s is generated by DB.
(3) DB randomly generates r i and encrypts it with s x to calculate one-time alias tag i 's identity AID � E sx (ID Ti ||r Ti ). (4) DB authenticates tag i based on AID T . (5) DB stores and delivers M 2 to tag.
Step PTR 3. Tag i stores the information M 2 � {ID Ti , K ts , AID} in its memory after receiving the messages from DB. e registered T starts the certification procedure, which is presented in Figure 2. e specific steps are as follows: Step PTA 1: (1) Nt is generated by an RFID tag with IDTi.
the authentication request.
Step PTA 2: (1) R i of the i th cluster verifies the timestamp freshness as ( Step PTA 3: (1) S proves (T 3 −T 2 )≤∆T and then stems from N t � K ts ⊕ N x and N r � K rs ⊕ N y .
are calculated and verified by S.
‖N r ‖K rs � � � �T 2 ) are calculated and verified by S.   e reader's impersonation attacks succeed.
3.3. Vulnerable to Asynchronous Attacks STEP 1. e above two processes lead to the wrong updating of cryptographic key AID' Ti(new) �(Z′ T ⊕ K Ts ).

STEP 2.
In the second round of conversation, T key AID' Ti(new) and T key AID Ti(new) �(Z T ⊕ K Ts ) stored in the database are inconsistent.
erefore, it results in asynchronous attacks.

The Improved RAPUS + Protocol
RAPUS + protocol is shown in Figure 3.
Step PTA 1: (1) N t is generated by RFID tag with ID Ti .
Step PTA 2: (1) R i of the i th cluster first proves the timestamp freshness as (T 2 −T 1 )≤∆T, as soon as the request is received from T.
Step PTA 3: (1) S proves (T 3 −T 2 ) ≤∆T and then stems from N t � K ts ⊕ N x and N r � K rs ⊕ N y . (2) S calculates and proves Step PTA 4: Step PTA 5: (1) Tag i checks the freshness of the timestamp after receiving M A4.
(2) Tag i calculates and renews AID Ti(new) �(Z T ⊕ K ts ), AID Ti � AID Ti(new) if Tag i verifies the message V 4 * ? � h(K ts ||ID Ti ||N t ||Z T ). (3) Tag i saves the information. (4) Otherwise, AID Ti will not update.

The Analysis of RAPUS + Protocol
e security of the RAPUS + protocol is analyzed through formal analysis and explains the informal security features under the GNY logic model and AVISPA model.

e Informal Analysis.
In the following subsections, we analyze the security of the RFID system.

Mutual Authentication between Tag and Server.
e DB verifies AID Ti and V 1 � h(AID Ti ||K ts ||N x ||R i ) in M 1 to authenticate RFID tag. Only the legitimate T can form the effective request M 1 , which includes the two parameters. As effective AID Ti , it only knows legal T. Besides, the legal tag only knows (ID T , K ts ). e RFID tag can use V 4 and M 3 in M 4 to authenticate the legitimacy of the DB. e mutual authentication property can be achieved by RAPUS + protocol.

Anonymity.
e most basic element of the secure protocol is anonymity. e personal information of the user is protected by a secure scheme so that the adversary has no access to any information. e protocol has achieved strong anonymity. During the registration part, the RFID tag used M � {ID Ti , K ts , AID} to identify the S through RFID-Reader. e messages M A1 � {AID Ti , N x , V 1 , T 1 } have been delivered to the S through a public channel in the authentication part. e adversary cannot attain the identity of the RFID tag even if the adversary gets the message M 1 for the reason that AID Ti is the one-time alias identity of T. e initial identity is kept encoded in AID Ti . It can only be encoded by the DB using K ts . erefore, the adversary cannot destroy the RFID tag's authentic identity. In this way, we achieve the anonymity of the protocol.

Traceability.
A safe protocol can protect any identity information of the participants from illegal users. e traceability of the RFID tag may be caused by identifying information. RAPUS + protocol cannot reveal the login information of the conversation that causes the security attack.
e protocol needs to use (N t , N r , r i ). It is impossible for the adversary to achieve any random number in the RFID system because the RFID tag's new one-time alias identity AID Ti has already been used. erefore, the protocol meets the untraceability.

Backward/Forward Secrecy.
It is essential for security protocols to ensure that the information transferred during a phase is not threatened and traced by the adversary, as it may generate defects in the certification phase between T and S. In our proposed scheme, the previous and next sessions will Journal of Healthcare Engineering not be affected, even if the identity ID T is lost. e encrypted AID Ti is updated in each new session to ensure it. erefore, the backward and forward secrecy of the RAPUS + protocol can be guaranteed.

Scalability.
In RAPUS + protocol, the detailed procedure is used to verify if any T is not performed by the RFID Server S. Oppositely, S disposes AID Ti to verify Tand makes a quick response to T. erefore, RAPUS + protocol gets more stable.

DoS Attacks.
For any random key which is in charge of verification or authentication of T, the protocol is not based on them. Instead, it is on the basis of AID Ti . Moreover, it is well encoded and renewed for each transaction. Hence, the proposed scheme defenses against DoS attack.

Replay Attacks.
In the replay attack, to authenticate S, the attacker may postpone and repeat the transferred information. (T, RE, DB) are included in RAPUS + protocol. For authentication, {M 1 , M 2 , M 3 , M 4 } are exchanged through the public channel. Accessible to the messages, the attackers try to launch the replay attacks.
Nevertheless, due to the messages delivered with the fresh timestamp T, the attempt will fail. e adversary request will be repulsed each time in the event of timestamp's ineffectiveness. Besides, the adversary cannot launch the attacks if it cannot calculate the parameters of the messages, because all messages' parameters are updated by the participants for every new session. RAPUS + protocol is able to resist replay attacks.

Location Tracking Attacks.
e authentic identity of the RFID tag is not delivered firsthand. erefore, it has been delivered in the encoded form for authentication between the RFID tag and S. And only the server can decrypt through its secret key. Besides, in every new session, the unpredictability of messages is continually renewed. erefore, the adversary cannot seek out the location. Any attempt to find the location will finally become a failure.

Impersonation Attacks.
To authenticate the server, adversary A holds up the messages of the valid Tand changes it. On this occasion, adversary A needs to issue the legitimate message request including (N y , R i , V 2 , M A1 , AID Ti ). In order to achieve it, AID Ti is encrypted and calculated and cannot be forged by adversary A.
Besides, to submit the legitimate request for certification as the valid T, adversary A demands various other timestamps and parameters too. Adversary A is impossible to know the real parameters used for certification. erefore, it has no ability to verify its validity as T to the DB. RAPUS + protocol for RFID system can resist any forgery attack reluctantly.

Stolen-Verifier Attacks.
All the validation and verification keys are encrypted and stored in the DB. Although the data and keys are both stolen from the DB, they cannot be decrypted and extracted by adversary A. Moreover, the original data saved in DB cannot be altered or modified by adversary A. erefore, the RAPUS + protocol resists the stolen-verifier attacks.

e Formal Analysis Using GNY Logic.
In order to guard against major attacks, the proposal of the security protocol design must be analyzed ahead of execution. e fundamental assumptions are presented in Table 3. e aims verified by RAPUS + protocol are as follows:  Journal of Healthcare Engineering e protocol messages generated by the parser are as follows: (G 1 , G 2 , G 3 , G 4 , G 5 , G 6 , G 7 , G 8 , G 9 ) are made to verify the RAPUS + protocol that has been certified. e sequence of logical assumptions is employed in the parser export to achieve the security goals by considering the various assumptions.
In M 1 , T 1 is the timestamp of T. Using the seeing rule, we can get (1) Applying the message-meaning rule and the previous step result, (2) Using the freshness-conjuncatenation rule and the previous step results, According to the jurisdiction rule and the previous step result, Applying the previous step result and the session-key rule, Using the nonce-verification rule, In M 2 , T 2 is the timestamp of R i . Applying the seeing rule, we get S j ⊲M1, N y : < N r > K rs , T2, V2.
According to the message-meaning rule and the previous step result, By the freshness-conjuncatenation rule and the previous step result, we get rough the jurisdiction rule and the previous step result, Applying the S 10 and the SK rule, According to the nonce-verification rule and the previous step result, In M 3 , T 3 is the timestamp of S j . rough the seeing rule, we get Applying the message-meaning rule and S 13 , rough S 14 and the freshness-conjuncatenation rule, According to the assumption S 15 and jurisdiction rule, Applying S 16 and the session-key rule, Using the nonce-verification rule, In M 4 , T 4 is the timestamp of R i . Applying the seeing rule, tag⊲V4, Zt < AIDT inew ≥ > ts , T4. (19) rough the message-meaning rule and S 19 , According to S 20 and the freshness-conjuncatenation rule, Applying the jurisdiction rule and S 21 , By the session-key rule, we get Finally, according to the nonce-verification rule, As a result, it is proved that (T, R i , S j ) achieve successful reciprocal certification and obtain the session-key agreement safely.

5.3.
e Protocol Verification Using AVISPA Tool. e AVISPA is the formal security protocol analysis tool. It uses the High-Level Protocol Specification Language (HLPSL) to specify the sequence of messages exchanged among different entities. e basic role is the module consisting of the action of each entity. e entities combine multiple basic roles into the composed role to interact with each other. e analyzed protocol's security goals are specified in the goal phase. It has four back ends, including OFMC, CL-Atse, SATMC, and TA4sp, which use different kinds of techniques to show whether the RAPUS + protocol is safe or not. e tool supplies tracking of the steps that lead to the attack and uses the Dolev-Yao intruder model that can eavesdrop, intercept messages, modify passing traffic, or insert bogus data. In our proposed scheme, we describe different entities' actions by defining the basic role and how the entities in the composed role interact with each other. e results present that our proposed scheme is "safe" against OFMC with regard to the security goal in Figure 4. Appendix A shows all source programs by AVISPA.

Performance Analysis and Comparison.
is phase conducts the comparative analysis between the RAPUS + protocol and the existing protocol. First of all, we compare the existing protocol with the RAPUS +protocol in terms of security requirements. Besides, based on the calculation cost, we compare the RAPUS + protocol with the existing protocol. Lastly, we compare the RAPUS + protocol with existing protocols in regard to model analysis.

Security Requirements.
is section analyzes the existing authentication protocols based on symmetric keys from the perspective of security requirements. Table 4 presents the comparisons between the proposed agreement and the existing agreement [1,10,15,[28][29][30]. Table 4 shows the insecurities of existing protocols [1,10,15,[28][29][30]. e result shows that only RAPUS +protocol can provide all the above security features, such as mutual authentication, untraceability, anonymity, the backward/forward secrecy, scalability, collision attacks, DoS attacks, replay attacks, location tracking attacks, stolenverifier attacks, database impersonation attacks, and reader impersonation attacks.
In the protocol proposed in [15], the computational cost of each (T, R, S) is 5 , 2 , and 7 , so the total cost is 14 . e protocol proposed in [29] needs 2 , 2 , and 3 for each (T, R, S), and the total cost is 7 . e cost of protocol  Journal of Healthcare Engineering demonstrated in [30] is 4 , 2 , and 6 correspondingly, amounting to 12 . In the protocol proposed in [10], each (T, R, S) generates 3 , 2 , and 5 cost correspondingly, so the total cost is 10 . e cost of protocol demonstrated in [28] is 2 , 3 , and 5 correspondingly, amounting to 10 . e protocol proposed in [1] needs 2 , 2 , and 4 +2Tse for each (T, R, S), so the total cost is 8 + 2Tse. In comparison, T requires 2 , the reader requires 1 , and the server requires 3 + 2Tse, and the total cost is 6 + 2Tse in the protocol proposed in this article. In general, the protocol proposed in this article has a relatively smaller computation cost and is the only one that can withstand all known attacks.

Comparisons of Model Analysis.
is section describes a model analysis of RAPUS + protocol with existing multitag authentication protocols in Table 6.
e results show that most protocols lack or have no model analysis. In the proposed authentication protocol, the secrecy attacks are analyzed by automatic cryptographic protocol verifier tools AVISPA. GNY logic is applied to verify the reciprocal certification.
6. Conclusion e auxiliary information system for epidemic prevention and control has integrated traditional medical systems with RFID technology. However, the antitampering of logistics data and management and control of epidemic materials are still challenges. In this article, we prove that the RAPUS protocol is susceptible to database impersonation attacks, reader impersonation attacks, and asynchronous attacks. en, the RAPUS + protocol is proposed. e security analysis of the RAPUS + protocol has been conducted through GNY logic, AVISPA model. Additionally, the comparisons of RAPUS + protocol with the existing protocols prove the superiority in security, computational cost, and model analysis. Based on RAPUS + protocol, the safety and management of epidemic prevention materials will be greatly improved.

Data Availability
e data served to support the findings of this study are contained within the article.

Conflicts of Interest
e authors declare no conflicts of interest.