A Lightweight and Privacy-Preserving Authentication Protocol for Healthcare in an IoT Environment

Round 1

Reviewer 1 Report

1. The authors should rewrite the abstract and make outstanding contributions.

2. The goal is mutual authentication between the user and the sensor node. Therefore, the authors should add the analysis of mutual authentication.

3. To achieve user-friendliness, the proposed scheme should allow users to change their passwords locally. So, the paper should add the password update phase.

4. In the registration process, the number 1 is combined when calculating Qi. If it is more reasonable to choose a random value?

5. Table 1 of the paper is redundant and needs to be modified. The chart does not correspond to what is described in the paper. The author should check it carefully and make corrections.

Minor editing of English language required

Author Response

Reviewer:1

Comments and Suggestions for Authors

1. The authors should rewrite the abstract and make outstanding contributions.

Response: Thank you for your valuable suggestion. We have revised the abstract.

2. The goal is mutual authentication between the user and the sensor node. Therefore, the authors should add the analysis of mutual authentication.

Response: Thank you for your valuable suggestion. We have added the analysis of mutual authentication in Section 5.

3. To achieve user-friendliness, the proposed scheme should allow users to change their passwords locally. So, the paper should add the password update phase.

Response: Thank you for your valuable suggestion. We have added the password update phase in Section 3.

4. In the registration process, the number 1 is combined when calculating Qi. If it is more reasonable to choose a random value?

Response: Thank you for your valuable comment. Generally speaking, choosing random numbers is more reasonable, but considering the avalanche effect of hash functions, combining the number 1 or random number in hash is functionally the same. If we replace 1 with a random number, it will need to increase the additional cost of storing random number.

5. Table 1 of the paper is redundant and needs to be modified. The chart does not correspond to what is described in the paper. The author should check it carefully and make corrections.

Response: Thank you for your valuable comment. We have checked and corrected Table 1.

 

Comments on the Quality of English Language

Minor editing of English language required

Response: Thank you for your valuable comment. We have proofread the manuscript, and corrected the typos and syntax errors.

Author Response File: Author Response.docx

Reviewer 2 Report

In this work, the authors presented a method of using the Elliptic Curve to implement an authentication protocol for Healthcare IoT system. The authors provided an appropriate literature review. The technical detail is well presented with a clear illustration. But there are some points that need to improve:

·       English is not fluency. The contents are only cited by numbering or bullet points. It makes it hard for the readers to follow the content of the paper. It requires a connection among sections and paragraphs. 

·      The contributions are not enough. The authors target to solve the issues in the paper of Masus et al. Therefore, despite the design scheme having some improvements over previous, it is not enough to contribute to the body of knowledge. I suggest that the authors should work more on this project before publishing.

·       The design of the provided authentication protocol is based on Elliptic Curve Cryptography (ECC), but there is not any analysis on ECC over the technical contents. The authors should explain about ECC and why it is suitable for this work.

·      The three highlighted keywords of this paper are “consideration to security”, “privacy protection”, and “lightweight”. For “lightweight” aspects, the author only analyzes the computational costs, which are much more related to the performance.  Therefore, I suggest that the authors should analyze more aspects (such as memory, computational power, etc.) or replace the term.

- In section 1, recent studies should be added. I found that [a] is a reference proposed a solution about Lightweight User Authentication Scheme.  As a result, what are the main differences between those two works?

[a] Kim, K.; Ryu, J.; Lee, Y.; Won, D. An Improved Lightweight User Authentication Scheme for the Internet of Medical Things. Sensors 2023, 23, 1122. https://doi.org/10.3390/s23031122.

- Table 4, section 6 (performance comparison): study in [30] and this work and have similar attacks/properties. The former takes 18T_H + 6T_M; meanwhile the latter requires 19T_H + 4T_M . Please explain which is the main factor when considering computational costs.

- Should add R_i, C_i  into Table 1 to be more clearly.

- What is the difference between R_i and Rep? I didn’t see the computation of Rep() function. Thus, why can the device in registration phase store Rep value? (see line 196, or Figure 2).

- The author should consider replacing figures 2-4. The current figures are unsuitable when demonstrating used phases in a protocol. The top-to-bottom dimension illustrates the time axis, so it should be clear which step/phase is taken before the next step/phase.

- The author should equation format for all used equation. The current style makes the readers difficult to follow the content of the article.

I suggest that the author can follow the method for illustrating a protocol and equation style in [a] or [33].

 

1) Page 2, line 85: “Rrecently” is an incorrectly used word.

2) Page 3, line 88: add reference for Masud’s work.

2) Add a blank space after the parentheses mark, for example: lines 112, 117, 120, and 123 (page 3)

3) Page 6, line 201: use the bullet format for Step SR2

4) Consider using center alignment for all used functions from Eq. 1 to Eq. 11

2) Remove dot symbol between Figures and their index. For example, use “Figure 2” instead of “Figure.2” (Page 5, line 168)

3) Page 5 lines 172 and 174: fix the incorrect figures’ index, they are “Figure 2” and “Figure 3”.

3) Page 4, Table 1: first four rows and last four rows are the same. Authors must remove them to avoid duplicate content.

4) Page 9, lines 333-334: check the equation in these lines.

English is not fluency. The contents are only cited by numbering or bullet points. It makes it hard for the readers to follow the content of the paper. It requires a connection among sections and paragraphs.

Author Response

Reviewer:2

Comments and Suggestions for Authors

In this work, the authors presented a method of using the Elliptic Curve to implement an authentication protocol for Healthcare IoT system. The authors provided an appropriate literature review. The technical detail is well presented with a clear illustration. But there are some points that need to improve:

• English is not fluency. The contents are only cited by numbering or bullet points. It makes it hard for the readers to follow the content of the paper. It requires a connection among sections and paragraphs. 

Response: Thank you for your valuable comment. We have proofread the manuscript, and corrected the typos and syntax errors. On the other hand, we have revised the inappropriate citation method.

•     The contributions are not enough. The authors target to solve the issues in the paper of Masus et al. Therefore, despite the design scheme having some improvements over previous, it is not enough to contribute to the body of knowledge. I suggest that the authors should work more on this project before publishing.

Response: Thank you for your valuable comment. The goal of this article is to address the security issues of the existing protocols for IoT based healthcare, such as Masud et al.’s protocol. As we know, many lightweight authentication protocols for IoT based healthcare have been proposed in recent years, but most of them may suffer from one or more security problems, especially, few protocols can resist sensor node captured attacks and achieve n-factor secrecy, which leads to unauthorized personnel being able to access the patient's physiological data, and obtain patients’ privacy. Therefore, a lightweight and privacy-preserving authentication protocol for healthcare based on Elliptic Curve Cryptography (ECC) and Physical Unclonable Function (PUF) is proposed to surmount the above obstacles. We design dynamic anonymity policy to achieve users’ anonymity and unlinkability, use PUF to protect information stored in users’ devices and sensor nodes. In addition, higher security such as three-factor secrecy, perfect forward secrecy, resistance to sensor node captured attacks and update asynchronous attacks are guaranteed. The proposed protocol is proved to be secure under the random oracle model, and keeps lightweight computing efficiency.

Therefore, the designed dynamic anonymity strategy, the proposed protocol can resist sensor node captured attacks, achieve three-factor secrecy, keep lightweight computing efficiency, and provably secure, are valuable.

•      The design of the provided authentication protocol is based on Elliptic Curve Cryptography (ECC), but there is not any analysis on ECC over the technical contents. The authors should explain about ECC and why it is suitable for this work.

Response: Thank you for your valuable comment. The security of the proposed scheme is based on Elliptic curve discrete logarithm problem (ECDLP) and Computational Diffie-Hellman problem (CDHP), ECC is currently a cryptographic tool that efficiently implements these two problems. In informal security analysis, we explained the role of ECC in security assurance. The efficiency comparison indicates that ECC has lower overhead.

•     The three highlighted keywords of this paper are “consideration to security”, “privacy protection”, and “lightweight”. For “lightweight” aspects, the author only analyzes the computational costs, which are much more related to the performance.  Therefore, I suggest that the authors should analyze more aspects (such as memory, computational power, etc.) or replace the term.

Response: Thank you for your valuable comment. Yes, lightweight generally refers to execution efficiency, especially low computational costs. Because many schemes (such as Masud et al. ’s scheme [32] and Kim et al.’s scheme [34]) only analyze computational complexity when analyzing lightweight, we have also only analyzed computational complexity. We have modified it to lightweight computational efficiency.

- In section 1, recent studies should be added. I found that [a] is a reference proposed a solution about Lightweight User Authentication Scheme.  As a result, what are the main differences between those two works?

[a] Kim, K.; Ryu, J.; Lee, Y.; Won, D. An Improved Lightweight User Authentication Scheme for the Internet of Medical Things. Sensors 2023, 23, 1122. https://doi.org/10.3390/s23031122.

Response: Thank you for your valuable comment and providing the article [a]. We have cited it as reference [34], and compared it with our protocol in terms of security and computational efficiency.

According to comparisons, we can know that the computation cost of Kim et al.’s scheme [34] lower than ours, but their scheme is insecure. It is well known, the security of protocols is the most important, and the lightweight protocol without ensuring security is not significant. The security analysis of their scheme is as follows:

In Kim et al.’s scheme [34], because TID^{new}_i= TID_i \xor GM_2\xor GM_3, where TID_i , GM_2 and GM_3 are public transmitted, so the attackers can trace the user because TID_i and TID^{new}_i are linkable. On the other hand, the attackers can use side-channel attack to obtain U_iM_2 and U_iM_3, and obtain GM_6 from public channel, and can compute S^1_i= GM_6\xor TID^{new}_i, S^2_i= S^1_i \xor U_iM_2 \xor U_iM_3, then the attacker can launch user impersonation attacks and sensor node impersonation attacks. Kim et al.’s scheme also cannot achieve session key secrecy, perfect forward secrecy, two-factor secrecy and known session key secrecy, and cannot resist sensor node captured attacks. Here, we omit the details of these attack processes.

- Table 4, section 6 (performance comparison): study in [30] and this work and have similar attacks/properties. The former takes 18T_H + 6T_M; meanwhile the latter requires 19T_H + 4T_M . Please explain which is the main factor when considering computational costs.

Response: Thank you for your valuable comment. In [30], additional ECC operations are used to hide the user's identity. In the proposed scheme, a dynamic anonymity strategy has been proposed to replace the anonymity strategy in [30], and achieving lower overhead.

- Should add R_i, C_i  into Table 1 to be more clearly.

Response: Thank you for your valuable suggestion. We have added them in Table 1.

- What is the difference between R_i and Rep? I didn’t see the computation of Rep() function. Thus, why can the device in registration phase store Rep value? (see line 196, or Figure 2).

Response: Thank you for your valuable comment. Ri is a unique and unpredictable response value of PUF. Fuzzy Extractor algorithm includes generation algorithm and reproduction algorithm Rep(). The function of generation algorithm is to input the user's biological information, which can output a random value and a deterministic information. The function of reproduction algorithm is to input the user's biological information with allowable deviation range and the deterministic information, which can recover this random value. Because the random value will be used in Login and Authentication phase, so Rep() and a deterministic information should be stored into the device in registration phase.

- The author should consider replacing figures 2-4. The current figures are unsuitable when demonstrating used phases in a protocol. The top-to-bottom dimension illustrates the time axis, so it should be clear which step/phase is taken before the next step/phase.

Response: Thank you for your valuable comment. The presentation method of our flowcharts (figures 2-4) is basically consistent with that of many references (such as Masud et al. ’s scheme [32] and Kim et al.’s scheme [34]).

- The author should equation format for all used equation. The current style makes the readers difficult to follow the content of the article.

I suggest that the author can follow the method for illustrating a protocol and equation style in [a] or [33].

Response: Thank you for your valuable comment. Most of the formulas in our protocol are arranged on a separate line and centered, which may make it easier for readers to read.

1) Page 2, line 85: “Rrecently” is an incorrectly used word.

2) Page 3, line 88: add reference for Masud’s work.

2) Add a blank space after the parentheses mark, for example: lines 112, 117, 120, and 123 (page 3)

3) Page 6, line 201: use the bullet format for Step SR2

4) Consider using center alignment for all used functions from Eq. 1 to Eq. 11

2) Remove dot symbol between Figures and their index. For example, use “Figure 2” instead of “Figure.2” (Page 5, line 168)

3) Page 5 lines 172 and 174: fix the incorrect figures’ index, they are “Figure 2” and “Figure 3”.

3) Page 4, Table 1: first four rows and last four rows are the same. Authors must remove them to avoid duplicate content.

4) Page 9, lines 333-334: check the equation in these lines.

Response: Thank you for your valuable suggestions. We have checked and corrected them.

 

Comments on the Quality of English Language

English is not fluency. The contents are only cited by numbering or bullet points. It makes it hard for the readers to follow the content of the paper. It requires a connection among sections and paragraphs.

Response: Thank you for your valuable comment. We have proofread the manuscript, and corrected the typos and Syntax errors. On the other hand, we have revised the inappropriate citation method.

Author Response File: Author Response.docx

Reviewer 3 Report

The paper has touched the issue of unauthorized authentication to wireless sensor network of healthcare IoT devices. It has first described the issue and state-of-the-art in this area by explaining the existing works and their drawbacks against various security attacks. Then they have proposed a lightweight privacy-preserving protocol for secure authentication using ECC to be able to protect those devices from various unauthorized authentication attacks by protecting user’s information stored in those devices. They have used random oracle model to prove the protocol security and compared it with the existing systems. To make the proposed protocol understandable and clear, all the notations used in the protocol have been separately listed with their explanations. Then the protocol has been illustrated in its various steps and phases, followed by a formal security analysis. Also, an informal security analysis has been added into the paper by considering different attacks to the system. At the end, a comparative study has been added for the protocol performance as well as comparing it with the existing works against various attacks.

 

The paper is very-well written. Introduction and proposed design are clearly presented. Security analysis of the proposed method is satisfiable.

 

Some minor editing required. For instance, it would be better to replace "private-preserving" with "privacy-preserving" inside the abstract.

Author Response

Reviewer:3

Comments and Suggestions for Authors

The paper has touched the issue of unauthorized authentication to wireless sensor network of healthcare IoT devices. It has first described the issue and state-of-the-art in this area by explaining the existing works and their drawbacks against various security attacks. Then they have proposed a lightweight privacy-preserving protocol for secure authentication using ECC to be able to protect those devices from various unauthorized authentication attacks by protecting user’s information stored in those devices. They have used random oracle model to prove the protocol security and compared it with the existing systems. To make the proposed protocol understandable and clear, all the notations used in the protocol have been separately listed with their explanations. Then the protocol has been illustrated in its various steps and phases, followed by a formal security analysis. Also, an informal security analysis has been added into the paper by considering different attacks to the system. At the end, a comparative study has been added for the protocol performance as well as comparing it with the existing works against various attacks.

 

The paper is very-well written. Introduction and proposed design are clearly presented. Security analysis of the proposed method is satisfiable.

 

 

Comments on the Quality of English Language

Some minor editing required. For instance, it would be better to replace "private-preserving" with "privacy-preserving" inside the abstract.

Response: Thank you for your valuable comment. We have proofread the manuscript, and corrected the typos and Syntax errors. On the other hand, we have corrected it.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

There are some points that could be improved to enhance reading satisfaction.

·         Section 1.2 should not be split into a discrete section. It is better to summarize all design goals in a single paragraph and it should be placed before the contributions. In addition, it is inappropriate when putting the paragraph “The organization of this paper is as follows, …” in “Design goals”.

·         In section 4, the bullets before “Definition …” redundant. The “Definition 1”, “Definition 2”, … are already an enumerate, there are not need an additional bullet. The same things should be fixed for section 3.2 and 3.3 (as Step 1, 2, 3…).

·          In section 4.1, 4.2, all the equations should be put in new line. For example, the equations in line 316, 324, 328, 332 in page 9.

·         Please the alignment of equations in section 4.2.

·         Finally, despite the conclusion is well summarized in the paper, I suggest that the authors should rewrite it by a more impressive way.

·         There are many English corrections that need to be fixed:

o    In page 1, line 37, a space is needed between [1] and “proposed”.

o    In page 2, line 48 and 49, when listing more than two items, the structure should be “item 1, item2, item3, .., and last item …”.

o    Page 2, line 80, “Shuai et al.’ s” should be “Shuai et al.’s”.

o    In page 6, line 186 and 194, the alignment is different despite the two line has the same property.

o    In page 8, line 254, “The authentication and the session key negotiation are completed.”  should be “The authentication and the session of key negotiation is completed.”

o    In page 8, line 256, the phrase should be “… replace sth by sth”, not “.. replace sth with sth”.

o    In page 9, line 338, “Proof. …” should be “Proof: ...”

o    In page 405, page 11, “user’ s biometric” should be “user’s biometric”.

- The explanation related to the concern “Table 4, section 6 (performance comparison): a study in [30] and this work and have similar attacks/properties. The former takes 18T_H + 6T_M; meanwhile the latter requires 19T_H + 4T_M …”is incomplete. The author should address this in detail.

- Most of the formulas in the updated manuscript were arranged on a separate line and centered, which may make it easier for readers to read. However, there are still numerous equations that are not to be fixed, such as equations on lines 344, 351, 358, etc. The authors should carefully proofread whole the manuscript.

- The author responded to my previous concern about the Fuzzy Extractor algorithm, including the generation algorithm Gen() and reproduction algorithm Rep(). Although their function was explained clearly in the responding letter, I think that the author should add information in the updated manuscript. For example, I didn’t see the formula of the reproduction algorithm Rep() in subsection 3.2.

Moderate editing of English language required.

Author Response

Response to the Rewiewer 2:

 

Thank you very much for your careful review and valuable comments on our article, we have revised them as follows.

 

Comments and Suggestions for Authors

There are some points that could be improved to enhance reading satisfaction.

• Section 1.2 should not be split into a discrete section. It is better to summarize all design goals in a single paragraph and it should be placed before the contributions. In addition, it is inappropriate when putting the paragraph “The organization of this paper is as follows, …” in “Design goals”.

Response: we have summarized all design goals in a single paragraph and placed it before the contributions.

• In section 4, the bullets before “Definition …” redundant. The “Definition 1”, “Definition 2”, … are already an enumerate, there are not need an additional bullet. The same things should be fixed for section 3.2 and 3.3 (as Step 1, 2, 3…).

Response: we have revised it.

• In section 4.1, 4.2, all the equations should be put in new line. For example, the equations in line 316, 324, 328, 332 in page 9.

Response: we have revised it.

• Please the alignment of equations in section 4.2.

Response: we have revised it.

 Finally, despite the conclusion is well summarized in the paper, I suggest that the authors should rewrite it by a more impressive way.

Response: we have revised it.

• There are many English corrections that need to be fixed:

o    In page 1, line 37, a space is needed between [1] and “proposed”.

Response: we have revised it.

o    In page 2, line 48 and 49, when listing more than two items, the structure should be “item 1, item2, item3, .., and last item …”.

Response: we have revised it.

o    Page 2, line 80, “Shuai et al.’ s” should be “Shuai et al.’s”.

Response: we have revised it.

o    In page 6, line 186 and 194, the alignment is different despite the two line has the same property.

Response: we have revised it.

o    In page 8, line 254, “The authentication and the session key negotiation are completed.”  should be “The authentication and the session of key negotiation is completed.”

Response: we have revised it.

o    In page 8, line 256, the phrase should be “… replace sth by sth”, not “.. replace sth with sth”.

Response: we have revised it.

o    In page 9, line 338, “Proof. …” should be “Proof: ...”

Response: we have revised it.

o    In page 405, page 11, “user’ s biometric” should be “user’s biometric”.

Response: we have revised it.

- The explanation related to the concern “Table 4, section 6 (performance comparison): a study in [30] and this work and have similar attacks/properties. The former takes 18TH + 6TM; meanwhile the latter requires 19TH + 4TM …”is incomplete. The author should address this in detail.

Recall the question of Round 1: - Table 4, section 6 (performance comparison): study in [30] and this work and have similar attacks/properties. The former takes 18T_H + 6T_M; meanwhile the latter requires 19T_H + 4T_M . Please explain which is the main factor when considering computational costs.

Response: In reference [30] and our proposed scheme, the factors that affect computational efficiency are point multiplication on elliptic curve, symmetric encryption, hash function, and XOR operations. However, point multiplication on elliptic curve requires more time than other operations. Therefore, reducing the number of point multiplication operations on elliptic curve is the main factor in considering protocol computational efficiency

 

- Most of the formulas in the updated manuscript were arranged on a separate line and centered, which may make it easier for readers to read. However, there are still numerous equations that are not to be fixed, such as equations on lines 344, 351, 358, etc. The authors should carefully proofread whole the manuscript.

Response: we have revised it.

- The author responded to my previous concern about the Fuzzy Extractor algorithm, including the generation algorithm Gen() and reproduction algorithm Rep(). Although their function was explained clearly in the responding letter, I think that the author should add information in the updated manuscript. For example, I didn’t see the formula of the reproduction algorithm Rep() in subsection 3.2.

Response: Because the function of generation algorithm Gen() is to input the user's biological information, which can output a random value and a deterministic information. The function of reproduction algorithm Rep() is to input the user's biological information with allowable deviation range and the deterministic information, which can recover this random value.

In the Registration Phase, user inputs his/her biological information, and use Gen() to output a random value and a deterministic information. Then the user uses random value to protect some secret parameters. In Mutual Authentication and Key Agreement Phase, the user inputs his/her biological information and uses Rep() and the deterministic information to recover the random value, and uses the random value to recover the protected secret parameters. So Gen() is used in Registration Phase, and Rep() is used in Mutual Authentication and Key Agreement Phase.

 

Author Response File: Author Response.docx

Round 3

Reviewer 2 Report

- The author said that they had summarized all design goals in a single paragraph and placed it before the contributions. Thus, there is only one subsection in this version. The author should clear the title of Section 1.1 or select an alternative method for presenting this content.

- Fix the name of subsection 5.3 (page 11)

- Move Table 2’s caption into page 4, Table 3’s caption into page 13

- Consider using one style of numbering for all subsections. The authors used three different formats: see subsection 2.1 (line 127), subsection 2.2 (line 141), subsection 3.1 (line 157), etc.

Table 1 and 3 should be placed in a same page.

Author Response

Dear Editor and reviewers,

 

Thank you very much of your valuable comments, our responses are as follows.

 

- The author said that they had summarized all design goals in a single paragraph and placed it before the contributions. Thus, there is only one subsection in this version. The author should clear the title of Section 1.1 or select an alternative method for presenting this content.

Response: We have removed the title of section 1.1 and highlighted the Design goals, Motivations and Contributions with emphasis marks.

- Fix the name of subsection 5.3 (page 11)

Response: We have revised it.

- Move Table 2’s caption into page 4, Table 3’s caption into page 13

Response: We have revised them.

- Consider using one style of numbering for all subsections. The authors used three different formats: see subsection 2.1 (line 127), subsection 2.2 (line 141), subsection 3.1 (line 157), etc.

Response: We have revised them.

Table 1 and 3 should be placed in a same page.

Response: We have modified the error numbers of two tables and placed them in the appropriate positions.

We hope the Editor and Reviewers will be satisfied with this revisions.

Qi XIE

7 Sep. 2023