*
Create and share pages like this: mail to web@snipek.com.
*

*Powered by Snipek.
Created at *

---------- Forwarded message ----------

From:**ISIT'2014** <isit2014-chairs@edas.info>

Date: Sat, Mar 29, 2014 at 7:31 PM

Subject: [ISIT'2014] Your paper #1569890297 ('On Topological Properties of Wireless Sensor Networks under the q-Composite Key Predistribution Scheme with On/Off Channels') has been accepted

To: Jun Zhao <junzhao@cmu.edu>

Dear Mr. Zhao,

We are happy to inform you that your paper

#1569890297: 'On Topological Properties of Wireless Sensor Networks under the q-Composite Key Predistribution Scheme with On/Off Channels'

submitted to the 2014 IEEE International Symposium on Information Theory,

has been accepted for presentation at the symposium and

for publication in the proceedings. The reviews are

attached below and can be found at

http://edas.info/showPaper.php?m=1569890297

We ask you to incorporate the comments of the reviewers in

preparing the final manuscript. Please also have a close look

at the following instructions:

1. The deadline for uploading the final manuscripts is

April 27, 2014, at 11:59pm EDT (New York time zone).

This deadline is strict and will not be extended.

Shortly after the deadline, uploads will no longer

be possible on EDAS. Please also note that every paper

must have at least one author registered by April 27.

(See below for registration instructions.)

2. You are encouraged to use LaTeX to format your final

manuscript, using the standard IEEE style. Please use

fonts no smaller than 10 points. For more information,

see the instructions posted at http://isit2014.org

(The instructions will be posted in early April.)

3. Remember to NOT include a sentence at the beginning

of the abstract saying that the paper is to be considered

for an IEEE Jack Keil Wolf ISIT Student Paper Award

(i.e., if you had this sentence in the initial submission,

please remove it both in the PDF of your final paper and

also in the abstract stored in EDAS).

4. Please upload the final version on EDAS. Clicking on

the link

http://edas.info/uploadPaper.php?m=1569890297

should take you to the upload page for this paper.

5. By the upload deadline, an IEEE Copyright form must

be provided for each accepted paper. The form should

be filled out electronically via the link that EDAS

provides during the upload.

In addition to the above, two more conditions need to be

satisfied in order for your paper to be included in the

ISIT'2014 Proceedings published on IEEE Xplore:

a. Every paper must have a registered author to appear

in the Proceedings. Up to 4 papers may be covered

by a single registration (student registration is OK).

Please register through the ISIT'2014 website at:

http://www.isit2014.org

(The registration page will be live in early April.)

The registration page will ask you for up to 4 paper

numbers. Please enter '1569890297' as one of them.

This must also be done by April 27. Please note that if

you are an IEEE member but not an IT Society member,

you can save more than the registration cost difference

by joining the IT Society (ieee.org/join) before registering.

b. Every paper must be presented at ISIT by one of its

authors. A paper with no presenter will be removed

from IEEE Xplore, unless there are extenuating

circumstances that need to be explained to the

General Chairs and approved by them.

Please note that if you require an invitation letter for a visa,

you should enter your passport number during registration.

After registering, you will receive an email with further

instructions for obtaining an electronic letter in PDF format.

You may also request a paper copy during the registration

process.

Thank you again for your contribution to ISIT. We look

forward to seeing you in Honolulu!

Sincerely,

Sae-Young Chung

Gerhard Kramer

Olgica Milenkovic

Urbashi Mitra

**************************************************************************

======= Summary review by TPC member 1 =======

> *** TPC recommendation: Given the reviews, what is your recommendation for this paper.

Accept (5)

> *** TPC Summary: Please give a justification for your recommendation, especially if

the review scores vary widely or if your recommendation differs significantly from those of the reviewers.

This is a very nice paper and the TPC agrees with the recommendations of the two reviewers.

======= Review 2 =======

> *** Suitability for ISIT: If correct, how interesting are the results for the ISIT crowd?

Quite interesting (4)

> *** Correctness: Do the results seem correct as far as you can tell?

Essentially correct (4)

> *** Novelty: How novel is the paper?

Reasonable novelty (3)

> *** Presentation: How well is the paper written?

Good (4)

> *** Recommendation: Assuming that 60% of papers are accepted, please give your

recommendation for including this paper in the ISIT'2014 program.

Accept (5)

> *** Confidence: How confident are you in your recommendation? This is important for us to know!

Rather confident (2)

> *** Strengths: What do you like about of this paper?

This paper investigates the topological properties (node degree) of WSNs that employ a key-predistribution scheme for establishment of secure communications. Although the first proposal of the key-predistribution scheme is already 12 years old, the analysis performed is relevant by addressing a hard problem where the number of required common keys between nodes is greater than 1, which appears to have not been solved yet.

> *** Weaknesses: What do you dislike about of this paper?

I feel that the paper could better motivate some of the implications of certain parameter choices from a practical standpoint. Although this is made, up to a certain point, in section III-D, I feel that a more hands-on explanation is missing. For example, it is mentioned that "K_n^2 / P_n = o(1) implies P_n >= 3K_n" - what is the practical implication of this if one wants to use this scheme to setup secure communication in WSNs?

W.r.t to the proofs, I strongly disagree with the practice of citing relevant proofs from unrevised technical reports (TR). I certainly did not go validate the proofs on the TR and, therefore, although the results in the paper seem correct I cannot (to the best of my knowledge) be 100% sure.

> *** Comments and Recommendation: Please give the reasoning for your overall

recommendation.

There are some minor quibbles that should be improved in case the paper gets accepted:

- provide a better practical feel of the implications of some choices (mentioned above)

- at some points it is mentioned that "K_n^2 / P_n = o(1) implies P_n >= 3K_n", while in others 2K_n appears instead -> clarify

- the relation between numerical and analytical results appears a bit loose -> more detail should be given about the process and how they relate

- at some point it is mentioned that "each h equals a particular non-negative number M", but M is never again mentioned

- in page 2, there are several concepts of probability: approaching, going, tending, and converging -> what is the difference between them?

- in formula (1) explain where does u=max[q,2Kn-Pn] comes from

- it is mentioned (page 1) that results about the node degree in [12] and [13] are even weaker than the analytical findings when the general q is set as 1 -> explain why

- the terminology "u" vs "q" in the appendix proofs is somewhat confusing -> clarify

- legend of fig. 1: analylsis -> analysis

======= Review 3 =======

> *** Suitability for ISIT: If correct, how interesting are the results for the ISIT crowd?

Very interesting (5)

> *** Correctness: Do the results seem correct as far as you can tell?

Definitely correct (5)

> *** Novelty: How novel is the paper?

Extremely novel (5)

> *** Presentation: How well is the paper written?

Excellent (5)

> *** Recommendation: Assuming that 60% of papers are accepted, please give your

recommendation for including this paper in the ISIT'2014 program.

Accept (5)

> *** Confidence: How confident are you in your recommendation? This is important for us to know!

Highly confident (3)

> *** Strengths: What do you like about of this paper?

The theoretical results of the paper are very interesting because they give another perspective to understanding the network topology under the q-composite KPS. The results of the computer simulations also demonstrate the correctness of the theoretical outcomes.

> *** Weaknesses: What do you dislike about of this paper?

My concern is the suitability of the Erdos-Renyi random graph model in modeling WSNs. This model has been extensively used, since the very first KPS proposed by Eschenauer-Gligor method in 2001, because of its simplicity and the existence of numerous theoretical results based on it. The problem with using this model in WSNs is the equal treatment of all nodes throughout the entire network in their chance for establishing links. This is a false assumption because of the limited communication range of sensor nodes, i.e., two neighbor nodes might have a chance of direct communication while two nodes that are far away from each other can never directly communicate. The main question is whether it is possible to replace the Erdos-Renyi random graph model with a more realistic model in the work performed by the authors.

> *** Comments and Recommendation: Please give the reasoning for your overall

recommendation.

======= Summary review by TPC member 4 =======

> *** TPC recommendation: Given the reviews, what is your recommendation for this paper.

Accept (5)

> *** TPC Summary: Please give a justification for your recommendation, especially if

the review scores vary widely or if your recommendation differs significantly from those of the reviewers.

The paper elicits strong reviews and seems to be an excellent fit.

From:

Date: Sat, Mar 29, 2014 at 7:31 PM

Subject: [ISIT'2014] Your paper #1569890297 ('On Topological Properties of Wireless Sensor Networks under the q-Composite Key Predistribution Scheme with On/Off Channels') has been accepted

To: Jun Zhao <junzhao@cmu.edu>

Dear Mr. Zhao,

We are happy to inform you that your paper

#1569890297: 'On Topological Properties of Wireless Sensor Networks under the q-Composite Key Predistribution Scheme with On/Off Channels'

submitted to the 2014 IEEE International Symposium on Information Theory,

has been accepted for presentation at the symposium and

for publication in the proceedings. The reviews are

attached below and can be found at

http://edas.info/showPaper.php?m=1569890297

We ask you to incorporate the comments of the reviewers in

preparing the final manuscript. Please also have a close look

at the following instructions:

1. The deadline for uploading the final manuscripts is

April 27, 2014, at 11:59pm EDT (New York time zone).

This deadline is strict and will not be extended.

Shortly after the deadline, uploads will no longer

be possible on EDAS. Please also note that every paper

must have at least one author registered by April 27.

(See below for registration instructions.)

2. You are encouraged to use LaTeX to format your final

manuscript, using the standard IEEE style. Please use

fonts no smaller than 10 points. For more information,

see the instructions posted at http://isit2014.org

(The instructions will be posted in early April.)

3. Remember to NOT include a sentence at the beginning

of the abstract saying that the paper is to be considered

for an IEEE Jack Keil Wolf ISIT Student Paper Award

(i.e., if you had this sentence in the initial submission,

please remove it both in the PDF of your final paper and

also in the abstract stored in EDAS).

4. Please upload the final version on EDAS. Clicking on

the link

http://edas.info/uploadPaper.php?m=1569890297

should take you to the upload page for this paper.

5. By the upload deadline, an IEEE Copyright form must

be provided for each accepted paper. The form should

be filled out electronically via the link that EDAS

provides during the upload.

In addition to the above, two more conditions need to be

satisfied in order for your paper to be included in the

ISIT'2014 Proceedings published on IEEE Xplore:

a. Every paper must have a registered author to appear

in the Proceedings. Up to 4 papers may be covered

by a single registration (student registration is OK).

Please register through the ISIT'2014 website at:

http://www.isit2014.org

(The registration page will be live in early April.)

The registration page will ask you for up to 4 paper

numbers. Please enter '1569890297' as one of them.

This must also be done by April 27. Please note that if

you are an IEEE member but not an IT Society member,

you can save more than the registration cost difference

by joining the IT Society (ieee.org/join) before registering.

b. Every paper must be presented at ISIT by one of its

authors. A paper with no presenter will be removed

from IEEE Xplore, unless there are extenuating

circumstances that need to be explained to the

General Chairs and approved by them.

Please note that if you require an invitation letter for a visa,

you should enter your passport number during registration.

After registering, you will receive an email with further

instructions for obtaining an electronic letter in PDF format.

You may also request a paper copy during the registration

process.

Thank you again for your contribution to ISIT. We look

forward to seeing you in Honolulu!

Sincerely,

Sae-Young Chung

Gerhard Kramer

Olgica Milenkovic

Urbashi Mitra

**************************************************************************

======= Summary review by TPC member 1 =======

> *** TPC recommendation: Given the reviews, what is your recommendation for this paper.

Accept (5)

> *** TPC Summary: Please give a justification for your recommendation, especially if

the review scores vary widely or if your recommendation differs significantly from those of the reviewers.

This is a very nice paper and the TPC agrees with the recommendations of the two reviewers.

======= Review 2 =======

> *** Suitability for ISIT: If correct, how interesting are the results for the ISIT crowd?

Quite interesting (4)

> *** Correctness: Do the results seem correct as far as you can tell?

Essentially correct (4)

> *** Novelty: How novel is the paper?

Reasonable novelty (3)

> *** Presentation: How well is the paper written?

Good (4)

> *** Recommendation: Assuming that 60% of papers are accepted, please give your

recommendation for including this paper in the ISIT'2014 program.

Accept (5)

> *** Confidence: How confident are you in your recommendation? This is important for us to know!

Rather confident (2)

> *** Strengths: What do you like about of this paper?

This paper investigates the topological properties (node degree) of WSNs that employ a key-predistribution scheme for establishment of secure communications. Although the first proposal of the key-predistribution scheme is already 12 years old, the analysis performed is relevant by addressing a hard problem where the number of required common keys between nodes is greater than 1, which appears to have not been solved yet.

> *** Weaknesses: What do you dislike about of this paper?

I feel that the paper could better motivate some of the implications of certain parameter choices from a practical standpoint. Although this is made, up to a certain point, in section III-D, I feel that a more hands-on explanation is missing. For example, it is mentioned that "K_n^2 / P_n = o(1) implies P_n >= 3K_n" - what is the practical implication of this if one wants to use this scheme to setup secure communication in WSNs?

W.r.t to the proofs, I strongly disagree with the practice of citing relevant proofs from unrevised technical reports (TR). I certainly did not go validate the proofs on the TR and, therefore, although the results in the paper seem correct I cannot (to the best of my knowledge) be 100% sure.

> *** Comments and Recommendation: Please give the reasoning for your overall

recommendation.

There are some minor quibbles that should be improved in case the paper gets accepted:

- provide a better practical feel of the implications of some choices (mentioned above)

- at some points it is mentioned that "K_n^2 / P_n = o(1) implies P_n >= 3K_n", while in others 2K_n appears instead -> clarify

- the relation between numerical and analytical results appears a bit loose -> more detail should be given about the process and how they relate

- at some point it is mentioned that "each h equals a particular non-negative number M", but M is never again mentioned

- in page 2, there are several concepts of probability: approaching, going, tending, and converging -> what is the difference between them?

- in formula (1) explain where does u=max[q,2Kn-Pn] comes from

- it is mentioned (page 1) that results about the node degree in [12] and [13] are even weaker than the analytical findings when the general q is set as 1 -> explain why

- the terminology "u" vs "q" in the appendix proofs is somewhat confusing -> clarify

- legend of fig. 1: analylsis -> analysis

======= Review 3 =======

> *** Suitability for ISIT: If correct, how interesting are the results for the ISIT crowd?

Very interesting (5)

> *** Correctness: Do the results seem correct as far as you can tell?

Definitely correct (5)

> *** Novelty: How novel is the paper?

Extremely novel (5)

> *** Presentation: How well is the paper written?

Excellent (5)

> *** Recommendation: Assuming that 60% of papers are accepted, please give your

recommendation for including this paper in the ISIT'2014 program.

Accept (5)

> *** Confidence: How confident are you in your recommendation? This is important for us to know!

Highly confident (3)

> *** Strengths: What do you like about of this paper?

The theoretical results of the paper are very interesting because they give another perspective to understanding the network topology under the q-composite KPS. The results of the computer simulations also demonstrate the correctness of the theoretical outcomes.

> *** Weaknesses: What do you dislike about of this paper?

My concern is the suitability of the Erdos-Renyi random graph model in modeling WSNs. This model has been extensively used, since the very first KPS proposed by Eschenauer-Gligor method in 2001, because of its simplicity and the existence of numerous theoretical results based on it. The problem with using this model in WSNs is the equal treatment of all nodes throughout the entire network in their chance for establishing links. This is a false assumption because of the limited communication range of sensor nodes, i.e., two neighbor nodes might have a chance of direct communication while two nodes that are far away from each other can never directly communicate. The main question is whether it is possible to replace the Erdos-Renyi random graph model with a more realistic model in the work performed by the authors.

> *** Comments and Recommendation: Please give the reasoning for your overall

recommendation.

======= Summary review by TPC member 4 =======

> *** TPC recommendation: Given the reviews, what is your recommendation for this paper.

Accept (5)

> *** TPC Summary: Please give a justification for your recommendation, especially if

the review scores vary widely or if your recommendation differs significantly from those of the reviewers.

The paper elicits strong reviews and seems to be an excellent fit.