A Reliable Transport Scheme for Human Opportunistic Networks

Round 1

Reviewer 1 Report

The paper can be accepted.

Reviewer 2 Report

Almost all comments have been replied for the paper "A Reliable Transport Scheme for Human Opportunistic Networks". The author's work seems satisfactory. 

 

There no further comments. 

 

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

The authors should address the following issues:

• The authors should state where Opportunistic Networks are working nowadays. And if they are not, why.
   
• The references sections state things like:

At present, existing reliable transport schemes in ONs are most designed
for deep space communication [12,13].

while these references are extremely old, so the expression "At present" is not correct.

The same with "In recent years", for references 16 and 17.

- You state that there are a few literatures till date consider the resource consumption in the minimization of
delay. Authors should discuss how the following paper, that also addresses
the problem of energy effciency in ON is different/better/worse from their aproach:

Borrego, Carlos, Joan Borrell, and Sergi Robles. "Efficient broadcast in opportunistic networks using optimal stopping theory." Ad Hoc Networks 88 (2019): 5-17.

- In section 4.2, the algorithm should be formatted using latex.

You say:

The inter-meeting interval between each node pair is generated using exponential
distribution with rate parameter λ,

How do you set the contact rate in TheOne? The One does not provide the way to do this.
You just have contacts and the rate is obtained, not set.

- Figures should be vectorised.

- Graphs should contain units. Very important.

- Grammar should be reviewed. It is full of grammar errors, for example:

for human opportunistic network that should be "for human opportunistic networks"

Reviewer 2 Report

The authors have presented a reliable transport scheme for the Opportunistic Networks. The topic is very important. Transport layer does have issue for MANET. 

I find the concept and writing interesting and well document. Novelty is good.

Comments: 

Since other transport layer algorithm interpret delay as packet loss most of the time. It would be interesting to see a comparison of the proposed scheme with recent transport algorithms.  

Reviewer 3 Report

Dear Authors,

Thanks for submitting this work for our consideration and for your hard work. However, at the stage the paper is in, I would recommend submitting it to a conference after properly addressing the following aspects:

• Please spend more space and focus on the proper description of the hypothesis assumed in this work. The network model seems to be a set of nodes connected in a single line but it is not explicitly drawn nor said in the paper. This is an important aspect to understand both the Mathematical model and the results. 
• You are considering wireless communication in the abstract, i.e., you don't consider any specific wireless technology. However, an intrinsic and mandatory aspect to address in a research work when considering such networks is the increase of random losses, which you don't mention at all.
• Please also consider that 60KB packets are simply out of any real application in wireless environments.  
• The proposed protocol is a simple stop-and-wait transport protocol: a fixed amount of packets are sent, then the source waits for ACKs that indicate the number of missing packets (at the destination) to properly decode the file. Consider that Ts,r and Tw,r might be the same unless properly justified in the paper. 
• Congestion is mentioned in the paper, but external load increase and the consequences are not evaluated. 
• What exactly is epidemic propagation in the considered context? Remember you are assuming a longitudinal, single line of nodes as your network model. Moreover, if you know the file size, the propagation delay, and that no losses are present, why is it that you don't present the optimal Mr and number cycles to contrast against all proposed tests?

Regards.