kumoh national institute of technology
Networked Systems Lab.

Review

NSL > Education> Review
Long - [Computers Standards & Interfaces Journal]
By :
Date : 2016-12-16
Views : 58

Reviewers' comments:


Reviewer #1: The main contribution of the paper is a heuristic algorithm eCoRSS, which is to determine two sets: main-assistant node set and sub-assistant node set. The objective is to select reliable relaying nodes in order to enhance the throughput, end-to-end delay, and energy efficiency. Comparing to existing studies, this paper took into account SNR. The organization is good, however there are following points that is not clear to me

1. Why did not the authors formulate the problem in the shape of an optimization problem? In my opinion, it will be clearer if the authors formulate the problem as an assistant node selection optimization problem, which is to determine main-assistant and sub-assistant nodes. The performance of heuristic alg., consequently, should be compared to the solution of optimization-based approach.
2. The simulation looks weird. The authors said that 100 nodes are distributed randomly, so the simulation should be run several times with different seeds. But, the results do not look like that. There is no confidential interval. Moreover, the number of hops is quite small. The transmission range is unreal, please give an example/explanation. What is the transmission power?
3. What is the value of Ebit in eq. 26?



Reviewer #2: The paper propose a novel cooperative mechanism used for selecting the overlay nodes in the MWN (eCoRSS). As presented in the paper, the science quality is quite high to analysis and consider the performance of network. The simulation results convinces the proposed algorithms in terms of throughput, E2E delay and energy consumption. However, there are many points needed to demonstrate clearly in the algorithms:

About wireless channel model for WSNs:
The authors should explain why they use the Rayleigh fading channel but other models and which case that network operates practically, the selected channel model is suitable? In fact, a majority of WSNs use the log normal model for considering.

About the definition and selection of assistant nodes:

1- Which node determine the assistant nodes? (relay nodes or source nodes? or by themselves)
2- The assistant nodes play roles of orverlaying or not?
3- In what cases, the assistant nodes perform their roles?

About the simulation results:

In some the figures (Fig. 9 and 10) presented in the paper, the results obtained from the analysis is shown. However, it is not clear how to get that results by analysis. Do that mean to calculate from the equations? It seems that the paper should add one section Analysis to verify the results.