kumoh national institute of technology
Networked Systems Lab.

Review Comment

NSL > Education> Review Comment
Hak-hui - ETFA2017
By :
Date : 2017-06-19
Views : 169

Reviewer 1
Overall evaluation: 0
The paper presents the implementation of a wireless network for the management of the sensors inside a greenhouse and their connection to the smart farm administration system.

There are several points along the paper that needs to be clarified. The structure of the paper seems quite standard, therefore the authors should clarify which is the novelty and the advantages of their solution. The main motivation for this work seems to be the poor communication due to plants, but it is not clear how hte proposed solution can solve this problem if it exists, please clarify. No transmission analysis or loss packet rate is reported.

The language needs to be deeply revised.

---------------------------------------------------------------------------------

Reviewer 2
Overall evaluation: -2
The paper "Design and Implementation of Smart Farm System with LoRa and IEEE 802.11ac" aims to describe a novel routing algorithm for wireless sensor connections in smart greenhouses. In this, it fails to deliver a convincing motivation and a comprehensible description of the algorithm. Furthermore, out of the 3.5 pages, ca. 1.5 pages are figures of marginal use to the reader.

ABSTRACT

Major issues:
The abstract claims to "prove stability of IEEE 802.11ac ad- hoc network in the plants and point to point networks" and that the paper will "provide the optimal solution to various greenhouses". The paper, however, neither does contain any proof (or evaluation), nor does it define a notion of optimality regarding greenhouse connection performance. While a work-in-progress paper may not require an evaluation, claiming one that is not delivered is not acceptable.

INTRODUCTION

Major issues:
The argument that plants could be in the way of the wireless communication path is a weak one: usually wifi can convey messages through concrete. Why are plants a problem that must be tackled? The paper is lacking any information of the brevity of the problem. If, as claimed, Korean farmers are mostly "illiterates", not being able to use wireless sensor networks, why should they be able to use WiFi? This argument doesn't make sense either. Moreover, Table 1 and Fig. 2 appear to be irrelevant to the paper. Also, both are not explained in the text, but only their existence is referenced.

Minor issues:
- Many definitive and indefinite articles missing.
- Please insert a blank between citations: "802.11ac [5]" is better readable than "802.11ac[5]" and do that consistently (cf. "paper [3]".
- Fig. 1 is too large for the conveyed information.
- Sentence "Tomatoes in the greenhouse" is missing a verb.
- The abbreviation WSN is not introduced.
- What is an "Intel Edison" and why is it relevant (but not referenced)?

PROPOSED SFNS

Major issues:
The paper does not explain how head nodes are elected, what the issues with multiple head nodes are, and how a node group is repaired to revert to having only a single head node. If this is not addressed, the network can diverge to consist of head nodes only. What is the effect of that? The example -- despite having the benefit of being illustrated using a sequence diagram -- is not explained comprehensibly:
- the node groups [1/*] and [N-1/*] do not contribute to the example
- which node becomes head node after disconnecting [N/3] is not explained
- the textual description ends prior to the Greenhouse Administration System requesting data from node [N/3] (Fig. 5)

Minor issues:
- The abbreviation RSSI is not introduced.
- Few definitive articles missing.

IMPLEMENTATION

Major issues:
Figure 6 is blurry, incomprehensible, and hence useless. The textual description of the application yielding "listview control and slider controls to show data from sensors and to control actuators." does not help either.

Minor issues:
- Fig. 8 conveys basically the same information than Fig. 1, but depicts an antenna instead of the greenhouse: Why?

OVERALL

The motivation in unclear as both claimed arguments are not supported by anything and the algorithm is not discussed properly as pointed out above. Hence, appreciating the contribution is complicated. Also, in the current form, this is not fit for publication.

---------------------------------------------------------------------------------

Reviewer 3
Overall evaluation: -2
Summary
The authors describe a 802.11ac networked smart farm setup for collecting parameters of plants and controlling watering, ventilation, etc. For this purpose, they develop a test bed integrating some Intel Galileo boards and visualized the data at a web front-end.

Strength
Smart farming is an important topic. The adjustment of a suitable communication infrastructure in combination with a fault tolerant protocols is important for this large scale and highly dynamically scenarios. The topic meets the call of ETFA 2017.

Weakness
I did not find the specific contribution of the paper. The authors address greenhouse scenarios, but they miss to identify and to explain the specific conditions here. I don't think that plants in the line of sight effect the signal significantly. Here we need to know
- which bandwidths have to be managed,
- which level of jitter and delay can be tolerated or
- how many nodes should be located in the scenario.
Based on this technical analysis a well-founded selection of hard- and software is possible. This includes an evaluation of existing mesh protocols!

At the moment we just find some strange statements related to the capabilities of farmers. Where is the relation between the alliteration of a user and the chosen WSN or WLAN infrastructure?
The mentioned scenario addresses typical WSN domains and the decision for a WLAN setup is not convincing.

The language and grammar of the paper have to be revised carefully.
 
Ʒ ̺
     Hong-Keu ETFA 2017
Ʒ  Rusyadi - ETFA 2017