5G Based on MNOs for Critical Railway Signalling Services: Future Railway Mobile Communication System

Round 1

Reviewer 1 Report

In the review of the manuscript titled: 5G based on MNOs for Railway Signalling Critical Data: Future Railway Mobile Communication System. The authors have provided a good description and the methodology is also fine. I would like to see this review publish but after some questions as follow;

 1.   Why did the authors just focus on mobile network operators (MNOs)?

2.    Why the authors select the length of the path up to 65 km long?

3.    Why does the availability increase in latency?

4.    Why does the signaling equipment requires a message transfer period lower than 1.8 seconds in a round-trip measurement?

5.    Why do the communication disconnections cause the duplication of the throughput when degraded radio level condition?

 

 

 

 

 

 

 

 

 

 

 

 

 

Author Response

Response to Reviewer 1 Comments

Point 0: In the review of the manuscript titled: 5G based on MNOs for Railway Signalling Critical Data: Future Railway Mobile Communication System. The authors have provided a good description and the methodology is also fine. I would like to see this review publish but after some questions as follow.

Response 0: Thank you for your questions. We have improved the document according to your comments.

Point 1: Why did the authors just focus on mobile network operators (MNOs)?

Response 1: MNOs are a feasible alternative to the typical solution based on the deployment of fibre optic cables for railway services. This is especially interesting for low-density railway traffic lines since fibre optics require civil work. Fibre optics and civil works’ costs can raise to 40% of the budget considering signalling and telecommunication systems. This means that, for some certain railway lines, updating the signalling systems is simply not economically viable, so the lines need to be closed. To sum up, MNOs can offer services to signalling systems in areas where it is not feasible to deploy a private network. More detail in this regard has been included in Section 1.

Point 2: Why the authors select the length of the path up to 65 km long?

Response 2: The objective of the field test is to analyse the performance of the proposed system in a realistic railway line. As interlockings are usually located at main stations, the idea was to test at least two main stations, which are Guardo and Mataporquera. Nevertheless, it is not usual but possible to find signalling equipment anywhere in the line. For that, the study shows the results along the railway line that connects these two main stations that are 65 km away including also two other secondary stations. This way, the study shows a good overview of the performance of the line. Section 3.3.3 includes further information concerning this question.

Point 3: Why does the availability increase in latency?

Response 3: The study shows empirical measurements, in which the study provides statistical latency over 99% of the time according to the radio parameters. In a process design, a set of thresholds are set with the aim of fulfilling the system requirements. Considering this, if we set the radio parameters with some extra margins, the system is more probable to fulfil the requirements more percent of the time. This study shows that, for example, where there is LTE RSRP at -108 dBm, the prototype provides a latency of 200 ms 99% of the time, whereas a signal of -109 increases the latency to 300 ms. Section 4.2.1 and Tab. 3 provide further detail in this regard.

Point 4: Why does the signaling equipment requires a message transfer period lower than 1.8 seconds in a round-trip measurement?

 

Response 4: The first step followed in this study was to seek the communication necessities of the signalling systems. The reason why signalling systems require to receive acknowledged information within 1.8 seconds is related to based on security and operative constraints, however, establishing this KPI is not part of this work. Nevertheless, further information has been included in Section 2.1.

In any case, from the study point of view, it is key to implement a communication system able to fulfil the application service requirements to not interfere in the operation of the service.

Point 5: Why do the communication disconnections cause the duplication of the throughput when degraded radio level condition?

Response 5: The throughput increases since the signalling system sends the same message more repeatedly during a shorter period until it receives the confirmation.

Reviewer 2 Report

Dear authors,

The research is interesting, but some revisions are necessary for the work to be published. Below are my comments:

1. In railway mobile communications systems, it is important to consider the Doppler effect, especially in 5G systems. However, I have not seen in this article an analysis of the Doppler effect.

2. Also, I believe that latency is high for certain situations presented for 5G communication systems; for example, you considered a 500 ms latency threshold for the service, but 500 ms is really appropriate for 5G systems? Or 200 ms is really appropriate for 5G systems?

Author Response

Point 0: Dear authors,

The research is interesting, but some revisions are necessary for the work to be published. Below are my comments:

Response 0: Thank you for your comments. We have further detailed the study according to them.

Point 1. In railway mobile communications systems, it is important to consider the Doppler effect, especially in 5G systems. However, I have not seen in this article an analysis of the Doppler effect.

Response 1: This study is focused on the communication between wayside signalling systems which are located in a fixed position. Although we agree that the Doppler effect is a key point to study for 5G systems, in this case, it is not required as the communication is fixed of all the elements in the communication system. For better understanding, a clarification has been included in Section 2.

Point 2. Also, I believe that latency is high for certain situations presented for 5G communication systems; for example, you considered a 500 ms latency threshold for the service, but 500 ms is really appropriate for 5G systems? Or 200 ms is really appropriate for 5G systems?

Response 2: A threshold is set to 500 ms since signalling systems required a latency lower than that threshold according to the study. Nevertheless, we agree that it is much more complete to include also the threshold for more restricted parameters. For that reason, Tab. 3 presents the statistical latency values according to the radio parameters. 

Reviewer 3 Report

Conclusively, I suggest accepting this article after a minor revision. This manuscript has only one problem. The authors introduced key performance indicators for the physical, network, and application layers. Nevertheless, I have read this article twice. I have not found any statement illustrating the authors’ key performance indicators. Thus, readers of this article may feel that the authors didn’t complete what they announced to do in this study.

Author Response

Point 0: Conclusively, I suggest accepting this article after a minor revision. This manuscript has only one problem. The authors introduced key performance indicators for the physical, network, and application layers. Nevertheless, I have read this article twice. I have not found any statement illustrating the authors’ key performance indicators. Thus, readers of this article may feel that the authors didn’t complete what they announced to do in this study.

Response 0: Thank you for your comment. The KPIs are the ones indicated in Tab. 2 and Tab. 3, whose captions have been rephrased. Nevertheless, for further clarification, Section 5 includes a statement regarding this issue.

Round 2

Reviewer 2 Report

Dear authors,

Thank you for responding to my comments.

I believe that the article is ready for publication.