An Introduction to Indoor Localization Techniques. Case of Study: A Multi-Trilateration-Based Localization System with User–Environment Interaction Feature
Round 1
Reviewer 1 Report
Brief summary
In the manuscript the authors made a review of the indoor positioning techniques. They also shortly described several positioning systems with a special focus on the on the solutions suitable for assistance of impaired people. The second part of the manuscript is devoted to description of the RESIMA - an ultrasound-based positioning systems and its application for guidance of people in the room.
Broad comments
The abstract is rather a part of introduction and does not reflect the manuscript content.
The manuscript is quite short for a review type. Descriptions of the positioning techniques as well as the working systems are quite brief and do not show the state of art as the authors promise in the Abstract.
As far as the section 3, the RESIMA system description is insufficiently detailed. The authors did not motivate, why multiple MTA algorithm was chosen instead of one of the estimation methods. Estimation methods deliver x and y user coordinates using all the measured distances allowing also outliers filtering (robust estimation).
We also do not know what is the positioning accuracy, which is important considering the main area of the system application.
The Conclusion section is rather a kind of summary and do not contain any conclusions.
Specific comments and minor mistakes
'US' abbreviation in the article title can be confusing for some readers. I suggest using full word 'Ultrasound'
Line 87: Section title 'Materials and Methods' is not adequate to the section content. Although this a title is an element of the MDPI manuscript pattern it concerns rather research articles.
Line 180: Mistake in the section title (one extra 'and')
Figure 1: There is a mistake in the formula for [xj, yj]. The right side of the equation should be divided by 2.
Conclusion:
The manuscript should be thoroughly rebuilt. Some minor mistakes should be corrected.
Author Response
We would like to thank the Reviewer for the precious suggestions, which allow for improving the paper quality.
Brief summary
In the manuscript the authors made a review of the indoor positioning techniques. They also shortly described several positioning systems with a special focus on the on the solutions suitable for assistance of impaired people. The second part of the manuscript is devoted to description of the RESIMA - an ultrasound-based positioning systems and its application for guidance of people in the room.
Broad comments
- The abstract is rather a part of introduction and does not reflect the manuscript content.
The abstract has been restyled following the Reviewer suggestion.
- The manuscript is quite short for a review type. Descriptions of the positioning techniques as well as the working systems are quite brief and do not show the state of art as the authors promise in the Abstract.
We are sorry for this misunderstanding. The paper aims to give a brief introduction to the State of the Art in the field of indoor localization, with particular regards to sensors based approaches, and to illustrate main results achieved by the RESIMA project.
Following the Reviewer suggestion the following actions have been implemented:
- The paper title has been slightly modified to properly contextualize the paper aim.
- However, the state of the art has been improved, with particular regards to Section 2. “Technologies for Indoor Localization”.
- As far as the section 3, the RESIMA system description is insufficiently detailed. The authors did not motivate, why multiple MTA algorithm was chosen instead of one of the estimation methods. Estimation methods deliver x and y user coordinates using all the measured distances allowing also outliers filtering (robust estimation).
We also do not know what is the positioning accuracy, which is important considering the main area of the system application.
Section 3 has been improved by including all requested details on the RESIMA system, the MTA algorithm, the system accuracy. Moreover, more results have been presented (see Figures 5b and 6).
- The Conclusion section is rather a kind of summary and do not contain any conclusions.
The abstract has been restyled following the Reviewer suggestion.
Specific comments and minor mistakes
- 'US' abbreviation in the article title can be confusing for some readers. I suggest using full word 'Ultrasound'
This aspect has been fixed in the manuscript.
- Line 87: Section title 'Materials and Methods' is not adequate to the section content. Although this a title is an element of the MDPI manuscript pattern it concerns rather research articles.
This point has been fixed in the manuscript.
- Line 180: Mistake in the section title (one extra 'and')
This point has been fixed in the manuscript.
- Figure 1: There is a mistake in the formula for [xj, yj]. The right side of the equation should be divided by 2.
This point has been fixed in the manuscript.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The title of the paper is somehow misleading, one would expect a survey paper with a certain level of details about possible sensing solutions for indoor localization. However, the authors described the state of the art just briefly without a comprehensive review of previously published papers oriented on sensing solutions.
What do the authors mean by "low-signal Bluetooth beacons"? I guess it should be Bluetooth low-energy (BTLE).
It is not clear what the authors mean by "UEI" as this abbreviation is not explained in the manuscript.
Moreover, the case study is not very detailed. The authors provide only an overview of the previously described system and algorithms. The only results presented in the manuscript are shown in figure 5 where time delay as a function of discretization step is presented. The discussion of the results is poor. Authors should provide more results and perform tests in multiple scenarios (i.e. different number of users, different environment, etc. ). Moreover, it is not clear what is the desired time delay in the system. Only maximum values are presented in the figure, however, there is no information about mean value nor about distributions of delays achieved for 50 test positions. Were the results similar for all cases or can these maximum values be considered as outliers?
Author Response
We would like to thank the Reviewer for the precious suggestions, which allow for improving the paper quality.
- The title of the paper is somehow misleading, one would expect a survey paper with a certain level of details about possible sensing solutions for indoor localization. However, the authors described the state of the art just briefly without a comprehensive review of previously published papers oriented on sensing solutions.
We are sorry for this misunderstanding. The paper aims to give a brief introduction to the State of the Art in the field of indoor localization, with particular regards to sensors based approaches, and to illustrate main results achieved by the RESIMA project.
Following the Reviewer suggestion the following actions have been implemented:
- The paper title has been slightly modified to properly contextualize the paper aim.
- However, the state of the art has been improved, with particular regards to Section 2. “Technologies for Indoor Localization”.
- What do the authors mean by "low-signal Bluetooth beacons"? I guess it should be Bluetooth low-energy (BTLE).
Thanks a lot for this note. This has been fixed in the manuscript.
- It is not clear what the authors mean by "UEI" as this abbreviation is not explained in the manuscript.
Sorry for this and thanks a lot for the note. This has been fixed in the manuscript.
- Moreover, the case study is not very detailed. The authors provide only an overview of the previously described system and algorithms. The only results presented in the manuscript are shown in figure 5 where time delay as a function of discretization step is presented. The discussion of the results is poor. Authors should provide more results and perform tests in multiple scenarios (i.e. different number of users, different environment, etc. ).
Section 3 has been improved by including all requested details on the RESIMA system, the MTA algorithm, the system accuracy. Moreover, more results and considerations have been included (see Figures 5b and 6) and further comments, presenting the system behavior under different conditions.
- Moreover, it is not clear what is the desired time delay in the system. Only maximum values are presented in the figure, however, there is no information about mean value nor about distributions of delays achieved for 50 test positions. Were the results similar for all cases or can these maximum values be considered as outliers?
The standard deviation of the delay-time computed across the 50 different user positions demonstrated a negligible dependence on the user position. Moreover, the choice to show the maximum value (instead of the average) is further motivated by the need to investigate the system performance in the worst case, which could compromise the system efficiency in real applications.
Above statements have been now included in the main text (see section 3.2).
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors significantly improved the manuscript. Almost all my comments have been properly addressed.
There are still some details which need to be considered.
1. The abbreviation PoA (line 126) has not been explained.
2. Two versions of abbreviations are used: AoA and AOA, as well as LoS and LOS. It should be unified.
3. Figure 1: There is still a mistake in the formula for [xj, yj]. The right side of the equation should be DIVIDED by 2 or multiplied by 0.5. Please check it.
Author Response
The authors significantly improved the manuscript. Almost all my comments have been properly addressed.
The authors would like to thank the Reviewer for the precious suggestions which allowed to improve the paper quality.
There are still some details which need to be considered.
- The abbreviation PoA (line 126) has not been explained.
Sorry for this oversight, which has been now fixed in the manuscript.
- Two versions of abbreviations are used: AoA and AOA, as well as LoS and LOS. It should be unified.
Sorry for this oversight, which has been now fixed in the manuscript.
Author Response File: Author Response.pdf
Reviewer 2 Report
The submitted manuscript is definitely not a review, it does not provide a deep overview of the localization solutions.
"Introduction to advanced sensing solutions" would require more detailed information about the sensing itself (aimed either on ultrasound or any other technology), describing challenges and solutions that were proposed to overcome them. However, the paper does not provide any information about problems or solutions related to sensing.
Section 2 Technologies for Indoor Localization is trying to describe the extremely wide area of localization solutions. If authors deal with an ultrasound-based positioning system, they should focus much more on this area. Unfortunately, only a small fraction of the section is devoted to ultrasound technology. Moreover, in the second section, there is not single information related to the topic of the manuscript - advanced sensing solutions.
In the description of the localization system used in experiments, it is mentioned that a multi-trilateration algorithm (MTA) is implemented, however, a detailed description of the algorithm is not provided. Moreover, the references provided in the description are misleading. In [24] RFID based positioning system using PSO algorithm is described, which has nothing to do with trilateration. On line 279 authors provide 3 references, where the suitability of the solution was demonstrated, however, none of the references provide information related to implemented algorithm.
Line 278:
"The user position is hence estimated by computing the mean value of the un-filtered positions."
It is not quite clear what authors mean by "un-filtered positions", taking into account the previous sentence, it looks like they do count with "fake or not accurate distances" that might be detected by the algorithm.
The results presented in figures 5 and 6 provide information about maximum delay time, however, the authors conclude that the time delay is negligible. The question is how will the delay time change when the system needs to handle multiple parallel localization requests. In order to investigate system performance under the "worst-case scenario" or in a real application, different types of tests should be run. In a real-world implementation in the worst-case scenario the server (or PC) can be affected by other applications running in the background, a number of localization requests etc.
"Results presented in Figs. 5 and 6 need to be scaled up proportionally to the number of users."
Do authors assume that the computation time will increase linearly with the number of users?
From figure 6 it can be seen that number of nodes has quite an impact on the delay time. What is the impact of the number of nodes on localization accuracy?
What is the desired accuracy of the localization system and how is it affected by the tested parameters?
How are the results provided in the manuscript linked to the sensing solutions?
Author Response
We would like to thank the Reviewer for these further suggestions.
- The submitted manuscript is definitely not a review, it does not provide a deep overview of the localization solutions.
As already discussed, this paper does not aim to be a review paper. To make this aspect clearer, the paper title has been further changed.
- "Introduction to advanced sensing solutions" would require more detailed information about the sensing itself (aimed either on ultrasound or any other technology), describing challenges and solutions that were proposed to overcome them. However, the paper does not provide any information about problems or solutions related to sensing.
This paper deals with “indoor localization system”. Advanced solutions for indoor localization exploit both SENSORS and paradigms/METHODOLOGIES to process measurement signals.
The sensing solutions are quite traditional, while the main efforts of researcher are focused on the development of novel paradigms/methodologies for indoor localization.
The main aim of the introductive section is to highlight the main classes of methodologies, namely Proximity Detection, Triangulation, Fingerprinting, Dead-Reckoning, and Hybrid Localization. Advantages and drawbacks of each of this classes are briefly summarized.
We are sorry the Reviewer disagree about its content, but in our opinion this is the main message which should be carried out by this section.
However, to avoid mis-understanding we have further changed the paper title in “An Introduction to Indoor Localization Techniques. Case of study: a Multi-Trilateration-based localization system with User-Environment Interaction feature”.
- Section 2. Technologies for Indoor Localization is trying to describe the extremely wide area of localization solutions. If authors deal with an ultrasound-based positioning system, they should focus much more on this area. Unfortunately, only a small fraction of the section is devoted to ultrasound technology. Moreover, in the second section, there is not single information related to the topic of the manuscript - advanced sensing solutions.
In line with considerations reported in the previous point, this section aims to present main technologies for indoor localization, with particular regards to sensing technology.
In the revised version of the paper the following section has been restyled to address ultrasound sensing approaches.
Sound based indoor tracking systems adopt microphones, speakers, sound (audible) and ultrasonic sensors to generate signals, detect the intensity of sounds or as a distance sensor by estimating the travel time of the sound signal between the source and the sensor. To this aim different distance measurement techniques have been proposed like, for example, the TOA and the TDOA. Ultrasonic sensors are used as a distance sensor in the location detection systems discussed in [20, 21]. Authors in [21], proposed a high-accuracy ultrasonic indoor positioning system exploiting several wireless ultrasonic beacons in the indoor environment. Each beacon has well-known coordinates and operates both as receiver collecting the signals from the target node and transmitter by emitting ultrasonic signals. The distance between the beacon and the target node is calculated by measuring the time-of-flights (TOF) for the ultrasonic signals, and then the position of the target is evaluated by computing the measured distances. Experimental location error of 10.2 mm in the positioning for a moving robot is reported when the system is operated in the line-of-sight signal conditions.
Main drawbacks affecting ultrasound-based localization systems are due to the difficulty of signals to penetrate well through an obstruction, the need for several numbers of ultrasonic sensors to monitor large areas, multipath sensitivity, temperature sensitivity, Doppler effect.
Concerning references, we suggested different papers which provide a general overview of localization techniques including also ultrasound based multi-sensor approaches. Specific references on ultrasound based solutions are recent contributions to the state of the art. Moreover, a further recent publication on the topic has been now included.
The term “Advanced sensing solution” has been used to highlight several efforts and different approaches proposed in the literature to address the indoor localization task. Examples have been reported in the manuscript, such as the one presented in references [20-23, 50, 51, 55].
We are sorry the reviewer did not appreciate our effort. Specific suggestions on what is missing would have been really appreciated.
- In the description of the localization system used in experiments, it is mentioned that a multi-trilateration algorithm (MTA) is implemented, however, a detailed description of the algorithm is not provided.
The algorithm is described in section 3.1.
The MTA algorithm estimates the user coordinates considering and computing all possible combinations of three measured user-node (not alligned) distances [24]. The approach adopted sorts all the estimated user coordinates in order to filter out outlier estimations. The latter, which is the main advantage of MTA as respect to standard trilateration algorithm, allows for enhancing the robustness of the localization method against fake or not accurate distance measurements. The user position is hence estimated by computing the mean value of the un-filtered positions. Results discussed in [24, 25, 29] demonstrate the suitability of the proposed approach.
A flow diagram has not been included because it is already available in [50, 51, 55]. Reporting the same content already presented in other Journal papers in our opinion is not appropriate.
- Moreover, the references provided in the description are misleading. In [24] RFID based positioning system using PSO algorithm is described, which has nothing to do with trilateration. On line 279 authors provide 3 references, where the suitability of the solution was demonstrated, however, none of the references provide information related to implemented algorithm.
We are sorry for this, which came from the reorganization of the paper introduction.
Now this point has been fixed in the paper.
- Line 278:
"The user position is hence estimated by computing the mean value of the un-filtered positions." It is not quite clear what authors mean by "un-filtered positions", taking into account the previous sentence, it looks like they do count with "fake or not accurate distances" that might be detected by the algorithm.
Un-filtered means before eliminating outliers. Now this point has been better clarified in the manuscript.
- The results presented in figures 5 and 6 provide information about maximum delay time, however, the authors conclude that the time delay is negligible. The question is how will the delay time change when the system needs to handle multiple parallel localization requests. In order to investigate system performance under the "worst-case scenario" or in a real application, different types of tests should be run. In a real-world implementation in the worst-case scenario the server (or PC) can be affected by other applications running in the background, a number of localization requests etc.
This aspect has been already addressed in the paper by considerations related to the use with multiple-users and observed times are negligible considering the addressed application. Future efforts will be dedicated to investigate also the effect of parallel running tasks. This statement has been now included in the conclusions.
- "Results presented in Figs. 5 and 6 need to be scaled up proportionally to the number of users." Do authors assume that the computation time will increase linearly with the number of users?
This is our idea, since the multi-user management will be handled in series and not in parallel.
- From figure 6 it can be seen that number of nodes has quite an impact on the delay time. What is the impact of the number of nodes on localization accuracy?
It should be observed that in the trilateration approach, and so also in the MTA algorithm, increasing the number of nodes could improve, to a certain exent, the accuracy of the localization system. However, this assessment must be performed case by case, by taking into acount the characteristic of the considered environment. This statement has been now included in the main text.
- What is the desired accuracy of the localization system and how is it affected by the tested parameters?
As stated in Section 3.1, “A statistical analysis of residuals between expected and measured user coordinates allow to estimate a system uncertainty less than 5.0 cm for both the x and the y axis.”.
This accuracy value is compliant with application in the field of Assistive Technology, which usually requires accuracy in the range of ten centimeters. This statement has been now included in the main text.
- How are the results provided in the manuscript linked to the sensing solutions?
Results shown in the paper are related to the ultrasound sensing system adopted to implement the localization task of the RESIMA system.
Author Response File: Author Response.pdf
