# An Empirical Study of the Transmission Power Setting for Bluetooth-Based Indoor Localization Mechanisms

^{1}

^{2}

^{*}

^{†}

## Abstract

**:**

## 1. Introduction

## 2. Related Work

## 3. BLE4.0 Indoor Localization: Set-Up, Tools and Algorithms

#### 3.1. Experimental Indoor Set-Up

#### 3.2. Bluetooth Receiver’s Characteristics

- We fixed the transmission power of all BLE4.0 beacons to the same level.
- We placed the mobile device at the centre of each one of the 151 m${}^{2}$ and measured the RSSI of each one of the five BLE4.0 beacons for a time period of one minute.
- We evaluated the mean and standard deviation of the RSSI for each one of the five BLE4.0 beacons.

#### 3.3. Bluetooth Signal Attenuation

#### Case 1: Sector 8

#### Case 2: Sector 4

#### Case 3: Sector 15

#### 3.4. Supervised Learning Algorithms

**k-NN**: Given a test instance, this algorithm selects the k nearest neighbours, based on a pre-defined distance metric of the training set. In our case, we use the Euclidean distance since our predictor variables (features) share the same type, i.e., the RSSI values, properly fitting the indoor localization problem [22]. Although k-NN uses the most common neighbour of the k located categories (that is the mode of the category) to classify a given test instance, some variations are used (e.g., weighted distances) to avoid removing relevant information. In this paper, we have set the hyperparameter to k = 5 as the best solution, based on some of our preliminary numerical analysis. We use both mentioned versions of the algorithm: the weighted distance (WD) and mode (MD).**SVM**: Given the training data, a hyperplane is defined to optimally discriminate between different categories. If linear classifier are used, SVM constructs a line that performs an optimal discrimination. For the non-linear classifier, kernel functions are used, which maximize the margin between categories. In this paper, we have explored the use of linear classifier and Polynomial kernel with two different grades, namely, 2 and 3. Finally, we present only the best results, which were obtained with a Polynomial kernel with a quadratic function [22].

## 4. On the Adequacy of the Bluetooth-Based Localization Platform

#### 4.1. Relevance of BLE4.0 Beacons

**ExtraTrees**stands for Extremely Randomized Trees, which is an ensemble method that builds multiple models (random trees) for each sample of the training data. Then, all of the predictions are averaged. Default sklearn python library hyperparameters were used.**Gradient Boosting Algorithm**is also an ensemble method using decision trees as base models and weighted voting selection method. Furthermore, it makes a prior model every time it is executed. Default sklearn python library hyperparameters were used.

#### 4.2. Baseline Evaluation

## 5. Asymmetric Transmission Power

#### 5.1. Fingerprint as a Function of the Transmission Power

#### 5.2. On Deriving the Best Asymmetric Transmission Power Setting

#### Case 1: Asymmetric Transmission Power for k-NN

#### Case 2: Asymmetric Transmission Power for SVM

#### 5.3. Asymmetric Transmission Power Setting

#### 5.4. On the Relevance of the Individual RSSI Values

#### 5.5. On Mitigating the Multipath Fading Impairment

- Although it is important to classify the sectors with a distinctive RSSI, the percentage of settings obtained is not a considerable matching of the combinations between the two classification algorithms.
- The RSSI value of a given BLE4.0 beacon proves to be a useful, but not definitely, the main source of information on setting the best transmission power setting.
- An asymmetric transmission power setting may prove useful on mitigating the information to be provided to the classification algorithms due to the multipath fading effect.

## 6. Conclusions

## Acknowledgments

## Author Contributions

## Conflicts of Interest

## Abbreviations

RSSI | Received Signal Strength Indication |

BLE4.0 | Bluetooth Low Energy 4.0 |

k-NN | k-Nearest Neighbour |

SVM | Support Vector Machine |

AP | Access Point |

$Tx$ | Transmission Power |

dB | Decibel |

dBm | Decibel-milliwatts |

MD | Mode |

WD | Weighted Distance |

## References

- Shuo, S.; Hao, S.; Yang, S. Design of an experimental indoor position system based on RSSI. In Proceedings of the 2nd International Conference on Information Science and Engineering, Hangzhou, China, 4–6 December 2010; pp. 1989–1992. [Google Scholar]
- Feldmann, S.; Kyamakya, K.; Zapater, A.; Lue, Z. An indoor bluetooth-based positioning system: Concept, implementation and experimental evaluation. In Proceedings of the International Conference on Wireless Networks, Las Vegas, NV, USA, 23–26 June 2003; pp. 109–113. [Google Scholar]
- Shukri, S.; Kamarudin, L.; Cheik, G.C.; Gunasagaran, R.; Zakaria, A.; Kamarudin, K.; Zakaria, S.S.; Harun, A.; Azemi, S. Analysis of RSSI-based DFL for human detection in indoor environment using IRIS mote. In Proceedings of the 3rd IEEE International Conference on Electronic Design (ICED), Phuket, Thailand, 11–12 August 2016; pp. 216–221. [Google Scholar]
- Rappaport, T. Wireless Communications: Principles and Practice, 2nd ed.; Prentice Hall PTR: Upper Saddle River, NJ, USA, 2001. [Google Scholar]
- Martínez-Gómez, J.; del Horno, M.M.; Castillo-Cara, M.; Luján, V.M.B.; Barbosa, L.O.; García-Varea, I. Spatial statistical analysis for the design of indoor particle-filter-based localization mechanisms. Int. J. Distrib. Sens. Netw.
**2016**, 12. [Google Scholar] [CrossRef] - Onishi, K. Indoor position detection using BLE signals based on voronoi diagram. In Proceedings of the International Conference on Intelligent Software Methodologies, Tools, and Techniques, Langkawi, Malaysia, 22–24 September 2014; pp. 18–29. [Google Scholar]
- Palumbo, F.; Barsocchi, P.; Chessa, S.; Augusto, J.C. A stigmergic approach to indoor localization using bluetooth low energy beacons. In Proceedings of the 12th IEEE International Conference on Advanced Video and Signal Based Surveillance, Karlsruhe, Germany, 25–28 August 2015; pp. 1–6. [Google Scholar]
- Wang, Q.; Feng, Y.; Zhang, X.; Su, Y.; Lu, X. IWKNN: An effective bluetooth positioning method based on isomap and WKNN. Mob. Inf. Syst.
**2016**, 2016, 8765874:1–8765874:11. [Google Scholar] [CrossRef] - Faragher, R.; Harle, R. An analysis of the accuracy of bluetooth low energy for indoor positioning applications. In Proceedings of the 27th International Technical Meeting of The Satellite Division of the Institute of Navigation, Tampa, FL, USA, 8–12 September 2014; Volume 812, pp. 201–210. [Google Scholar]
- Peng, Y.; Fan, W.; Dong, X.; Zhang, X. An Iterative Weighted KNN (IW-KNN) Based Indoor Localization Method in Bluetooth Low Energy (BLE) Environment. In Proceedings of the 2016 International IEEE ConferencesUbiquitous Intelligence & Computing, Advanced and Trusted Computing, Scalable Computing and Communications, Cloud and Big Data Computing, Internet of People, and Smart World Congress, Toulouse, France, 18–21 July 2016; pp. 794–800. [Google Scholar]
- Zhang, L.; Liu, X.; Song, J.; Gurrin, C.; Zhu, Z. A comprehensive study of bluetooth fingerprinting-based algorithms for localization. In Proceedings of the 27th IEEE International Conference on Advanced Information Networking and Applications Workshops (WAINA), Barcelona, Spain, 25–28 March 2013; pp. 300–305. [Google Scholar]
- Leitinger, E.; Meissner, P.; Rüdisser, C.; Dumphart, G.; Witrisal, K. Evaluation of position-related information in multipath components for indoor positioning. IEEE J. Sel. Areas Commun.
**2015**, 33, 2313–2328. [Google Scholar] [CrossRef] - Wang, Q.; Guo, Y.; Yang, L.; Tian, M. An indoor positioning system based on ibeacon. In Transactions on Edutainment XIII; Pan, Z., Cheok, A.D., Müller, W., Zhang, M., Eds.; Springer: Berlin/Heidelberg, Germany, 2017; pp. 262–272. [Google Scholar]
- Kriz, P.; Maly, F.; Kozel, T. Improving indoor localization using bluetooth low energy beacons. Mob. Inf. Syst.
**2016**, 2016, 2083094:1–2083094:11. [Google Scholar] [CrossRef] - Faragher, R.; Harle, R. Location fingerprinting with bluetooth low energy beacons. IEEE J. Sel. Areas Commun.
**2015**, 33, 2418–2428. [Google Scholar] [CrossRef] - Paek, J.; Ko, J.; Shin, H. A measurement study of ble ibeacon and geometric adjustment scheme for indoor location-based mobile applications. Mob. Inf. Syst.
**2016**, 2016, 1–13. [Google Scholar] [CrossRef] - Perera, C.; Aghaee, S.; Faragher, R.; Harle, R.; Blackwell, A. A contextual investigation of location in the home using bluetooth low energy beacons. arXiv, 2017; arXiv:cs.HC/1703.04150. [Google Scholar]
- Pei, L.; Zhang, M.; Zou, D.; Chen, R.; Chen, Y. A survey of crowd sensing opportunistic signals for indoor localization. Mob. Inf. Syst.
**2016**, 2016, 1–16. [Google Scholar] [CrossRef] - Jaalee. Beacon IB0004-N Plus. Available online: https://www.jaalee.com/ (accessed on 6 March 2017).
- Anagnostopoulos, G.G.; Deriaz, M.; Konstantas, D. Online self-calibration of the propagation model for indoor positioning ranging methods. In Proceedings of the International Conference on Indoor Positioning and Indoor Navigation (IPIN), Alcala de Henares, Spain, 4–7 October 2016; pp. 1–6. [Google Scholar]
- Trendnet. Micro Bluetooth USB Adapter. Available online: https://www.trendnet.com/products/USB-adapters/TBW-107UB/ ( accessed on 6 March 2017).
- Brownlee, J. Spot-check classification algorithms. In Machine Learning Mastery with Python; Machine Learning Mastery Pty Ltd.: Vermont Victoria, Australia, 2016; pp. 100–120. [Google Scholar]
- Breiman, L. Statistical modeling: The two cultures (with comments and a rejoinder by the author). Stat. Sci.
**2001**, 16, 199–231. [Google Scholar] [CrossRef] - Brownlee, J. Feature selection. In Machine Learning Mastery with Python; Machine Learning Mastery Pty Ltd.: Vermont Victoria, Australia, 2016; pp. 52–56. [Google Scholar]
- Rivas, T.; Paz, M.; Martín, J.; Matías, J.M.; García, J.; Taboada, J. Explaining and predicting workplace accidents using data-mining techniques. Reliab. Eng. Syst. Saf.
**2011**, 96, 739–747. [Google Scholar] [CrossRef] - Geurts, P.; Ernst, D.; Wehenkel, L. Extremely randomized trees. Mach. Learn.
**2006**, 63, 3–42. [Google Scholar] [CrossRef] - Brownlee, J. Ensemble methods. In Machine Learning Mastery with Python; Machine Learning Mastery Pty Ltd.: Vermont Victoria, Australia, 2016; pp. 91–95. [Google Scholar]
- Pedregosa, F.; Varoquaux, G.; Gramfort, A.; Michel, V.; Thirion, B.; Grisel, O.; Blondel, M.; Prettenhofer, P.; Weiss, R.; Dubourg, V.; et al. Scikit-learn: Machine learning in python. J. Mach. Learn. Res.
**2011**, 12, 2825–2830. [Google Scholar] - Li, J.; Cheng, K.; Wang, S.; Morstatter, F.; Trevino, R.P.; Tang, J.; Liu, H. Feature selection: A data perspective. arXiv, 2016; arXiv:1601.07996. [Google Scholar]
- Rahim, A.; Dimitrova, R.; Finger, A. Techniques for Bluetooth Performance Improvement. Available online: https://pdfs.semanticscholar.org/3205/2262d3c152a3cc947acbc7b325debe9cbeef.pdf (accessed on 7 June 2017).
- Chen, L.; Li, B.; Zhao, K.; Rizos, C.; Zheng, Z. An improved algorithm to generate a Wi-Fi fingerprint database for indoor positioning. Sensors
**2013**, 13, 11085–11096. [Google Scholar] [CrossRef] [PubMed]

**Figure 3.**Pictures from each one of the four corners of the lab. (

**a**) from Be07; (

**b**) from Be08; (

**c**) from Be10; (

**d**) from Be11.

**Figure 4.**RSSI (dBm) for BLE4.0 Antenna and smartphone with transmission power $Tx=0\mathrm{x}04$ for each sector (1.15) of our environment. (

**a**) for Be07; (

**b**) for Be09.

**Figure 5.**Sector 8: Comparison of the RSSI from different BLE4.0 beacons for $Tx=0\mathrm{x}04$. (

**a**) for Be07; (

**b**) for Be10.

**Figure 6.**Sector 4: Comparison of the RSSI from different BLE4.0 beacons for $Tx=0\mathrm{x}04$. (

**a**) for Be07; (

**b**) for Be10.

**Figure 7.**Sector 15: Comparison of the RSSI from different BLE4.0 beacons for $Tx=0\mathrm{x}04$. (

**a**) for Be07; (

**b**) for Be10.

**Figure 8.**Relevance score of each BLE4.0 beacon for ExtraTrees algorithm for different transmission power ($Tx$) levels. (

**a**) $Tx=0\mathrm{x}01$; (

**b**) $Tx=0\mathrm{x}02$; (

**c**) $Tx=0\mathrm{x}03$; (

**d**) $Tx=0\mathrm{x}04$; (

**e**) $Tx=0\mathrm{x}05$; (

**f**) $Tx=0\mathrm{x}06$.

**Figure 9.**Relevance score of each BLE4.0 beacon for Gradient Boosting Classifier algorithm for different transmission power ($Tx$) levels. (

**a**) $Tx=0\mathrm{x}01$; (

**b**) $Tx=0\mathrm{x}02$; (

**c**) $Tx=0\mathrm{x}03$; (

**d**) $Tx=0\mathrm{x}04$; (

**e**) $Tx=0\mathrm{x}05$; (

**f**) $Tx=0\mathrm{x}06$.

**Figure 10.**RSSI values for the best (

**top**) and worst (

**bottom**) transmission power ($Tx$) level for BLE4.0 beacons ’Be07’, ’Be09’ and ’Be10’ throughout the area captured by the BLE4.0 antenna. (

**a**) Be07 with $Tx=0\mathrm{x}03$; (

**b**) Be09 with $Tx=0\mathrm{x}03$; (

**c**) Be10 with $Tx=0\mathrm{x}03$; (

**d**) Be07 with $Tx=0\mathrm{x}05$; (

**e**) Be09 with $Tx=0\mathrm{x}05$; (

**f**) Be10 with $Tx=0\mathrm{x}05$.

**Figure 11.**RSSI values for different transmission power levels (Tx) for BLE4.0 beacons ’Be11’, ’Be07’ and ’Be08’. (

**a**) ’Be11’ with $Tx=0\mathrm{x}03$; (

**b**) ’Be07’ with $Tx=0\mathrm{x}01$; (

**c**) ’Be08’ with $Tx=0\mathrm{x}05$; (

**d**) ’Be11’ with $Tx=0\mathrm{x}04$; (

**e**) ’Be07’ with $Tx=0\mathrm{x}04$; (

**f**) ’Be08’ with $Tx=0\mathrm{x}01$.

**Figure 12.**Positioning error for k-NN (with k = 5) using (

**a**) weighted distance; (

**b**) mode. In both plots, the three best and the three worst combined transmission power for each BLE4.0 beacon are shown.

**Figure 13.**RSSI values using the most relevant transmission power ($Tx$) level setting for each BLE4.0 beacon: [4,1,2,1,1]. (

**a**) ’Be07’ with $Tx=0\mathrm{x}04$; (

**b**) ’Be08’ with $Tx=0\mathrm{x}01$; (

**c**) ’Be09’ with $Tx=0\mathrm{x}02$; (

**d**) ’Be10’ with $Tx=0\mathrm{x}01$; (

**e**) ’Be11’ with $Tx=0\mathrm{x}01$.

**Figure 14.**Positioning error for SVM (with a quadratic polynomial kernel function). In both plots, the three best and the three worst combined transmission power for each BLE4.0 beacon are shown.

**Table 1.**Global accuracy for k-NN using mode (with k = 5) and SVM (with a quadratic polynomial kernel function) algorithms for transmission power $Tx=0\mathrm{x}04$. Best results are shown in bold.

Algorithm | Smartphone | BLE4.0 Antenna |
---|---|---|

k-NN | 21% | 64.6% |

SVM | 22.4% | 60.6% |

**Table 2.**Sample sizes of the RSSI captured using the BLE4.0 at various transmission power ($Tx$) levels.

Transmission Power | Sample Size per BLE4.0 Beacon |
---|---|

$Tx=0\mathrm{x}01$ | 5003 |

$Tx=0\mathrm{x}02$ | 5246 |

$Tx=0\mathrm{x}03$ | 4844 |

$Tx=0\mathrm{x}04$ | 5134 |

$Tx=0\mathrm{x}05$ | 4697 |

$Tx=0\mathrm{x}06$ | 4198 |

**Table 3.**Global accuracy using BLE4.0 antenna for k-NN (with k = 5) using mode and SVM (with a quadratic polynomial kernel function) algorithms for different transmission power ($Tx$) levels. Best results are shown in bold.

Transmission Power | Algorithm | |
---|---|---|

k-NN | SVM | |

$Tx=0\mathrm{x}01$ | 62.3% | 57.7% |

$Tx=0\mathrm{x}02$ | 61.5% | 52.6% |

$Tx=0\mathrm{x}03$ | 65.0% | 58.0% |

$Tx=0\mathrm{x}04$ | 64.6% | 60.6% |

$Tx=0\mathrm{x}05$ | 56.6% | 50.4% |

$Tx=0\mathrm{x}06$ | 63.8% | 61.7% |

**Table 4.**Local accuracy in each sector of our experimental area with the most relevant transmission power level for k-NN using mode (with k = 5). The centre shows the accuracy (in %) of each sector. Corners and middle-left hand are the position of BLE4.0 beacons with BeXY name. The most relevant transmission power level was [4,1,2,1,1].

81.31 | 71.43 | 84.62 | ||

30.10 | 70.69 | 84.11 | ||

Be09 | 100.00 | 18.10 | 52.88 | |

28.95 | 71.17 | 53.19 | ||

72.10 | 86.92 | 77.59 | ||

**Table 5.**Local accuracy in each sector of our experimental area with the most relevant transmission power level for SVM (with a quadratic polynomial kernel function). The centre shows the accuracy (in %) of each sector. Corners and middle-left hand are the position of BLE4.0 beacons with BeXY name. The most relevant transmission power level was [4,1,2,1,1].

85.00 | 80.70 | 99.07 | ||

11.50 | 69.17 | 76.64 | ||

Be09 | 70.43 | 19.83 | 51.97 | |

20.18 | 68.38 | 27.66 | ||

52.33 | 87.85 | 91.38 | ||

**Table 6.**Cumulative positioning error with different transmission power ($Tx$) level settings for k-NN (with k = 5) using weighted distance (WD) and mode (MD); and SVM (with a quadratic polynomial kernel function). Best results are shown in bold.

Algorithm - $\mathit{Tx}$ Setting | Cumulative Positioning Error | ||||
---|---|---|---|---|---|

0 m | ≤1 m | ≤2 m | ≤3 m | ≤4 m | |

k-NN (WD) - [3,3,3,3,3] | 33.27% | 57.22% | 77.53% | 88.15% | 95.12% |

k-NN (WD) - [4,1,2,1,1] | 36.15% | 64.47% | 82.92% | 94.03% | 98.70% |

k-NN (MD) - [3,3,3,3,3] | 65.00% | 65.00% | 74.57% | 81.36% | 89.26% |

k-NN (MD) - [4,1,2,1,1] | 77.89% | 77.89% | 86.74% | 92.52% | 99.68% |

SVM - [6,6,6,6,6] | 61.70% | 61.70% | 72.81% | 77.22% | 88.40% |

SVM - [4,1,2,1,1] | 75.57% | 75.57% | 84.92% | 90.33% | 99.12% |

**Table 7.**Mean error for k-NN (with k = 5) using weighted distance (WD) and mode (MD); and SVM (with a quadratic polynomial kernel function) with the same and the most relevant transmission power level ($Tx$). Best results are shown in bold.

Algorithm - $\mathit{Tx}$ Setting | Mean Error (m) |
---|---|

k-NN (WD) - [3,3,3,3,3] | 1.16 |

k-NN (WD) - [4,1,2,1,1] | 0.57 |

k-NN (MD) - [3,3,3,3,3] | 1.11 |

k-NN (MD) - [4,1,2,1,1] | 0.51 |

SVM - [6,6,6,6,6] | 1.17 |

SVM - [4,1,2,1,1] | 0.58 |

**Table 8.**Accuracy results for the k-NN using mode (with k = 5) (right) and SVM localization (with a quadratic polynomial kernel function) (left) algorithms. Worst and best settings using different asymmetric transmission power settings, and the best symmetric transmission power level settings (shown in italic font). Best results are shown in bold.

SVM | k-NN | ||
---|---|---|---|

$\mathit{Tx}$ Setting | Accuracy | $\mathit{Tx}$ Setting | Accuracy |

[1-2-4-5-3] | 35.70% | [1-5-4-5-3] | 43.23% |

[1-5-4-5-3] | 35.91% | [1-2-4-5-3] | 44.48% |

[1-5-5-5-3] | 36.28% | [1-5-4-5-5] | 44.53% |

[1-5-5-2-3] | 36.69% | [1-3-4-5-3] | 44.54% |

[1-2-4-2-3] | 36.73% | [1-3-4-2-3] | 44.58% |

[2-2-2-2-2] | 52.68% | [5-5-5-5-5] | 56.70% |

[5-5-5-5-5] | 50.41% | [2-2-2-2-2] | 61.50% |

[1-1-1-1-1] | 57.74% | [1-1-1-1-1] | 62.10% |

[3-3-3-3-3] | 57.90% | [6-6-6-6-6] | 63.80% |

[4-4-4-4-4] | 60.70% | [4-4-4-4-4] | 64.70% |

[6-6-6-6-6] | 61.70% | [3-3-3-3-3] | 65.00% |

[4-1-2-3-2] | 73.86% | [3-1-2-1-1] | 75.96% |

[4-1-4-1-1] | 74.29% | [4-1-2-3-4] | 76.87% |

[4-1-2-6-1] | 75.36% | [4-1-2-6-1] | 77.23% |

[4-1-2-3-1] | 75.36% | [4-1-2-1-4] | 77.45% |

[4-1-2-1-1] | 75.57% | [4-1-2-1-1] | 77.89% |

**Table 9.**Ranking of the transmission power values used by each BLE4.0 beacon for k-NN using mode (with k = 5) results.

BLE4.0 Beacon | Ranking |
---|---|

Be07 | 3.7% |

Be08 | 0.9% |

Be09 | 0.5% |

Be10 | 5.0% |

Be11 | 2.5% |

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Castillo-Cara, M.; Lovón-Melgarejo, J.; Bravo-Rocca, G.; Orozco-Barbosa, L.; García-Varea, I.
An Empirical Study of the Transmission Power Setting for Bluetooth-Based Indoor Localization Mechanisms. *Sensors* **2017**, *17*, 1318.
https://doi.org/10.3390/s17061318

**AMA Style**

Castillo-Cara M, Lovón-Melgarejo J, Bravo-Rocca G, Orozco-Barbosa L, García-Varea I.
An Empirical Study of the Transmission Power Setting for Bluetooth-Based Indoor Localization Mechanisms. *Sensors*. 2017; 17(6):1318.
https://doi.org/10.3390/s17061318

**Chicago/Turabian Style**

Castillo-Cara, Manuel, Jesús Lovón-Melgarejo, Gusseppe Bravo-Rocca, Luis Orozco-Barbosa, and Ismael García-Varea.
2017. "An Empirical Study of the Transmission Power Setting for Bluetooth-Based Indoor Localization Mechanisms" *Sensors* 17, no. 6: 1318.
https://doi.org/10.3390/s17061318