New Cluster Selection and Fine-grained Search for k-Means Clustering and Wi-Fi Fingerprinting

Torres-Sospedra, Joaquín; Quezada-Gaibor, Darwin; Mendoza-Silva, Germán Martín; Nurmi, Jari; Koucheryavy, Yevgeni; Huerta, Joaquín

doi:10.1109/icl-gnss49876.2020.9115419

Cited by 22 publications

(39 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the reduction in the computational load often comes with accuracy issues [ 31 ]. These advances of the area resulted in different approaches concerning traditional clustering such as K -means [ 32 ], hierarchical [ 33 , 34 ], and other novel clustering techniques [ 35 ]. Graph-based data structures have also been proposed recently to improve IPSs.…”

Section: Related Workmentioning

confidence: 99%

A Small World Graph Approach for an Efficient Indoor Positioning System

Lima

Guimarães²,

Santos

et al. 2021

Sensors

View full text Add to dashboard Cite

The main goal of an Indoor Positioning System (IPS) is to estimate the position of mobile devices in indoor environments. For this purpose, the primary source of information is the signal strength of packets received by a set of routers. The fingerprint technique is one of the most used techniques for IPSs. By using supervised machine learning techniques, it trains a model with the received signal intensity information so it can be used to estimate the positions of the devices later in an online phase. Although the k-Nearest Neighbors (kNN) is one of the most widely used classification methods due to its accuracy, it has no scalability since a sample that needs to be classified must be compared to all other samples in the training database. In this work, we use a novel hierarchical navigable small world graph technique to build a search structure so the location of a sample can be efficiently found, allowing the IPSs to be used in large-scale scenarios or run on devices with limited resources. To carry out our performance evaluation, we proposed a synthetic IPS dataset generator as well as implemented a complete real-world, large-scale IPS testbed. We compared the performance of our graph-based solution with other known kNN variants, such as Kd-Tree and Ball-Tree. Our results clearly show the performance gains of the proposed solution at 98% when compared to the classic kNN and at least 80% when compared to tree-based approaches.

show abstract

Section: Related Workmentioning

confidence: 99%

A Small World Graph Approach for an Efficient Indoor Positioning System

Lima

Guimarães²,

Santos

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…However, quantization only partially reduces the complexity of fingerprinting as the number of APs and reference fingerprints are not altered. Other relevant works that have focused on reducing the radio map include a focus on removing useless APs from the radio map [53], applying unsupervised learning to cluster/split the radio map [33,38,54,55], using radio-signal propagation knowledge to filter fingerprints out the radio-map [34,[56][57][58], combining clustering with RSSI-based rules [59], and, even, reducing on-the-fly the reference samples and APs of the radio map [60]. In this study, we focused on the nominal case where no other optimizations have been performed.…”

Section: Related Workmentioning

confidence: 99%

Analysis of Received Signal Strength Quantization in Fingerprinting Localization

Khandker

Torres-Sospedra

Ristaniemi

2020

Sensors

View full text Add to dashboard Cite

In recent times, Received Signal Strength (RSS)-based Wi-Fi fingerprinting localization has become one of the most promising techniques for indoor localization. The primary aim of RSS is to check the quality of the signal to determine the coverage and the quality of service. Therefore, fine-resolution RSS is needed, which is generally expressed by 1-dBm granularity. However, we found that, for fingerprinting localization, fine-granular RSS is unnecessary. A coarse-granular RSS can yield the same positioning accuracy. In this paper, we propose quantization for only the effective portion of the signal strength for fingerprinting localization. We found that, if a quantized RSS fingerprint can carry the major characteristics of a radio environment, it is sufficient for localization. Five publicly open fingerprinting databases with four different quantization strategies were used to evaluate the study. The proposed method can help to simplify the hardware configuration, enhance security, and save approximately 40–60% storage space and data traffic.

show abstract

“…Given the rapid growth of wearable and IoT devices that use positioning and localization services, it is essential to provide models that empower indoor positioning solutions in power-constrained devices. For instance, in [3], the authors provided three new variants for k-means clustering, which allowed a better distribution of Wi-Fi fingerprints among the clusters, reducing the computational load (by approx. 40%) in comparison with the original K-means clustering.…”

Section: Introductionmentioning

confidence: 99%

Lightweight Hybrid CNN-ELM Model for Multi-building and Multi-floor Classification

Quezada-Gaibor,

Torres-Sospedra,

Nurmi

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Machine learning models have become an essential tool in current indoor positioning solutions, given their high capabilities to extract meaningful information from the environment. Convolutional neural networks (CNNs) are one of the most used neural networks (NNs) due to that they are capable of learning complex patterns from the input data. Another model used in indoor positioning solutions is the Extreme Learning Machine (ELM), which provides an acceptable generalization performance as well as a fast speed of learning. In this paper, we offer a lightweight combination of CNN and ELM, which provides a quick and accurate classification of building and floor, suitable for power and resource-constrained devices. As a result, the proposed model is 58% faster than the benchmark, with a slight improvement in the classification accuracy (by less than 1%).

show abstract

New Cluster Selection and Fine-grained Search for k-Means Clustering and Wi-Fi Fingerprinting

Cited by 22 publications

References 27 publications

A Small World Graph Approach for an Efficient Indoor Positioning System

A Small World Graph Approach for an Efficient Indoor Positioning System

Analysis of Received Signal Strength Quantization in Fingerprinting Localization

Lightweight Hybrid CNN-ELM Model for Multi-building and Multi-floor Classification

Contact Info

Product

Resources

About