scholarly journals Application of Deep Convolutional Neural Networks and Smartphone Sensors for Indoor Localization

2019 ◽  
Vol 9 (11) ◽  
pp. 2337 ◽  
Author(s):  
Imran Ashraf ◽  
Soojung Hur ◽  
Yongwan Park
Keyword(s):  
Magnetic Field ◽  
Nearest Neighbor ◽  
Indoor Localization ◽  
Scene Recognition ◽  
Localization Error ◽  
K Nearest Neighbor ◽  
Current Location ◽  
Smartphone Sensors ◽  
The Impact ◽  
Localization Systems

Indoor localization systems are susceptible to higher errors and do not meet the current standards of indoor localization. Moreover, the performance of such approaches is limited by device dependence. The use of Wi-Fi makes the localization process vulnerable to dynamic factors and energy hungry. A multi-sensor fusion based indoor localization approach is proposed to overcome these issues. The proposed approach predicts pedestrians’ current location with smartphone sensors data alone. The proposed approach aims at mitigating the impact of device dependency on the localization accuracy and lowering the localization error in the magnetic field based localization systems. We trained a deep learning based convolutional neural network to recognize the indoor scene which helps to lower the localization error. The recognized scene is used to identify a specific floor and narrow the search space. The database built of magnetic field patterns helps to lower the device dependence. A modified K nearest neighbor (mKNN) is presented to calculate the pedestrian’s current location. The data from pedestrian dead reckoning further refines this location and an extended Kalman filter is implemented to this end. The performance of the proposed approach is tested with experiments on Galaxy S8 and LG G6 smartphones. The experimental results demonstrate that the proposed approach can achieve an accuracy of 1.04 m at 50 percent, regardless of the smartphone used for localization. The proposed mKNN outperforms K nearest neighbor approach, and mean, variance, and maximum errors are lower than those of KNN. Moreover, the proposed approach does not use Wi-Fi for localization and is more energy efficient than those of Wi-Fi based approaches. Experiments reveal that localization without scene recognition leads to higher errors.

scholarly journals Indoor Localization Based on Fingerprint Clustering

2020 ◽  
Vol 5 (2) ◽  
pp. 40
Author(s):  
Shi Chen
Keyword(s):  
Nearest Neighbor ◽  
Indoor Localization ◽  
Rapid Development ◽  
Computational Cost ◽  
Gaussian Model ◽  
Location Based Services ◽  
Indoor Environments ◽  
Localization Error ◽  
Localization Algorithm ◽  
K Nearest Neighbor

With the rapid development of the huge promotion of the Internet and artificial intelligence, the demand for location-based services in indoor environments has grown rapidly. At present, for the localization of the indoor environment, researchers from all walks of life have proposed many indoor localization solutions based on different technologies. Fingerprint localization technology, as a commonly used indoor localization technology, has led to continuous research and improvement due to its low accuracy and complex calculations. An indoor localization system based on fingerprint clustering is proposed by this paper. The system includes offline phase and online phase. We collect the RSS signal in the offline phase. We preprocess it with the Gaussian model to build a fingerprint database, and then we use the K-Means++ algorithm to cluster the fingerprints and group the fingerprints with similar signal strengths into a clustering subset. In the online phase, we classify the measured received signal strength (RSS), and then use the weighted K-Nearest neighbor (WKNN) algorithm to calculate the localization error. The experimental results show that we can reduce the localization error and effectively reduce the computational cost of the localization algorithm in the online phase, and effectively improve the efficiency of real-time localization in the online phase.

Improved Indoor Geo-Localization System

2019 ◽  
pp. 183-206
Author(s):  
Abdaoui Noura ◽  
Ismahène Hadj Khalifa ◽  
Sami Faiz
Keyword(s):  
Sensor Networks ◽  
Recent Progress ◽  
Nearest Neighbor ◽  
Low Energy ◽  
Wireless Sensor ◽  
Detection Accuracy ◽  
K Nearest Neighbor ◽  
Smart Object ◽  
Low Energy Consumption ◽  
Localization Systems

In the concept of internet of things (IOT), physical position of smart object is very useful for relevant function over sensor networks. However, the invalid information of indoor geo-localization systems relative to these wireless sensor compromises the intelligence of IOT network. Therefore, this chapter produces the recent progress in the indoor geo-localization systems and the IOTs area. It defines the best indoor geo-localization technologies that meet their needs while respecting the constraints related to sensor networks. This framework combines between simplicity of Bluetooth low energy (BLE), popular wi-fi infrastructure, and the k-nearest neighbor (KNN) algorithm (in order to filter the initial fingerprint dataset). This new conception increases real-time detection accuracy and guarantees the low energy consumption.

scholarly journals The Impact of Emotional Music on Active ROI in Patients with Depression Based on Deep Learning: A Task-State fMRI Study

2019 ◽  
Vol 2019 ◽  
pp. 1-14
Author(s):  
Renzhou Gui ◽  
Tongjie Chen ◽  
Han Nie
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Correlation Matrix ◽  
Nearest Neighbor ◽  
Learning Model ◽  
Fmri Data ◽  
Support Vector ◽  
K Nearest Neighbor ◽  
The Impact ◽  
Deep Learning Model

With the continuous development of science, more and more research results have proved that machine learning is capable of diagnosing and studying the major depressive disorder (MDD) in the brain. We propose a deep learning network with multibranch and local residual feedback, for four different types of functional magnetic resonance imaging (fMRI) data produced by depressed patients and control people under the condition of listening to positive- and negative-emotions music. We use the large convolution kernel of the same size as the correlation matrix to match the features and obtain the results of feature matching of 264 regions of interest (ROIs). Firstly, four-dimensional fMRI data are used to generate the two-dimensional correlation matrix of one person’s brain based on ROIs and then processed by the threshold value which is selected according to the characteristics of complex network and small-world network. After that, the deep learning model in this paper is compared with support vector machine (SVM), logistic regression (LR), k-nearest neighbor (kNN), a common deep neural network (DNN), and a deep convolutional neural network (CNN) for classification. Finally, we further calculate the matched ROIs from the intermediate results of our deep learning model which can help related fields further explore the pathogeny of depression patients.

scholarly journals SDR Based Indoor Beacon Localization Using 3D Probabilistic Multipath Exploitation and Deep Learning

Electronics ◽  
2019 ◽  
Vol 8 (11) ◽  
pp. 1323 ◽  
Author(s):  
Donald L. Hall ◽  
Ram M. Narayanan ◽  
David M. Jenkins
Keyword(s):  
Deep Learning ◽  
Software Defined Radio ◽  
High Performance ◽  
Nearest Neighbor ◽  
Indoor Localization ◽  
Signal Propagation ◽  
Multipath Channel ◽  
K Nearest Neighbor ◽  
Positioning Systems ◽  
Channel Characteristics

Wireless indoor positioning systems (IPS) are ever-growing as traditional global positioning systems (GPS) are ineffective due to non-line-of-sight (NLoS) signal propagation. In this paper, we present a novel approach to learning three-dimensional (3D) multipath channel characteristics in a probabilistic manner for providing high performance indoor localization of wireless beacons. The proposed system employs a single triad dipole vector sensor (TDVS) for polarization diversity, a deep learning model deemed the denoising autoencoder to extract unique fingerprints from 3D multipath channel information, and a probabilistic k-nearest-neighbor (PkNN) to exploit the 3D multipath characteristics. The proposed system is the first to exploit 3D multipath channel characteristics for indoor wireless beacon localization via vector sensing methodologies, a software defined radio (SDR) platform, and multipath channel estimation.

Optimized Cost Model for k-NN Queries in R*-Trees over Random Distribution

2011 ◽  
Vol 403-408 ◽  
pp. 3315-3321
Author(s):  
Sirisala Nageswara Rao
Keyword(s):  
Nearest Neighbor ◽  
Cost Model ◽  
Neighbor Join ◽  
Multidimensional Data ◽  
Analysis Model ◽  
K Nearest Neighbor ◽  
Storage And Retrieval ◽  
Key Issues ◽  
Improved Performance ◽  
The Impact

Efficient storage and retrieval of multidimensional data in large volumes has become one of the key issues in the design and implementation of commercial and application software. The kind of queries posted on such data is also multifarious. Nearest neighbor queries are one such category and have more significance in GIS type of application. R-tree and its sequel are data partitioned hierarchical multidimensional indexing structures that help in this purpose. Today’s research has turned towards the development of powerful analytical method to predict the performance of such indexing structures such as for varies categories of queries such as range, nearest neighbor, join, etc .This paper focuses on performance of R*-tree for k nearest neighbor (kNN) queries. While general approaches are available in literature that works better for larger k over uniform data, few have explored the impact of small values of k. This paper proposes improved performance analysis model for kNN query for small k over random data. The results are tabulated and compared with existing models, the proposed model out performs the existing models in a significant way for small k

scholarly journals Random Forest-based Fingerprinting Technique for Device-free Indoor Localization System

2021 ◽  
pp. 79-96
Author(s):  
Dwi Suroso ◽  
Refa Rupaksi ◽  
Aditya Krisnawan ◽  
Nur Siddiq
Keyword(s):  
Random Forest ◽  
Nearest Neighbor ◽  
Indoor Localization ◽  
Random Forest Algorithm ◽  
K Nearest Neighbor ◽  
Received Signal Strength Indicator ◽  
Actual Measurement ◽  
Artificial Neural Network Ann ◽  
Device Free ◽  
Simple Parameter

The device-free indoor localization (DFIL) research is gaining attention due to the popularity of location-based service (LBS)-based advertisement. In DFIL, a user or an object does not need to bring any device to be localized. In this paper, we propose the Wi-Fi-based DFIL and the random forest algorithm for the fingerprint-based technique. The simple parameter commonly used in indoor localization is the Received Signal Strength Indicator (RSSI). We apply the fingerprint technique because of its reliability to handle the RSSI fluctuation and time-varying effect in a static indoor environment. We conducted an actual measurement campaign to observe the DFIL's implementation visibility. The DFIL system works by comparing the database fingerprint in an empty open office with the database in which a person is inside the measurement area without bringing any devices. Thus, we have the device-free RSSI database for fingerprint technique from both empty rooms and RSSI affected by a person inside the room. We validated the random forest algorithm results by comparing them with the k-nearest neighbor (kNN) and artificial neural network (ANN). The results show that our proposed system's accuracy is better than kNN and ANN with a mean error of 0.63 m than kNN with 0.80 m and ANN with 1.01 m. Meanwhile, the precision of the random forest is 0.63 m, whereas kNN and ANN are 0.67 m and 0.80 m, showing that the random forest performed better. We concluded that our simple DFIL system is visible to apply with acceptable accuracy performance.

scholarly journals Impact of a pulsed xenon disinfection system on hospital onset Clostridioides difficile infections in 48 hospitals over a 5-year period

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Sarah Simmons ◽  
Grady Wier ◽  
Antonio Pedraza ◽  
Mark Stibich
Keyword(s):  
Nearest Neighbor ◽  
Negative Binomial ◽  
Learning Algorithm ◽  
Compliance Rate ◽  
Negative Binomial Regression ◽  
Inverse Association ◽  
K Nearest Neighbor ◽  
Data Set ◽  
Clostridioides Difficile ◽  
The Impact

Abstract Background The role of the environment in hospital acquired infections is well established. We examined the impact on the infection rate for hospital onset Clostridioides difficile (HO-CDI) of an environmental hygiene intervention in 48 hospitals over a 5 year period using a pulsed xenon ultraviolet (PX-UV) disinfection system. Methods Utilization data was collected directly from the automated PX-UV system and uploaded in real time to a database. HO-CDI data was provided by each facility. Data was analyzed at the unit level to determine compliance to disinfection protocols. Final data set included 5 years of data aggregated to the facility level, resulting in a dataset of 48 hospitals and a date range of January 2015–December 2019. Negative binomial regression was used with an offset on patient days to convert infection count data and assess HO-CDI rates vs. intervention compliance rate, total successful disinfection cycles, and total rooms disinfected. The K-Nearest Neighbor (KNN) machine learning algorithm was used to compare intervention compliance and total intervention cycles to presence of infection. Results All regression models depict a statistically significant inverse association between the intervention and HO-CDI rates. The KNN model predicts the presence of infection (or whether an infection will be present or not) with greater than 98% accuracy when considering both intervention compliance and total intervention cycles. Conclusions The findings of this study indicate a strong inverse relationship between the utilization of the pulsed xenon intervention and HO-CDI rates.

scholarly journals K-Nearest Neighbor Classification for Detection of The Effect of Game Addiction on Cognitive Activity in The Late Adolescent Phase based on Brainwave Signals

2019 ◽  
Vol 1 (2) ◽  
pp. 46-62
Author(s):  
Ahmad Azhari ◽  
Ajie Kurnia Saputra Swara
Keyword(s):  
Cognitive Abilities ◽  
Nearest Neighbor ◽  
Cognitive Activity ◽  
World Health ◽  
Gaming Disorder ◽  
Negative Consequences ◽  
K Nearest Neighbor ◽  
Late Adolescent ◽  
Game Addiction ◽  
The Impact

World Health Organization (WHO) has determined that Gaming disorder is included in the International Classification of Diseases (ICD-11). The behavior of playing digital games included in the Gaming disorder category is characterized by impaired control of the game, increasing the priority given to the game more than other activities insofar as the game takes precedence over other daily interests and activities, and the continuation or improvement of the game despite negative consequences. The influence of video games on children's development has always been a polemic because in adolescence not only adopts cognitive abilities in learning activities, but also various strategies related to managing activities in learning, playing and socializing to improve cognitive abilities. Therefore, this research was conducted to analyze the cognitive activity of late teens in learning and playing games based on brainwave signals and to find out the impact of games on cognitive activity in adolescents. Prediction of the effect of the game on cognitive activity will be done by applying Fast Fourier Transform for feature extraction and K-Nearest Neighbor for classification. The results of the expert assessment showed the percentage of respondents with superior cognitive category but game addiction was 63.3% and respondents with cognitive categorization were average but were addicted by 36.6%. The percentage of accuracy produced by the system shows 80% in games and cognitive by using k values of 1, 6, and 7. The correlation test results show a percentage of 0.089, so it is concluded that there is no influence of the game on cognitive activity in late adolescents.

scholarly journals Machine Learning Based Indoor Localization using Wi-Fi Fingerprinting

2019 ◽  
Vol 8 (3) ◽  
pp. 502-506
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Nearest Neighbor ◽  
Indoor Localization ◽  
The Internet ◽  
Location Prediction ◽  
K Nearest Neighbor ◽  
Data Types ◽  
Specific Location ◽  
Web Access

The aim of indoor localization is to locate the objects inside a location wirelessly. This paper reports the models that predict the location along with floor and coordinates from the WAPs (Web Access Points) signal strengths of a user who connects to the internet at a specific location which had three locations. Starting with the cleaning of data, then assigning attributes into proper data types, making subset of dataset for each location, examining each column, and normalizing WAPs rows in order to build models. Different algorithms have been used to predict the location, floor, and coordinates of a logged in user. The models that have been used in this paper are k-Nearest Neighbor (k-NN) for location prediction, random forest for floor prediction and regression with k-NN for coordinate prediction.

Sign in / Sign up

Export Citation Format

Share Document