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

This paper focuses on the review of the recent advancement and development of indoor localization focusing on device-free indoor localization (DFIL). DFIL offers more flexibility for the user or object point-of-view. The basic idea is to free the object from the device attached for localization process. The most popular technologies for DFIL are based on wireless technology. However, if we discuss the low-cost system, we have to discuss the fusion or hybrid of many sensors output data to increase accuracy. Some of the accessible technology used for these low-cost systems are wireless-based, i.e., Wi-Fi, ZigBee, radio frequency identification (RFID) and an optical-based, visible light communication (VLC). Wireless-based probably the most used in the low-cost and simple configuration of an indoor localization since the parameter used can be directly obtained from its device. The most used parameter in the wireless-based system is received signal strength indicator (RSSI), which merely the derivation of power by the distance. By this parameter, the technique such as trilateration, min-max, and RSSI database can be used as the location. In the DFIL, RSSI can be applied as the Wi-Fi -based radio map from two or more links and applying an algorithm such as the Bayes Inference, resulting in the accuracy of object tracking up to 86%. On the other hand, ZigBee applies to mesh network and need more nodes to gather more information for higher accuracy results. The RFID has advantages in no-contact, and a good accuracy even in non-line-of-sight (LoS) condition. Optical-based VLC can provide high accuracy but dependent on the light noise and shadow effect in the object. From these benefits and disadvantages of some low-cost systems, we can make the best of them by fusing the sensor or developing the hybrid system.

LoRaWAN technology’s reliability is challenged by weather parameters, which can influence the communication channel design, especially when dealing with outdoor devices. We propose to analyze this effect by evaluating the relationship between the received signal strength indicator (RSSI) and different weather parameters, as well as its temporal changes. A rigorous statistical analysis of the RSSI sequences is conducted to assess if they could be represented by a specific statistical model. For this purpose, several models are investigated. The Artificial Intelligence (AI) algorithms cover machine learning (ML) and deep learning methods are appealing when dealing with time series forecasting. Nevertheless, the classical autoregressive integrated moving average (ARIMA) model can be an attractive alternative due to its simplicity. Therefore, this work proposes a comparative study of ARIMA, AI, and hybrid approaches to forecast the RSSI using weather parameters as regressors. The considered AI algorithms are the artificial neural network (ANN), support vector machine (SVM), random forest (RF), and Long Short-Term Memory (LSTM). Also, hybrid models are constructed, coupling the ARIMA with them. The models are evaluated in time series of RSSI, measured by eight different LoRaWAN transmitter nodes and considering the temperature, pressure, relative humidity, and rain as weather parameters. Our analysis reveals that temperature is the dominant factor among weather parameters, and negatively affects RSSI. The ARIMA model that uses only the temperature as a regressor provides consistently better fits than the ARIMA without regressors. Moreover, coupling the ARIMA with the temperature as a regressor and the ANN (ARIMA-ANN) is the best option among the pure AI and hybrid approaches. However, it provided accuracy measures very close to those obtained from the ARIMA model fitted in the first stage, with similar performance. Therefore, the ARIMA model considering the temperature is the most competitive alternative when analyzing RSSI measurements, with the advantage of being the most straightforward method. These results suggest that the RSSI from the analyzed LoRaWAN receiver nodes may not present nonlinear patterns and, considering several weather parameters, they are affected mainly by the outdoor temperature.

The localization of wireless sensor networks (WSN) is an important field of research study, that is gaining considerable attention. This growth in wireless sensor network installations is expected to expand more in the future. Indoor localization based on the Received Signal Strength Indicators (RSSI) is a well-known localization technique used in WSN systems due to its availability and simplicity. Radio location fingerprinting is one of the most effective of the many indoor positioning techniques due to its accuracy and low cost. In this study, we investigate the possibility of applying RSSI fingerprinting based Artificial Neural Network (ANN) jointly to determine the correct location of an object or person in indoor environments. According to our analysis and experiment result, Fingerprinting techniques can achieve accurate localization with high interference in a computer laboratory. In addition, this research uses a small-structure neural network (one hidden layer) to provide efficient computation with restricted resource requirements.

A radio frequency (RF) based device-free indoor localization (DFL) has attracted a lot of research effort due to its simplicity, less costly and compatibility with the existing hardware equipped with RF interface compared to the existing positioning system. An advanced monitoring system, known as Ambient Assisted Living (AAL) has been developed using DFL and Internet of Things (IoT) technologies. In this paper, we present a probabilistic DFL system using passive radio map method based on a non-parametric histogram-based approach to locate and map the passive target position in an indoor area. The proposed technique is based on a radio map concept in locating human position using received signal strength indicator (RSSI). The Bayesian inversion was introduced in the proposed approach for estimating the density function and the Probability of Error metric (PoE) was used to evaluate the tracking accuracy of the system. We firstly performed system analysis on the deterministic approach for comparison with the proposed probability approach. The results show that the probabilistic approach can accurately locate a passive target with an error probability of 0.0782 compared to the deterministic approach that gives high PoE of 0.2463.

Received Signal Strength Indicator (RSSI) is used in indoor positioning for measuring object distance to the base station. However, acquiring accurate RSSI values is challenging because wireless interference factors, such as multipath decline interference, make RSSI values of the same object fluctuate over time. Therefore, instead of a single RSSI, RSSI acquisition will collect a set of RSSI values from which the most moderate RSSI is derived. For this purpose, we propose an Enhanced Gaussian Mixture Model (EGMM) to derive a more precise RSSI for improving indoor positioning accuracy. EGMM enhances Gaussian Mixture Model (GMM) by applying Akaike information criterion (AIC) to determine the best K value for GMM to divide RSSI values into K sets representing signals from different paths. Then, EGMM identifies the most appropriate set of RSSI values to derive a more precise RSSI and thus improves the accuracy of indoor positioning. Our EGMM solution performs well in an open indoor space. The experiment is conducted with iBeacon devices, and the average error distance of EGMM is about 64% of those generated by existing Gaussian filtering. The average positioning error of EGMM is about 0.48 meter, which is adequate to indoor positioning accuracy.

Nowadays, location awareness becomes the key to numerous Internet of Things (IoT) applications. Among the various methods for indoor localisation, received signal strength indicator (RSSI)-based fingerprinting attracts massive attention. However, the RSSI fingerprinting method is susceptible to lower accuracies because of the disturbance triggered by various factors from the indoors that influence the link quality of radio signals. Localisation using body-mounted wearable devices introduces an additional source of error when calculating the RSSI, leading to the deterioration of localisation performance. The broad aim of this study is to mitigate the user’s body shadowing effect on RSSI to improve localisation accuracy. Firstly, this study examines the effect of the user’s body on RSSI. Then, an angle estimation method is proposed by leveraging the concept of landmark. For precise identification of landmarks, an inertial measurement unit (IMU)-aided decision tree-based motion mode classifier is implemented. After that, a compensation model is proposed to correct the RSSI. Finally, the unknown location is estimated using the nearest neighbour method. Results demonstrated that the proposed system can significantly improve the localisation accuracy, where a median localisation accuracy of 1.46 m is achieved after compensating the body effect, which is 2.68 m before the compensation using the classical K-nearest neighbour method. Moreover, the proposed system noticeably outperformed others when comparing its performance with two other related works. The median accuracy is further improved to 0.74 m by applying a proposed weighted K-nearest neighbour algorithm.

Indoor localization can provide a number of different services such as location-aware advertisement, indoor navigation and automating different appliances based on the user location. A number of different techniques such as time-difference- of-arrival, angle-of-arrival, time-of-flight, and received signal strength indicator (RSSI) have been used to provide Location Based Services (LBS). RSSI is one of the widely used methods as it is cost efficient and easy to implement. However, RSSI's performance is limited by multipath fading and indoor noise. Particle Filter (PF) is an accurate Bayesian Filtering algorithm that can improve the performance of RSSI-based indoor localization. However, PF is not able to satisfy the high accuracy requirement (possibly 10cm) of indoor localization. In this paper, we present Particle Filter-Extended Kalman Filter (PFEKF) cascaded algorithm that combines PF and EKF in series to reduce the impact of multipath effects and noise on the RSSI. Our experimental results show that PFEKF improves the localization accuracy by 31.3% and 33.9% in 3D and 2D environments respectively when compared with using only a PF.

Wireless sensor network (WSN) indoor location systems are based on effective distance estimation between the target node and anchors to find the node's precise location. Almost all low-cost wireless devices use the received signal strength indicator (RSSI) metric for estimating distance. This study aims to minimise the error in the estimated distance between the anchor nodes and the target node by proposing a model based on an Artificial Neural Network (ANN) for determining the distance using the RSSI. The proposed model was used to apply 3D indoor localisation. ANN is implemented in an Esp32 embedded system to make it compatible with WSN and IoT applications by using the TinyML approach for designing and implementing an efficient base station. The results show that the time required for the indoor localisation process does not exceed 40 ms, which is a very promising result for real-time localisation applications.

Towards a better indoor positioning system: A location estimation process using artificial neural networks based on a semi-interpolated database

Nowadays, location-based services are common in our daily lives. Traditional Global Positioning System (GPS) location can provide real-time location function in outdoor complex environments, but it is insufficient for indoor location. There are many indoor location technologies, such as ultrasound, Zigbee, RFID and WIFI. RFID location technology has attracted the attention of researchers due to its high precision and low cost. Most existing RFID location algorithms are based on RSSI (Received Signal Strength Indicator) measurement. When converting RSSI to distance, the inaccurate estimation of the path loss parameter may lead to large error. In order to reduce the deviation, this paper proposes a new RFID location algorithm. Specifically, the RSSI of the target tag is read in different directions of the antenna, and the position information is predicted by the general regression neural network, which is optimized by the optimization algorithm. The experimental results show the efficiency of our proposed algorithm.

Satellite remote sensing is widely being used by the researchers and geospatial scientists due to its free data access for land observation and agricultural activities monitoring. The world is suffering from food shortages due to the dramatic increase in population and climate change. Various crop genotypes can survive in harsh climatic conditions and give more production with less disease infection. Remote sensing can play an essential role in crop genotype identification using computer vision. In many studies, different objects, crops, and land cover classification is done successfully, while crop genotypes classification is still a gray area. Despite the importance of genotype identification for production planning, a significant method has yet to be developed to detect the genotypes varieties of crop yield using multispectral radiometer data. In this study, three genotypes of wheat crop (Aas-‘2011’, ‘Miraj-‘08’, and ‘Punjnad-1) fields are prepared for the investigation of multispectral radio meter band properties. Temporal data (every 15 days from the height of 10 feet covering 5 feet in the circle in one scan) is collected using an efficient multispectral Radio Meter (MSR5 five bands). Two hundred yield samples of each wheat genotype are acquired and manually labeled accordingly for the training of supervised machine learning models. To find the strength of features (five bands), Principle Component Analysis (PCA), Linear Discriminant Analysis (LDA), and Nonlinear Discernment Analysis (NDA) are performed besides the machine learning models of the Extra Tree Classifier (ETC), Random Forest (RF), Support Vector Machine (SVM), Decision Tree (DT), Logistic Regression (LR), k Nearest Neighbor (KNN) and Artificial Neural Network (ANN) with detailed of configuration settings. ANN and random forest algorithm have achieved approximately maximum accuracy of 97% and 96% on the test dataset. It is recommended that digital policymakers from the agriculture department can use ANN and RF to identify the different genotypes at farmer's fields and research centers. These findings can be used for precision identification and management of the crop specific genotypes for optimized resource use efficiency.

An indoor localization system in wireless sensor networks has become a hot development area. Received signal strength indicator (RSSI)-based localization is a promising technique since it requires a relatively low configuration, battery power and easy control. However, the received signal strength is influenced by channel interference and propagation environments. This characteristic affects channel stability and location accuracy in RSSI-based localization. As a result, we propose a novel indoor localization system consisting of a preprocessing method and a post-processing method. To improve channel stability, the pre-processing method selects an optimal channel in terms of the smallest distance error. The optimal channel is less affected to IEEE 802.11. To develop location accuracy, the post-processing method performs maximum likelihood estimation-based location tracking scheme considering the reliability of the RSSI value measurement. We apply three methods to the existing MLE to improve the reliability of the RSSI value. By using this indoor localization system with preprocessing and post-processing, the location error can be reduced. We also proved the performance of the indoor localization system by conducting experiments in a real indoor environment.

Various technologies such as WiFi, Bluetooth, and RFID are being used to provide indoor location-based services (LBS). In particular, a WiFi base using a WiFi AP already installed in an indoor space is widely applied, and the importance of indoor location recognition using deep running has emerged. In this study, we propose a WiFi-based indoor location recognition system using a smart watch, which is extended from an existing smartphone. Unlike the existing system, we use both the Received Signal Strength Indication (RSSI) and Basic Service Set Identifier (BSSID) to solve the problem of position recognition owing to the similar signal strength. By performing two times of filtering, we want to improve the execution time and accuracy through the learning of random forest based location awareness. In an unopened indoor space with five or more WiFi APs installed. Experiments were conducted by comparing the results according to the number of data for supposed system and a system based on existing WiFi fingerprint based random forest. The proposed system was confirmed to exhibit high performance in terms of execution time and accuracy. It has significance in that the system shows a consistent performance regardless of the number of data for location information.

Skin disease is a very vulnerable and severe issue in today's world. Skin disorder classification is crucial for diagnosis. Several new data mining algorithms have been developed to classify and interpret medical images. The functionality of the K-Nearest Neighbors (KNN) Random Forest (RF) Algorithm is described in this article, along with an analysis of its results. Furthermore, this study demonstrates a high-performing approach that saves both effort and money. The proposed model is designed based on KNN and the Random Forest algorithm. The patient can use this model to classify his skin disease as a primary detection, and the doctor also can ensure his judgment by using this proposed model. Traditional skin disease diagnosis is an expensive and time-consuming procedure. This paper's proposed classification model will identify ten different skin diseases. The Random Forest algorithm has a testing accuracy of 94.22 percent, and K-Nearest Neighbors (KNN) has a testing accuracy of 95.23 percent. The KNN algorithm has an F1 Score of 95.98 percent, whereas the Random Forest (RF) algorithm has an F1 Score of 95.94 percent. It can be increased by expanding the dataset and more feature extraction. This approach may benefit individuals with skin illness who are looking to save money and time as well as avoid skin cancer by identifying cancer at an early stage.

Recently, machine learning methodologies are playing an important role in detecting network intrusions (or attacks), which further helps the network administrator to take precautionary measures for preventing intrusions. In this project report, the authors propose to use various techniques like support vector machine, logistic regression, random forest, decision tree, and a hybrid model using artificial neural networks and random forest. The new hybrid model for network intrusion detection aims to achieve better precision, recall, accuracy, and F1 score. To reify this, combined artificial neural network and random forest algorithms are used. In the hybrid model, ANN works as a feature extractor and random forest as a classifier. The data is provided to ANN and the ANN generates new features on the basis of given input. The new features are then provided to random forest as an input. The random forest then classifies the data into different classes on the basis of new data. Here, KddCup99 dataset has been used to evaluate the models and accuracy achieved in each model.