상황인식 서비스에서 위치설정은 매우 중요한 기술 요소이다. RSSI와 같은 전파특성 지수가 편리하고 저렴한 이유로 널리 사용되고 있다. 그러나 RSSI는 시간에 따른 변화가 크고 다중경로에 취약성으로 실내 환경에서 위치 설정에 적절하지 않다. 본 논문은 WLAN의 RF 전파특성 지수인 CSI(Channel State Information)를 이용하여 실내에서 임의 단말의 위치설정을 위한 거리추정에 소요되는 절차와 기법들을 제시한다. 먼저 거리추정의 포괄적인 절차를 정의하고, 거리대비 전파손실 모델의 환경 특성값을 보정하는 알고리즘을 제시한다. 상용 WLAN 통신모듈을 이용한 실험을 통하여 제안된 절차와 기법의 유용성에 대해서 분석한다.

we present Wireless Indoor Localization System (WILS) that enables commercial WiFi infrastructures to implement localization in the complex indoor environment. Receive Signal Strength Indication (RSSI) is utilized by some systems to get a high accuracy of localization, but they are easily be affected by the environment. Other use modified hardware or firmware, such as increasing the number of antennas to obtain enormous location precision. To solve this problem, WILS uses physical layer information, called Channel State Information (CSI). And it does not require modification in any hardware and firmware. WILS makes four technical contributions for achieving a high accuracy in localization. At first, it proposes a new algorithm, called Modified MUSIC (MMUSIC), and it can estimate Angle of Arrival (AoA) and Time of Flight (ToF) at the same time from CSI information. Next, it introduces some methods to sanitize the phase, and improve the accuracy of measured phase. Third, it presents an idea of window slides to do CSI Smoothing for CSI measurements. Finally, it utilizes clustering algorithm to identify Line of Sight (LoS) path, and combines the AoA for LoS path and RSSI to locate the target. Our implementation on commodity WiFi cards demonstrates that WILS’s accuracy is comparable to advanced localization systems, it can achieve a median accuracy of 60 cm. Keywords-indoor localization; channel state information (CSI); angle of arrival (AoA); time of flight (ToF); modified MUSIC, line of sight(LoS); phase sanitization

ii Acknowledgements iii

With the rapid growth of indoor positioning requirements without equipment and the convenience of channel state information acquisition, the research on indoor fingerprint positioning based on channel state information is increasingly valued. In this article, a multi-level fingerprinting approach is proposed, which is composed of two-level methods: the first layer is achieved by deep learning and the second layer is implemented by the optimal subcarriers filtering method. This method using channel state information is termed multi-level fingerprinting with deep learning. Deep neural networks are applied in the deep learning of the first layer of multi-level fingerprinting with deep learning, which includes two phases: an offline training phase and an online localization phase. In the offline training phase, deep neural networks are used to train the optimal weights. In the online localization phase, the top five closest positions to the location position are obtained through forward propagation. The second layer optimizes the results of the first layer through the optimal subcarriers filtering method. Under the accuracy of 0.6 m, the positioning accuracy of two common environments has reached, respectively, 96% and 93.9%. The evaluation results show that the positioning accuracy of this method is better than the method based on received signal strength, and it is better than the support vector machine method, which is also slightly improved compared with the deep learning method.

With the rapid development of wireless network technology, wireless passive indoor localization has become an increasingly important technique that is widely used in indoor location-based services. Channel state information (CSI) can provide more detailed and specific subcarrier information, which has gained the attention of researchers and has become an emphasis in indoor localization technology. However, existing research has generally adopted amplitude information for eigenvalue calculations. There are few research studies that have used phase information from CSI signals for localization purposes. To eliminate the signal interference existing in indoor environments, we present a passive human indoor localization method named FapFi, which fuses CSI amplitude and phase information to fully utilize richer signal characteristics to find location. In the offline stage, we filter out redundant values and outliers in the CSI amplitude information and then process the CSI phase information. A fusion method is utilized to store the processed amplitude and phase information as a fingerprint database. The experimental data from two typical laboratory and conference room environments were gathered and analyzed. The extensive experimental results demonstrate that the proposed algorithm is more efficient than other algorithms in data processing and achieves decimeter-level localization accuracy.

With the rapid development of Internet of Things (IoT) technology, location-based services have been widely applied in the construction of smart cities. Satellite-based location services have been utilized in outdoor environments, but they are not suitable for indoor technology due to the absence of global positioning system (GPS) signal. Therefore, many indoor localization technologies and systems have emerged by utilizing many other signals. In particular, fingerprinting localization has recently garnered attention because its promising performance. In this work, we aim to study recent indoor localization technologies and systems based on various fingerprints, which use machine learning and intelligent algorithms. We also present the architecture of intelligent localization. The development of indoor localization technology should have the ability of self-adaptation and self-learning in the future. And the architecture shows how to make localization become more “smart” by advanced techniques. The state-of-the-art localization systems’ working principles are summarized and compared in terms of their localization accuracy, latency, energy consumption, complexity, and robustness. We also discuss the challenges of existing indoor localization technologies, potential solutions to these challenges, and possible improvement measures.

With the rapid development of Internet of Things (IoT) techn iques, RF sensing has found wide applications for, e.g., indoor localization, activit y recognition, and healthcare. In this dissertation, we investigate the problem of RF sensing for IoT using channel state information (CSI) and machine learning techniques. In particular, our work mainl y focuses on indoor localization using deep learning and vital sign monitoring for RF sensing. In this dissertation, we first study the problem of CSI based in door localization. For first three works, we exploit deep learning for three different in door localization systems using CSI amplitudes, CSI calibrated phases, and CSI bimodal data, resp ectively. Moreover, we study and analyze CSI data, which is stable for indoor localization. We consider deep autoencoder networks to train CSI data, and employ the weights of the deep netw ork to represent fingerprints. A greedy learning algorithm is leveraged to train the weights layer-by-layer to reduce computational complexity, where a sub-network between two consecutive la y rs forms a Restricted Boltzmann Machine (RBM). In the online stage, we use a probabilistic meth od for online location estimation. Then, we exploit deep convolutional neural networks (DCNN) f or indoor localization. Since DCNN is a supervised method, it only requires to train one grou p f weights for all the training data with related labels, which is different with our prior w orks that requires training weights for every training location. Specially, we use estimated angle of arrival (AOA) images from CSI data as input to the DCNN. By executing four convolutional and subsa mpling layers, the system can automatically extract the features of the estimated AOA ima ges, to obtain training weights. To improve indoor localization accuracy, we propose deep resi dual sharing learning for training two channels CSI tensor data. Moreover, we can stack many residua l haring blocks for adding the depth of the deep network, thus achieving higher learning an d representation ability for CSI tensor

With the proliferation of Internet of Things (IoT) applications, billions of household appliances, phones, smart devices, security systems, environment sensors, vehicles, buildings, and other radio-connected devices will transmit data and communicate with each other or people, and it will be possible to constantly measure and track virtually everything. Among the various approaches to measuring what is happening in the surrounding environment, wireless sensing has received increasing attention in recent years because of the ubiquitous deployment of wireless radio devices. In addition, human activities affect wireless signal propagation, so understanding and analyzing how these signals react to human activities can reveal rich information about the activities around us.

With the popularization of mobile devices, location based services (LBS) have received significant attention. Global Navigation Satellite System (GNSS) that widely used outdoors is unable to achieve high-precision positioning indoors, thus various indoor positioning methods are proposed for precise positioning. Wi-Fi based indoor positioning is widely used to provide accurate and efficient location services due to the convenience of hardware facilities. Channel state information (CSI) is an essential feature provided by Wi-Fi signals, which depicts the fine-grained features of channels, and attracts great attention in recent years. This paper comprehensively reviews the positioning technology based on CSI, and compares the primary positioning systems that makes great progress. This paper are mainly represented from two aspects: geometry-based positioning method and fingerprint-based positioning method. The former introduces angle-based method, range-based method, and the combination of both. The later mines the development process from traditional machine learning method to deep learning method. At the same time, this article also gives a brief overview of CSI’s multi-source fusion technologies. Finally, this paper summarizes the challenges and points out the trends in this rapidly developing field.

With the marvelous development of wireless techniques and ubiquitous deployment of wireless systems indoors, myriad indoor location-based services (ILBSs) have permeated into numerous aspects of modern life. The most fundamental functionality is to pinpoint the location of the target via wireless devices. According to how wireless devices interact with the target, wireless indoor localization schemes roughly fall into two categories: device based and device free. In device-based localization, a wireless device (e.g., a smartphone) is attached to the target and computes its location through cooperation with other deployed wireless devices. In device-free localization, the target carries no wireless devices, while the wireless infrastructure deployed in the environment determines the target’s location by analyzing its impact on wireless signals. This article is intended to offer a comprehensive state-of-the-art survey on wireless indoor localization from the device perspective. In this survey, we review the recent advances in both modes by elaborating on the underlying wireless modalities, basic localization principles, and data fusion techniques, with special emphasis on emerging trends in (1) leveraging smartphones to integrate wireless and sensor capabilities and extend to the social context for device-based localization, and (2) extracting specific wireless features to trigger novel human-centric device-free localization. We comprehensively compare each scheme in terms of accuracy, cost, scalability, and energy efficiency. Furthermore, we take a first look at intrinsic technical challenges in both categories and identify several open research issues associated with these new challenges.

With the increasing popularity of location-based services, channel state information (CSI) has received widespread attention in positioning due to the fine-grained information it provides. However, the raw amplitudes of CSI are especially sensitive to noise and easily effected by the frequency-selective fading. As the expanding of test area, positioning is disturbed by random noise and insufficient resolution of features severely. In this paper, we propose Cluster-Mapping (C-Map), an adaptive pre-processing system for CSI amplitude-based fingerprint localization. This system models positioning as a classification problem, an adaptive processing mechanism is constructed according to the characteristics of CSI amplitude. C-Map mainly contains two parts: dynamic denoising and feature enhancement. In the dynamic denoising, effective features are screened out by clustering iteratively, without specifying parameters according to the experimental environment. In the feature enhancement, polynomial fitting with regularization is used to reduce jagged fluctuations, then the trend of variation over 30 sub-carriers are described by combining derivation with nonlinear kernel function. Extensive experiments are conducted in typical environments to verify the superior performance of C-Map for the preprocessing of CSI amplitude. Compared with the combination of mean and multidimensional scaling (MDS), the average positioning error of C-Map is reduced 28.1% in comprehensive indoor environment, and 19% in garage. Furthermore, compared with the combination of DBSCAN and principal component analysis (PCA), the average positioning error of C-Map is reduced 33.1% in comprehensive indoor environment, and 28% in garage.

With the increasing need for indoor localization based application and widely use of Wi-Fi device, Wi-Fi signal is employed to the human positioning of indoor environments due to the absence of global position system. Differing from prior works of position-based scheme and track-based scheme, the path reconstruction for indoor localization proposed in this paper utilizes path segmentation and combination to reconstruct trajectory of human motion. To enhance the universality in different environments and roam easily to new scenarios, our proposed method applies unsupervised learning identification models based on artificial neural network to recognize different benchmark paths. Then, the segments of test motion path will be matched with the most similar benchmark paths, respectively, to reconstruct whole motion path in chronological order. Experimental results of two experiment scenarios indicate that the proposed system could achieve satisfactory accuracy and robustness.

With the increasing demands for indoor location-based services, fingerprinting-based localization attracts considerable attention, due to that it could achieve high localization accuracy using simple equipment. However, the main problem of fingerprinting localization is that with the change of the indoor environment, the fingerprint database would be outdated, which inevitably leads to localization performance degradation. To tackle this issue, based on deep learning, this article proposes a self-calibration time-reversal (TR) fingerprinting localization approach to mitigate the effects of environmental changes without updating the fingerprint database. In the offline stage, the amplitude autoencoder (A-AE) and the phase autoencoder (P-AE) are, respectively, trained without labels to record features of the current environment. In the online stage, the trained A-AE and P-AE are used to adaptively calibrate the real-time measurements which may have been distorted due to the environmental changes. Based on the calibrated measurements, a modified TR resonating strength (TRRS) is presented for localization. The experimental results confirm the effectiveness of the proposal.

With the increasing demand of location-based services, indoor ranging method based on Received Signal Strength Indicator (RSSI) or Channel State Information (CSI) has become an increasingly important technique due to its low hardware requirement and high accuracy. Due to robustness against the multipath effect, frequency domain Channel State Information (CSI) of Orthogonal Frequency Division Multiplexing (OFDM) systems is supposed to provide an excellent ranging measurement for indoor localization. In this paper, we propose a novel signal attenuation model for indoor ranging with RSSI and CSI. The proposed attenuation model scheme is implemented and validated with experiments in a typical indoor environment. Experimental results are presented to confirm that the proposed model can effectively reduce ranging error, compared with two existing methods in a typical indoor environment.

With the increasing demand of location-based services, indoor localization based on fingerprinting has become an increasingly important technique due to its high accuracy and low hardware requirement. In this paper, we propose PhaseFi, a fingerprinting system for indoor localization with calibrated channel state information (CSI) phase information. In PhaseFi, the raw phase information is first extracted from the multiple antennas and multiple subcarriers of the IEEE 802.11n network interface card by accessing the modified device driver. Then a linear transformation is applied to extract the calibrated phase information, which we prove to have a bounded variance. For the offline stage, we design a deep network with three hidden layers to train the calibrated phase data, and employ the weights of the deep network to represent fingerprints. A greedy learning algorithm is incorporated to train the weights layer-by-layer to reduce computational complexity, where a subnetwork between two consecutive layers forms a restricted Boltzmann machine. In the online stage, we use a probabilistic method based on the radial basis function for online location estimation. The proposed PhaseFi scheme is implemented and validated with extensive experiments in two representation indoor environments. It is shown to outperform three benchmark schemes based on CSI or received signal strength in both scenarios.

With the increasing demand of location-based services, indoor localization based on fingerprinting has become an increasingly important technique due to its high accuracy and low hardware requirement. In this paper, we propose PhaseFi, a fingerprinting system for indoor localization with calibrated channel state information (CSI) phase information. In PhaseFi, the raw phase information is first extracted from the multiple antennas and multiple subcarriers of the IEEE 802.11n network interface card (NIC) by accessing the modified driver. Then a linear transform is used to extract the calibrated phase information, which is proven to have a bounded variance. For the offline stage, we design a deep network with three hidden layers to train the calibrated phase data, and employ weights to represent fingerprints. A greedy learning algorithm is incorporated to train the weights layer-by-layer to reduce computational complexity, where a sub-network between two continuous layers forms a Restricted Boltzmann Machine (RBM). In the online stage, we use a probabilistic method based on the radial basis function (RBF) for online location estimation. The proposed PhaseFi scheme is implemented and validated with intensive experiments in two representation indoor environments. It outperforms other three benchmark schemes based on CSI or RSS in both scenarios.

With the increasing demand of location-based services, indoor localization based on fingerprinting has become an increasingly important technique due to its high accuracy and low hardware requirement. In this paper, we propose PhaseFi, a fingerprinting system for indoor localization with calibrated channel state information (CSI) phase information. In PhaseFi, the raw phase information is first extracted from the multiple antennas and multiple subcarriers of the IEEE 802.11n network interface card (NIC) by accessing the modified driver. Then a linear transform is used to extract the calibrated phase information, which is proven to have a bounded variance. For the offline stage, we design a deep network with three hidden layers to train the calibrated phase data, and employ weights to represent fingerprints. A greedy learning algorithm is incorporated to train the weights layer-by-layer to reduce computational complexity, where a sub-network between two continuous layers forms a Restricted Boltzmann Machine (RBM). In the online stage, we use a probabilistic method based on the radial basis function (RBF) for online location estimation. The proposed PhaseFi scheme is implemented and validated with intensive experiments in two representation indoor environments. It outperforms other three benchmark schemes based on CSI or RSS in both scenarios.

With the fast-growing demand of location-based services in indoor environments, indoor positioning based on fingerprinting has attracted significant interest due to its high accuracy. In this paper, we present a novel deep-learning-based indoor fingerprinting system using channel state information (CSI), which is termed DeepFi. Based on three hypotheses on CSI, the DeepFi system architecture includes an offline training phase and an online localization phase. In the offline training phase, deep learning is utilized to train all the weights of a deep network as fingerprints. Moreover, a greedy learning algorithm is used to train the weights layer by layer to reduce complexity. In the online localization phase, we use a probabilistic method based on the radial basis function to obtain the estimated location. Experimental results are presented to confirm that DeepFi can effectively reduce location error, compared with three existing methods in two representative indoor environments.

With the fast growing demand of location-based services in indoor environments, indoor positioning based on fingerprinting has attracted a lot of interest due to its high accuracy. In this thesis, we present a novel deep learning based indoor fingerprinting system using Channel State Information (CSI), which is termed DeepFi. Based on three hypotheses on CSI, the DeepFi system architecture includes an off-line training phase and an on-line localization phase. In the off-line training phase, deep learning is utilized to train all the weights as fingerprints. Moreover, a greedy learning algorithm is used to train all the weights layer-bylayer to reduce complexity. In the on-line localization phase, we use a probabilistic method based on the radial basis function to obtain the estimated location. Experimental results are presented to confirm that DeepFi can effectively reduce location error compared with three existing methods in two representative indoor environments.

With the fast growing demand of location-based services in indoor environments, indoor positioning based on fingerprinting has attracted a lot of interest due to its high accuracy. In this paper, we present a novel deep learning based indoor fingerprinting system using Channel State Information (CSI), which is termed DeepFi. Based on three hypotheses on CSI, the DeepFi system architecture includes an off-line training phase and an on-line localization phase. In the off-line training phase, deep learning is utilized to train all the weights as fingerprints. Moreover, a greedy learning algorithm is used to train all the weights layer-by-layer to reduce complexity. In the on-line localization phase, we use a probabilistic method based on the radial basis function to obtain the estimated location. Experimental results are presented to confirm that DeepFi can effectively reduce location error compared with three existing methods in two representative indoor environments.