A Machine Learning Approach to Ranging Error Mitigation for UWB Localization

Sensor networks can benefit greatly from location-awareness, since it allows information gathered by the sensors to be tied to their physical locations. Ultra-wide bandwidth (UWB) transmission is a promising technology for location-aware sensor networks, due to its power efficiency, fine delay resolution, and robust operation in harsh environments. However, the presence of walls and other obstacles presents a significant challenge in terms of localization, as they can result in positively biased distance estimates. We have performed an extensive indoor measurement campaign with FCC-compliant UWB radios to quantify the effect of non-line-of-sight (NLOS) propagation. From these channel pulse responses, we extract features that are representative of the propagation conditions. We then develop classification and regression algorithms based on machine learning techniques, which are capable of: (i) assessing whether a signal was transmitted in LOS or NLOS conditions; and (ii) reducing ranging error caused by NLOS conditions. We evaluate the resulting performance through Monte Carlo simulations and compare with existing techniques. In contrast to common probabilistic approaches that require statistical models of the features, the proposed optimization-based approach is more robust against modeling errors.

Due to the wide deployment of wireless local area networks (WLANs) and the easy acquirement of received signal strength (RSS), indoor localization based on RSS has attracted considerable attention in both academia and industry. In this paper, we extract the scatter factor from RSS and then propose a novel indoor localization scheme based on location fingerprinting and curve fitting techniques. The scheme is set up in two phases. In the offline phase, we create a fingerprint and apply the curve fitting technique to construct fitted RSS-distance functions in LOS (Line Of Sight) and NLOS (None Line Of Sight) condition, respectively. In online positioning phase, the proposed optimized fingerprinting-based localization algorithm (OFPL) uses trilateration technique to assist the fingerprinting localization in choosing the proper location of the receiver. Through conducting field and extensive experiments, we can draw the conclusion that our proposed algorithm can obtain some improvement in localization accuracy compared with the classical fingerprinting-based localization.

Localization is a key issue for wireless sensor network. The traditional localization methods achieve high location accuracy using received signal strength (RSS) in line of sight (LOS) environment. But the localization accuracy degrades significantly in nonline of sight (NLOS) environment. So the localization in NLOS environment is one of the most challenging for wireless sensor network. In this paper, we propose an RSS based localization algorithm in NLOS environment. A NLOS identification algorithm is firstly proposed. This algorithm does not need the parameters of the NLOS in prior. Then we correct the NLOS measurements by subtracting the mean of NLOS errors. Finally, the Kalman filter is employed to mitigate the process noise. In simulations, we demonstrate that the proposed algorithm achieves higher localization accuracy compared with other methods.

Our work has developed the Internet of Things (IoT) LoRa prototype to support the remote measurement of temperature at the frequency of 915 MHz. The device is structured by a temperature sensor, memory storage, data processing and data visualization using Cloud Server on the Web Browser. The performance is evaluated by both LoS and Non-LoS field trial of 14 dBm transmit power. By using the IoT platform of LoRa, end devices with low energy consumption that connected to gateways, shall perform a more efficient data transmission to other devices and network servers. We conduct field trial on Line of Sight (LoS) and Non-LoS condition in the surrounding area of University Indonesia campus. The results have indicated that prototype is working well on both condition, while RSSI may vary from −50 dB to −105 dB. For LoS condition, RSSI is at intervals of −55 dB to −90 dB for distance of 100 meters to 700 meters, while for Non-LoS condition is at intervals of −62 dB to −99 dB for distance of 100 meters to 400 meters. Accordingly, the LoRa is proven to be efficiently used for the IoT applications of both LoS and Non-LoS condition, while a specific design should be adjusted.

Removed.

Indoor localization system based on received signal strength (RSS) often operate under non-line-of-sight (NLOS) conditions that can cause ranging errors. To identify non-line-of-sight status and line-of-sight (LOS) status and improve the accuracy of indoor localization, a D-S evidence theory based NLOS identification algorithm was proposed. The algorithm is divided into two parts. In the first part, respectively extract a variety of RSS features in NLOS and LOS status and choose the appropriate filter for data processing; in the second part, RSS features fused by D-S evidence theory are used to identify NLOS and LOS status. In an actual test environment, the algorithm was applied to fingerprinting localization, then analyzed the advantages of fusing RSS features and the influence of RSS features selection on experimental results. Finally, compared with several commonly used NLOS and LOS identification algorithm, it shows the proposed algorithm can deal with the indoor localization environment with obstacles with a high localization accuracy and good stability.

Relative positioning accuracy between two devices is dependent on the precise range measurements. Ultra-wideband (UWB) technology is one of the popular and widely used technologies to achieve centimeter-level accuracy in range measurement. Nevertheless, harsh indoor environments, multipath issues, reflections, and bias due to antenna delay degrade the range measurement performance in line-of-sight (LOS) and non-line-of-sight (NLOS) scenarios. This article proposes an efficient and robust method to mitigate range measurement error in LOS and NLOS conditions by combining the latest artificial intelligence technology. A GP-enhanced non-linear function is proposed to mitigate the range bias in LOS scenarios. Moreover, NLOS identification based on the sliding window and Bayesian Conv-BLSTM method is utilized to mitigate range error due to the non-line-of-sight conditions. A novel spatial–temporal attention module is proposed to improve the performance of the proposed model. The epistemic and aleatoric uncertainty estimation method is also introduced to determine the robustness of the proposed model for environment variance. Furthermore, moving average and min-max removing methods are utilized to minimize the standard deviation in the range measurements in both scenarios. Extensive experimentation with different settings and configurations has proven the effectiveness of our methodology and demonstrated the feasibility of our robust UWB range error mitigation for LOS and NLOS scenarios.

At present, location technology divides into outdoor location technology and indoor location technology. Comparing with other indoor location methods, ultra-wide band (UWB)indoor location technology has the advantages of strong anti-multipath interference ability and signal penetration ability. But, when this method faces large obstacles or metal obstacles, non line of sight (NLOS) will still occur. Indoor location system is a part of our project Home Care System for The Elderly. The NLOS of Indoor location system reduces the reliability of Home Care System for The Elderly. In order to improve the accuracy of location and efficiency of operation, an NLOS detection method based on k-nearest neighbors (KNN) algorithm is proposed. This algorithm modeled the training samples to find the K value with the highest recognition accuracy. The experimental results show that The NLOS detection accuracy of this method is 78% in the simulated apartment scene. With the same number of samples, the accuracy of the widely used Convolutional Neural Networks (CNN) algorithm is 67%. The experimental results show that KNN algorithm is more suitable for Home Care System for The Elderly for it can keep high accuracy with less requirement for the number of samples than CNN.

High-precision indoor localization systems (ILSs) are critical for applications such as human smartphone navigation, autonomous robotics and automated warehouse and factory design. This paper presents a novel fingerprinting-based ILS, which features a decimeter-level localization accuracy, the ability to function in a constantly changing non line-of-sight (NLoS) environment, and user privacy protection without the need for heavy computations. The proposed ILS is able to maintain its localization accuracy in a constantly changing environment and to camouflage the user’s location by leveraging multipath propagation. The method was successfully tested both by experimental verification using the ultra-wideband communication standard and a ray-tracing simulation. An average localization error of 6 cm is demonstrated for a stationary or slow-moving receiver. An average error of 30 cm is demonstrated for a receiver that is moving at a fast walking pace. The obtained localization accuracy is comparable to the accuracy of the state-of-the-art localization algorithms. At the same time, the proposed approach solves two practical challenges faced by ILSs: robustness to changing environments with moving objects and the high computation requirements of user privacy protection. The high degree of user privacy was evaluated using a set of corresponding metrics.

Indoor tracking and navigation (ITN) mainly depend on indoor localization. An impulse radio ultra-wideband (IR-UWB) is the most advanced technology for precision indoor localization. Besides its precision, the IR-UWB also has low complex hardware, low power consumption, and a flexible data rate that makes it the ideal candidate for ITN. However, two significant challenges impede the achievement of high-resolution accuracy and optimum performance: non-line-of-sight (NLOS) channel condition and multipath propagation (MPP). To enhance the performance under these conditions, the ranging error is estimated and corrected using parameters’ uncertainties. The uncertainties in the channel’s parameters have a relationship with the error, and these uncertainties are induced due to the NLOS and MPP propagation conditions. The parameters are collected in real-time experimental setups in two different environments. A proposed fuzzy inference model utilizes these uncertainties and the relationship to estimate ranging errors. The model is evaluated, and its performance is gauged in terms of residual ranging error cumulative distribution, root mean square error, and outage probability parameters using experimental measurements and compared with the state-of-the-art work. Moreover, the proposed fuzzy model is evaluated for computational complexity in terms of execution time and compared with the state-of-the-art work. The time is estimated on the targeted embedded system. The experimental and simulated results show that the proposed model effectively minimizes the ranging errors and computational burden. Moreover, the model does not induce a delay in estimating ranging error due to the non-statistical based solution.

Positioning is one of the key technologies to support the application of intelligent connected vehicles (ICV). GNSS, as one of main measures of ICV positioning, cannot be available in indoor scenarios. The emergence of indoor wireless positioning technologies, such as UWB and WIFI, has solved this problem. However, they are affected by the non line of sight (NLoS) propagation of the signal, resulting in lower positioning accuracy. To improve the vehicle indoor positioning performance with NLoS state, we propose an ICV indoor positioning method called CGPC-EMPF. First, the Chan-GN algorithm is improved with permutation and combination consistency, whose result is used as the observation of particle filter. For the vehicle motion prediction without a Strapdown inertial navigation system, we propose a vehicle motion prediction model based on a priori information. Then, the NLoS correction model is established by a particle filter. Finally, the corrected measurements are substituted into the Chan-GN algorithm to obtain the final results. It realizes accuracy enhancement by introducing a vehicle motion state to correct errors coursed by NLoS. Compared with the multi-sensor fusion positioning algorithm, our method has lower complexity and cost due to a single sensor application.

With the rise of the Internet of Things and smart cities, indoor positioning technology becomes more and more important. Among the existing network technologies, ultra-wide band (UWB) has the advantages of high transmission rate, and strong penetration ability. However, its signal propagation is susceptible to N on- Line-of-Sight (NLOS) effects. Therefore, this paper focuses on indoor positioning in NLOS environments, and studies the Chan-Taylor co-weighted positioning algorithm improved by machine learning. The main research contents include: 1. A machine learning method is proposed to identify and classify NLOS signals. Using a sliding window to dynamically calculate the variance and mean of multiple groups of TDOA measurements in a short period of time as the input to the recognition algorithm, and classify them into three categories: Line-of-Sight (LOS), low NLOS error, and high NLOS error. 2. Based on the Chan- Taylor co-location algorithm, the algorithm introduces the concept of weighting coefficient, optimizes the observation data, and finally obtains the target position coordinates. The simulation results show that the positioning accuracy and system stability of the algorithm are improved compared with the Chan algorithm and the Chan-Taylor co-location algorithm.

Long-Range Wide-Area Networks (LoRaWAN) allow the transmission of data via radio link from sensors, which are potentially isolated or difficult to access, to gateways and servers that are connected to cellular networks for data processing, exchange, or relay, with low transmission power. This concept employs Long-Range (LoRa) modulation and has led to the emergence of many applications for the monitoring and tracking of objects. However, due to its characteristic of a low data rate for low-power communication, the transmission of information with LoRa technology is not suitable for the fast real-time monitoring of data. Additionally, due to its narrow bandwidth, an attempt to perform localization through the LoRa modulation technique will result in very limited accuracy because of its inability to resolve multipath problems. Thus, in this paper, we propose a multi-standard Ultra-Wide Bandwidth (UWB) and LoRa end-device that is capable of measuring location with high accuracy using UWB technology and then transmitting the location information through LoRa method to gateways and the Internet of Things Network. The results of measurements in indoor and outdoor scenarios show a UWB localization accuracy that is of sub-meter level, being between 10 and 33 cm, and a UWB range of 124 m in Line-of-Sight (LOS) and 55 m in Non-Line-of-Sight (NLOS) applications, respectively.

Determination of position accuracy for geolocation is a fundamental issue in wireless sensor networks. This paper derives the position error bound (PEB), a fundamental limit for localization accuracy, by using information inequality. In particular, the authors consider all multipath propagation parameters, and hence our bound is tighter than those of previous work. To alleviate computation complexity, the authors put forth the notion of equivalent Fisher information (EFI) to characterize the localization accuracy. This approach also unifies the contributions from line-of-sight (LOS), non-line-of-sight (NLOS), and a priori knowledge to the PEB in a consistent form. These results are applicable to ultra-wide bandwidth (UWB) systems as a specific case.

Ultra Wideband (UWB) signals are a promising choice for indoor positioning applications, since they are able to penetrate walls to a certain extent. Nevertheless, signal reflections and Non-Line-of-Sight (NLOS) propagation cause bias in the measured range. This ranging error can be corrected with Machine Learning (ML) methods such as Convolutional Neural Networks (CNNs). However, these ML models often generalize poorly between different environments. In this work we present an instance-based Transfer Learning (TL) approach, that enables generalizing a CNN-based ranging error mitigation approach to a new situation with only a few unlabeled training samples. The performance of the UWB error correction approach is demonstrated in a real-life infrastructure-free cooperative positioning setting.