Mitigating multipath of cellular long-term evolution (LTE) signals for robust positioning in urban environments is considered. A computationally efficient receiver, which uses a phase-locked loop (PLL)–aided delay-locked loop (DLL) to track the received LTE signals, is presented. The PLL-aided DLL uses orthogonal frequency division-multiplexing (OFDM)–based discriminator functions to estimate and track the time-of-arrival. The code phase and carrier phase performances in an additive white Gaussian noise (AWGN) channel are evaluated numerically. The effects of multipath on the code phase and carrier phase are analyzed, demonstrating robust multipath mitigation for high transmission LTE bandwidths. The average of the DLL discriminator functions over multiple LTE symbols is presented to reduce the pseudorange error. The proposed receiver is evaluated on a ground vehicle in an urban environment. Experimental results show a root mean square error (RMSE) of 3.17 m, a standard deviation of 1.06 m, and a maximum error of 6.58 m between the proposed LTE receiver and the GPS navigation solution over a 1.44 km trajectory. The accuracy of the obtained pseudoranges with the proposed receiver is compared against two algorithms: estimation of signal parameters by rotational invariance techniques (ESPRIT) and EKAT (ESPRIT and Kalman filter).

LTE receiver design and multipath analysis for navigation in urban environments

Global Navigation Satellite System (GNSS) shadow matching is a new positioning technique that determines position by comparing the measured signal availability and strength with predictions made using a three-dimensional (3D) city model. It complements conventional GNSS positioning and can significantly improve cross-street positioning accuracy in dense urban environments. This paper describes how shadow matching has been adapted to work on an Android smartphone and presents the first comprehensive performance assessment of smartphone GNSS shadow matching. Using GPS and GLONASS data recorded at 20 locations within central London, it is shown that shadow matching significantly outperforms conventional GNSS positioning in the cross-street direction. The success rate for obtaining a cross-street position accuracy within 5 m, enabling the correct side of a street to be determined, was 54·50% using shadow matching, compared to 24·77% for the conventional GNSS position. The likely performance of four-constellation shadow matching is predicted, the feasibility of a large-scale implementation of shadow matching is assessed, and some methods for improving performance are proposed. A further contribution is a signal-to-noise ratio analysis of the direct line-of-sight and non-line-of-sight signals received on a smartphone in a dense urban environment.

/pdf/smartphone-shadow-matching-for-better-cross-street-gnss-1n17f6kv7d.pdf

Smartphone Shadow Matching for Better Cross-street GNSS Positioning in Urban Environments

Current GPS positioning accuracy in urban areas is still unsatisfactory for various applications, including pedestrian navigation and autonomous driving. Due to the ineffectiveness of special corrector designs against non-line-of-sight (NLOS) reception, the research regarding NLOS signals has been increasing in the recent years. This study first develops an algorithm to detect NLOS signals from the pseudorange measurements by using a 3D building model, ray-tracing simulation, and known receiver position. According to the analysis of 24 h of collected NLOS data, a new finding is that NLOS pseudorange delay is highly correlated with the elevation angle of satellite instead of the received signal strength. Thus, we further propose an innovative NLOS model using two variables, the elevation angle and the distance between the receiver and building that reflect the NLOS. The proposed model is evaluated in both pseudorange and position domains. Based on the experiment results regarding pseudorange error, the difference between the proposed model and the collected NLOS measurement is very small. Finally, the proposed model is applied to a hypothesis-based positioning method and achieves about 6.3 m in terms of horizontal positioning accuracy, which is only slightly worse than the method applied with ray-tracing simulation.

/pdf/analysis-and-modeling-gps-nlos-effect-in-highly-urbanized-1f9pinh0ye.pdf

Analysis and modeling GPS NLOS effect in highly urbanized area

Insufficient localization accuracy of global navigation satellite system (GNSS) receivers is one of the challenges to implement advanced intelligent transportation system in highly urbanized areas. Multipath and non-line-of-sight (NLOS) effects strongly deteriorate GNSS positioning performance. This paper aims to train a classifier by supervised machine learning to separate the type of GNSS pseudorange measurement into three categories, clean, multipath and NLOS. Several features obtained or calculated from the GNSS raw data are evaluated. This paper also proposes a new feature to indicate the consistency between measurements of pseudorange and Doppler shift. According to the experiment result, about 75% of classification accuracy can be achieved using a support vector machine (SVM) classifier trained by the proposed feature and received signal strength.

/pdf/gnss-multipath-detection-using-a-machine-learning-approach-2exe2qq06k.pdf

GNSS multipath detection using a machine learning approach

In urban areas, GPS signals are often reflected or blocked by buildings, which causes multipath effects and non-line-of-sight (NLOS) reception respectively consequently degrading GPS positioning performance. While improved receiver design can reduce the effect of multipath to some extent, it cannot deal with NLOS. Modelling methods based on measurements have shown promise to reduce the effect of NLOS signal reception. However, this depends on their ability to accurately and reliably classify line-of-sight (LOS), multipath and NLOS signals. The traditional method is based on one feature using signal strength as measured by the carrier to noise ratio, C/N0. However, this feature is ineffective in capturing the characteristics of multipath and NLOS in all environments. In this paper, to improve the accuracy of signal reception classification, we are using the three features of C/N0, pseudorange residuals and satellite elevation angle with a gradient boosting decision tree (GBDT) based classification algorithm. Experiments are carried out to compare the proposed algorithm with classifiers based on decision tree, distance weighted k-nearest neighbour (KNN) and the adaptive network-based fuzzy inference system (ANFIS). Test results from static receivers in urban environments, show that the GBDT based algorithm achieves a classification accuracy of 100%, 82% and 86% for LOS, multipath and NLOS signals, respectively. This is superior to the other three algorithms with the corresponding results of 100%, 82% and 84% for the Distance-Weighted KNN, 99%, 70% and 65% for the ANFIS and 98%, 35% and 95% for the traditional decision tree. With the NLOS detection and exclusion, the proposed GBDT with multi-feature based method can provide a positioning accuracy improvement of 34.1% compared to the traditional C/N0 based method.

A gradient boosting decision tree based GPS signal reception classification algorithm

GNSS signal classification to LOS and NLOS signals is of great value for conventional ranging-based and shadow matching algorithms. The most common attribute for performing this classification is the signal strength. Alas, such classification is often insufficient, in particular, in urban environments. In this paper, we present a novel approach for LOS/NLOS classification utilizing supervised machine learning algorithms. Provided with a sufficiently large labeled training set, the proposed approach is able to predict with high certainty (>85 percent) the satellites’ visibility status in dense urban regions. This achievement was possible due to the vast raw measurements supplied for the algorithm and using sophisticated feature-selection techniques.

A Robust GNSS LOS/NLOS Signal Classifier

Commercial GNSS devices tend to perform poorly in urban canyon environments. The dense and tall buildings block the signals from many of the satellites. In this work, we present a particle filter algorithm for a Shadow Matching framework to face this problem. Given a 3D city map and given the satellites' signal properties, the algorithm calculates in real-time invalid regions inside the Region Of Interest (ROI). This approach reduces the ROI to a fraction of its original size. We present a general framework for a Shadow Matching positioning algorithm based on a modified particle filter. Using simulation experiments we have shown that the suggested method can improve the accuracy of existing GNSS devices in urban regions. Moreover, the proposed algorithm can be efficiently extended to 3D positioning in high sampling rate, inherently applicable for UAVs and Drones. Copyright © 2015 Institute of Navigation.

https://onlinelibrary.wiley.com/doi/pdf/10.1002/navi.117

A Robust Shadow Matching Algorithm for GNSS Positioning

Receiver modules in Global Navigation Satellite Systems (GNSS) are capable of providing positioning and velocity estimations that are sufficiently accurate for the purpose of road navigation. However, even in optimal open-sky conditions, GNSS-based positioning carries an average error of 2–4 m. This imposes an effective limitation on GNSS-based vehicle lane detection, a desired functionality for various navigation and safety applications. In this paper, we present a novel framework for lane-level accuracy using GNSS devices and 3-D shadow matching. The suggested framework is based on detection and analysis of rapid changes in navigation satellites' signal strength, which are caused by momentary blockages due to utility and light poles. A method for detecting such momentary changes between line of sight and non line of sight is presented, followed by a geometric algorithm that improves location accuracy of commercial GNSS devices. We have tested the framework's applicability using both simulations and field experiments. We provide the results of these tests and discuss receiver-side sampling rate requirements for high-performance lane-level positioning.

GNSS Accuracy Improvement Using Rapid Shadow Transitions

A Robust Shadow Matching Algorithm for GNSS Positioning: Particle Filter for Shadow Matching

Commercial GNSS devices tend to perform poorly in urban canyon environments. The buildings' density and height in such environments block the signals from many of the satellites. In this work, we present a novel yet simple approach for tackling this accuracy problem. The novelty of the suggested method lies in analyzing insufficient LOS to satellites and improving the estimated location using few shadow matching methods. Given an estimated location of the GNSS device and given the satellite signal-strength, the algorithm calculates in real-time invalid points inside the ROI (Region Of Interest). This approach reduces the ROI to a fraction of its original size. Using both simulation and field experiment we have shown that the suggested method can improve existing GNSS devices accuracy in urban canyons. Moreover, in urban regions with accurate 3D building map the positioning accuracy might break the 1 meter accuracy bound.

Roi Yozevitch

Papers

A Robust GNSS LOS/NLOS Signal Classifier

A Robust Shadow Matching Algorithm for GNSS Positioning

GNSS Accuracy Improvement Using Rapid Shadow Transitions

A Robust Shadow Matching Algorithm for GNSS Positioning: Particle Filter for Shadow Matching

Breaking the 1 meter accuracy bound in commercial GNSS devices