Ultra-Wideband Antenna-Induced Error Prediction Using Deep Learning on Channel Response Data

Tiemann, Janis; Pillmann, Johannes; Wietfeld, Christian

doi:10.1109/vtcspring.2017.8108571

Cited by 23 publications

(24 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The authors of [16] focused on correcting ranging errors caused by different antenna orientations. To this end, they propose a DNN based approach for ranging error prediction.…”

Section: A Uwb Ranging Error Correctionmentioning

confidence: 99%

“…Machine learning approach Environment Dataset error Open-source data Open-source model [16] Ranging correction with deep neural network LOS conditions (indoor and outdoor)…”

Section: Papermentioning

confidence: 99%

“…Clearly the model is able to reconstruct the (denoised) signal with reduced dimensionality at layer h. This indicates that h contains useful features that can describe the signal. Table 2 gives an overview of the proposed autoencoder models and compares it with the DNN architecture proposed by [16]. Additionally a CNN, part of the autoencoder-based models, serves as a benchmark for SOTA deep learning solutions.…”

Section: B a Dual Loss Semi-supervised Autoencodermentioning

confidence: 99%

“…In this section we compare baseline models, which include the DNN proposed in [16], our CNN (part of the autoen- coder), the pre-trained stacked autoencoder and the dualloss autoencoder predictor. Their performance is expressed in MAE, while their complexity is denoted in parameter amount.…”

Section: A Semi-supervised Learning: Accuracy Complexity and Generamentioning

confidence: 99%

See 3 more Smart Citations

Edge Inference for UWB Ranging Error Correction Using Autoencoders

et al. 2020

View full text Add to dashboard Cite

Indoor localization knows many applications, such as industry 4.0, warehouses, healthcare, drones, etc., where high accuracy becomes more critical than ever. Recent advances in ultra-wideband localization systems allow high accuracies for multiple active users in line-of-sight environments, while they still introduce errors above 300 mm in non-line-of-sight environments due to multi-path effects. Current work tries to improve the localization accuracy of ultra-wideband through offline error correction approaches using popular machine learning techniques. However, these techniques are still limited to simple environments with few multi-path effects and focus on offline correction. With the upcoming demand for high accuracy and low latency indoor localization systems, there is a need to deploy (online) efficient error correction techniques with fast response times in dynamic and complex environments. To address this, we propose (i) a novel semi-supervised autoencoder-based machine learning approach for improving ranging accuracy of ultra-wideband localization beyond the limitations of current improvements while aiming for performance improvements and a small memory footprint and (ii) an edge inference architecture for online UWB ranging error correction. As such, this paper allows the design of accurate localization systems by using machine learning for low-cost edge devices. Compared to a deep neural network (as state-of-the-art, with a baseline error of 75 mm) the proposed autoencoder achieves a 29% higher accuracy. The proposed approach leverages robust and accurate ultra-wideband localization, which reduces the errors from 214 mm without correction to 58 mm with correction. Validation of edge inference using the proposed autoencoder on a NVIDIA Jetson Nano demonstrates significant uplink bandwidth savings and allows up to 20 rapidly ranging anchors per edge GPU.

show abstract

“…The authors of [16] focused on correcting ranging errors caused by different antenna orientations. To this end, they propose a DNN based approach for ranging error prediction.…”

Section: A Uwb Ranging Error Correctionmentioning

confidence: 99%

“…Machine learning approach Environment Dataset error Open-source data Open-source model [16] Ranging correction with deep neural network LOS conditions (indoor and outdoor)…”

Section: Papermentioning

confidence: 99%

Section: B a Dual Loss Semi-supervised Autoencodermentioning

confidence: 99%

Section: A Semi-supervised Learning: Accuracy Complexity and Generamentioning

confidence: 99%

See 2 more Smart Citations

Edge Inference for UWB Ranging Error Correction Using Autoencoders

et al. 2020

View full text Add to dashboard Cite

show abstract

“…In order to meet this explosive demand, various indoor positioning approaches have recently been developed, including Wi-Fi [3,4], UWB [5], BLE [6] and visual methods [7]. However, these methods are costly and ineffective when the radio signal is weak or not available in some scenarios, such as underground parking lots.…”

Section: Introductionmentioning

confidence: 99%

An Infrastructure-Free Indoor Localization Algorithm for Smartphones

Wang

Luo

Men

et al. 2018

Sensors

View full text Add to dashboard Cite

Accurate indoor positioning technology provides location-based service for a variety of applications. However, most existing indoor localization approaches (e.g., Wi-Fi and Bluetooth-based methods) rely heavily on positioning infrastructure, which prevents their large-scale deployment and limits the range at which they are applicable. Here, we proposed an infrastructure-free indoor positioning and tracking approach, termed LiMag, which used ubiquitous magnetic field and ambient lights (e.g., fluorescent, incandescent, and light-emitting diodes (LEDs)) without containing modulated information. We conducted an in-depth study on both the advantages and the challenges in leveraging magnetic field and ambient light intensity for indoor localization. Based on the insights from this study, we established a hybrid observation model that took full advantage of both the magnetic field and ambient light signals. To address the low discernibility of the hybrid observation model, LiMag first generated a single-step fingerprint model by vectorizing consecutive hybrid observations within each step. In order to accurately track users, a lightweight single-step tracking algorithm based on the single-step fingerprints and the particle filter framework was designed. LiMag leveraged the walking information of users and several single-step fingerprints to generate long trajectory fingerprints that exhibited much higher location differentiation ability than the single-step fingerprint. To accelerate particle convergence and eliminate the accumulative error of single-step tracking algorithm, a long trajectory calibration scheme based on long trajectory fingerprints was also introduced. An undirected weighted graph model was constructed to decrease the computational overhead resulting from this long trajectory matching. In addition to typical indoor scenarios including offices, shopping malls and parking lots, we also conducted experiments in more challenging scenarios, including large open-plan areas as well as environments characterized by strong sunlight. Our proposed algorithm achieved a 75th percentile localization accuracy of 1.8 m and 2.2 m, respectively, in the office and shopping mall tested. In conclusion, our LiMag algorithm provided location-based service of infrastructure-free with significantly improved localization accuracy and coverage, as well as satisfactory robustness inside complex indoor environments.

show abstract

UWB anchor nodes self-calibration in NLOS conditions: a machine learning and adaptive PHY error correction approach

et al. 2021

View full text Add to dashboard Cite

Ultra-wideband (UWB) positioning performance is highly related to the accuracy of the coordinates of the fixed anchor nodes, which form the system infrastructure. The process of determining the position of the anchors is called calibration. In an anchor-based system, it is crucial for the fixed nodes to know their locations with the highest possible accuracy. However, in certain situations, it is almost impossible to perform the calibration manually, e.g., during emergency interventions. Moreover, calibration is always delicate and time-consuming. We designed an effortless and accurate selfcalibration algorithm that does not require any manual intervention to precisely pinpoint the position of the anchors. This paper presents an innovative algorithm that combines machine learning and exploits the time resolution capabilities of UWB with adaptive physical settings to enable the automatic calibration of the fixed anchor nodes, even in realistic NLOS (non-line-of-sight) conditions. The self-calibration algorithm combines iterative gradient descent to pinpoint the positions of the anchors and uses error detection and correction from a convolutional neural network. Moreover, the algorithm can use a different set of settings for each anchor pair. This is done to ensure the most robust and accurate communication between nodes. Extensive measurements were carried out to allow anchors to estimate distances among each others. Distances were then combined and processed by the self-calibration algorithm. Experimental evaluation in two complex and large environments with many obstacles and reflections shows that accuracy reached by the algorithm is about 2.4 cm on average and 95th percentile is 5.7 cm, in best case. The results refer to the relative positions among the anchors. Results prove that in order to precisely calibrate the anchors nodes in an UWB positioning system, high correctness can be obtained by combining the accuracy of UWB together with deep learning and adaptive PHY modulation schemes.

show abstract

Ultra-Wideband Antenna-Induced Error Prediction Using Deep Learning on Channel Response Data

Cited by 23 publications

References 4 publications

Edge Inference for UWB Ranging Error Correction Using Autoencoders

Edge Inference for UWB Ranging Error Correction Using Autoencoders

An Infrastructure-Free Indoor Localization Algorithm for Smartphones

UWB anchor nodes self-calibration in NLOS conditions: a machine learning and adaptive PHY error correction approach

Contact Info

Product

Resources

About