A Structured Deep Neural Network for Data Driven Localization in High Frequency Wireless Networks

Comiter, Marcus; Crouse, Michael B.; Kung, H. T.

doi:10.5121/ijcnc.2017.9302

Cited by 6 publications

(4 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the processed signals, the available information, and the system setup in massive MIMO channel fingerprinting are significantly different. Comiter et al [ 68 ] propose a beam estimation for using deep neural networks that derives the angle of arrival by phase differences. Using different antenna arrays a structured pair of neural networks is used to estimate the antenna beam.…”

Section: Related Workmentioning

confidence: 99%

A Deep Learning Approach to Position Estimation from Channel Impulse Responses

Niitsoo

Edelhäußer

Eberlein

et al. 2019

Sensors

View full text Add to dashboard Cite

Radio-based locating systems allow for a robust and continuous tracking in industrial environments and are a key enabler for the digitalization of processes in many areas such as production, manufacturing, and warehouse management. Time difference of arrival (TDoA) systems estimate the time-of-flight (ToF) of radio burst signals with a set of synchronized antennas from which they trilaterate accurate position estimates of mobile tags. However, in industrial environments where multipath propagation is predominant it is difficult to extract the correct ToF of the signal. This article shows how deep learning (DL) can be used to estimate the position of mobile objects directly from the raw channel impulse responses (CIR) extracted at the receivers. Our experiments show that our DL-based position estimation not only works well under harsh multipath propagation but also outperforms state-of-the-art approaches in line-of-sight situations.

show abstract

Section: Related Workmentioning

confidence: 99%

A Deep Learning Approach to Position Estimation from Channel Impulse Responses

Niitsoo

Edelhäußer

Eberlein

et al. 2019

Sensors

View full text Add to dashboard Cite

show abstract

“…A slightly different approach is taken in [52], where the authors adapt the structure of a deep neural network so that it can take as input the phase differences of a sub-6 GHz signal measured at the elements of an antenna array and relate them to a user location, even with a few tens of training samples. The accuracy is sufficient to point a directive beam accurately.…”

Section: The Importance Of Location Informationmentioning

confidence: 99%

Scaling Millimeter-Wave Networks to Dense Deployments and Dynamic Environments

et al. 2019

View full text Add to dashboard Cite

Millimeter wave (mmWave) communications have emerged as one of the most promising options to vastly increase wireless data rates due to the high bandwidth they offer. Given the high path loss at mmWave frequencies, such systems require directional antennas to achieve a good communication range. The communicating devices thus need to align the beam directions of their mmWave antennas. Due to the high penetration loss, the paths between the antennas also need to be free of blocking obstacles. This makes efficient and reliable operation of mmWave networks in dynamic environments very challenging. At the same time, the directionality reduces interference and allows to scale these networks to much higher access point and device densities. In this article, we discuss the above challenges and present techniques that allow mmWave networks to scale to high-density deployments, to adapt to dynamic and mobile environments, and to consistently achieve high data rates. This includes learning the environment to find different propagation paths, reacting timely to channel impairments such as blockage, and integrating mmWave networks with networks operating at lower-frequency for robustness. A key ingredient to enable these forms of adaptivity is the use of location information. Such mechanisms then turn a collection of very-high-speed but brittle mmWave links into an efficient, low-latency, and reliable network.

show abstract

“…In general, the best accuracy and lowest localization errors are enabled by range-based algorithms [10], [29]; however, ranging requires accurate path loss models, whose parameters are environmentspecific and may have to be re-trained over time. Multipath propagation can be exploited along with AoA information in order to localize a client [30] even with a single AP [31], although mmWave beamforming upon link establishment may turn out to sparsify the channel and reduce the number of useful multipath components [8]. Given sufficiently many reflected multi-path components, a mmWave location system can also be used to map the environment [32].…”

Section: Related Workmentioning

confidence: 99%

LEAP: Location Estimation and Predictive Handover with Consumer-Grade mmWave Devices

Palacios

Casari

Assasa

et al. 2019

IEEE INFOCOM 2019 - IEEE Conference on Computer Communications

View full text Add to dashboard Cite

Future millimeter-wave networks will support very high densities of devices and access points. This vastly increases the overhead required for access point selection and beam training. Fortunately, the quasi-optical properties of millimeterwave channels make location-based network optimization a highly promising technique to reduce control overhead in such millimeter-wave WLANs. In this paper, we extract channel state information from off-the-shelf routers, we use it to design a high accuracy location system, and then show how location information enables the optimization of network operations. The resulting scheme, named LEAP, can predict blockage, optimize access point association, and select the most suitable antenna beam patterns while significantly reducing the beam training overhead. We show that compared to standard state-of-the-art 802.11ad systems, LEAP's location driven management greatly improves network performance and link stability.

show abstract

A Structured Deep Neural Network for Data Driven Localization in High Frequency Wireless Networks

Abstract: Next-generation wireless networks such as 5G and 802.11ad

Cited by 6 publications

References 11 publications

A Deep Learning Approach to Position Estimation from Channel Impulse Responses

A Deep Learning Approach to Position Estimation from Channel Impulse Responses

Scaling Millimeter-Wave Networks to Dense Deployments and Dynamic Environments

LEAP: Location Estimation and Predictive Handover with Consumer-Grade mmWave Devices

Contact Info

Product

Resources

About