SDR-based Tool for Fine Indoor Localization presented at INFOCOM 2021

The demonstration for the article “FIND: an SDR-based Tool for Fine Indoor Localization” by Evgeny Khorov, Alexey Kureev, and Vladislav Molodtsov has been presented at the IEEE INFOCOM 2021, a top-ranked conference on networking in the research community. The speaker was Vladislav Molodtsov, a 4th-year student of the DREC MIPT and a WNL member.

In the modern world, it is often necessary to locate devices indoors, where GPS is not applicable. An alternative approach is to utilize Wi-Fi access points, which are ubiquitous indoors. Most access points have an antenna array that can be used to estimate the direction of arrival from the client. Thus, multiple access points can locate devices indoors. Despite several promising techniques including machine learning methods, there is no exact answer as to what the most accurate direction of arrival estimation algorithm is. The problem is that the existing algorithms have not been compared with each other in the same scenarios, since some of them operate with a signal in the time domain, while others work in the frequency domain with the so-called CSI (Channel State Information). At the same time, there are no available datasets or even devices, which can provide both kinds of that information. In this article, the WNL scientists demonstrate a testbed that solves this problem.

The testbed consists of a 4-channel Wi-Fi receiver, a rotation platform with an antenna array, and a Wi-Fi transmitter. The main advantage of the testbed is the devised prototype of the Wi-Fi receiver, which extracts from real Wi-Fi frames all necessary information for various direction of arrival estimation algorithms. The prototype is based on NI USRP. An array of antennas is connected to the receiver and placed on the rotation platform, which allows the array to be automatically set at a certain angle relative to the transmitter. The authors also proposed a new calibration method eliminating the constant phase shift between the receiver channels. As a result of the study, a large-scale comparison of various methods for the direction of arrival estimation has been carried out. Moreover, the authors have published a dataset gathered using the developed testbed for further research of the direction of arrival estimation algorithms. The dataset contains channel data extracted from 300.000 Wi-Fi frames in a classroom and 80.000 Wi-Fi frames in an anechoic chamber.

IEEE International Conference on Computer Communications (IEEE INFOCOM) is a top-ranked conference on networking in the research community. It is a major conference venue for researchers to present and exchange significant and innovative contributions and ideas in the field of networking and closely related areas. IEEE INFOCOM covers both theoretical and systems research. For INFOCOM 2021, the conference includes a main technical program, a number of workshops, a keynote speech, panels, a student poster session, and demo/poster sessions. Due to the COVID-19 pandemic, the INFOCOM conference, including the demo session, went online. Therefore, the WNL team prepared a video for an online presentation of the work accomplished.

Wireless Networks Lab is a ‘Megagrant’ lab established in 2017 around the project on Cloudified Wireless Networks for 5G and beyond, led by Prof. Ian F. Akyildiz. The team regularly reports at leading IEEE conferences, runs industrial projects, and contributes to the standardization of wireless networks.