In this paper, we propose FreeMap, an approach that uses RF signals to generate accurate indoor maps of the line of sight as well as the non line of sight parts of buildings. To generate the maps, FreeMap leverages the movements of occupants in buildings. When an RF signal reflects from an occupant, the reflected signal arrives back at the receiver both directly and as multipaths, i.e., the copies of the signal that reflect from further walls before arriving back at the receiver. These direct and multipath reflections form a triangular geometry that enables FreeMap to locate various parts of the walls in the environment. As time progresses and the occupants move around, FreeMap localizes more and more parts of all the walls. It eventually obtains the locations of enough parts of all the walls that it can then connect them together to generate the indoor map. We have implemented FreeMap using a commercial FMCW radar (which provides in-home monitoring services to elderly) and have extensively evaluated it at eight measurement sites. Our results show that FreeMap generated maps with average precision, recall, and positioning error of $\mathbf{9 7 . 3 \%}, \mathbf{9 5 . 5 \%}$, and $20.3 \mathrm{~cm}$, respectively.

Generating maps of indoor environments beyond the line-of-sight finds applications in several areas such as planning, navigation, and security. While researchers have previously explored the use of RF signals to generate maps, prior work has two important limitations: (i) it requires moving the mapping setup along the entire lengths of the sides of the building, and (ii) it generates maps that are not fully connected, rather are scatter plots of locations from where some obstacles reflected the signals. Thus, prior approaches require human interpretation to locate the walls and determine how they merge. In this article, we address these limitations and propose RFMap, which generates fully connected maps, and does not require the measurement setup to be moved along the sides of the buildings. To generate the map, RFMap first transmits RF signals in many different directions and then measures the distances of different reflectors inside the building. Next, it identifies these reflectors and classifies them into various types based on the properties of the reflections. A key challenge is that RFMap does not receive reflections from all the directions due to the specular nature of the reflectors. Due to this, it only gets sparse data about the objects in the environment. To address this challenge, RFMap trains a deep generative adversarial network (GAN) to intelligently predict the missing information. At runtime, it feeds the locations and types of the detected reflectors to the trained GAN and generates complete and accurate map. We implemented RFMap using software defined radios and extensively evaluated it in several real-world environments. Our results show that RFMap generated the maps of all the buildings that we tested it on with high accuracy.

In this paper, we propose CoMEt, a radio frequency based approach that measures soil moisture at multiple depths underneath the ground surface without installing any objects in the soil and without making any contact with the ground surface. The main insight behind CoMEt is that the phase of an RF signal depends on its wavelength in the medium through which it is propagating, which in turn depends on the amount of soil moisture. To measure soil moisture, CoMEt leverages the phase changes across successive antennas in a receive antenna array along with the time of flight of the received signal to jointly estimate the depth of each layer of soil and the wavelength of the signal in each layer. It then uses these estimates to obtain the amount of moisture in each soil layer. We have implemented CoMEt using a software defined radio and a Raspberry Pi to measure soil moisture in real-time. We have extensively evaluated CoMEt in both indoor and outdoor environments. Our results show that CoMEt estimated soil moisture for up to three layers of soil with a median error of just 1.1%.

Email services use spam filtering algorithms (SFAs) to filter emails that are unwanted by the user. However, at times, the emails perceived by an SFA as unwanted may be important to the user. Such incorrect decisions can have significant implications if SFAs treat emails of user interest as spam on a large scale. This is particularly important during national elections. To study whether the SFAs of popular email services have any biases in treating the campaign emails, we conducted a large-scale study of the campaign emails of the US elections 2020 by subscribing to a large number of Presidential, Senate, and House candidates using over a hundred email accounts on Gmail, Outlook, and Yahoo. We analyzed the biases in the SFAs towards the left and the right candidates and further studied the impact of the interactions (such as reading or marking emails as spam) of email recipients on these biases. We observed that the SFAs of different email services indeed exhibit biases towards different political affiliations.

Generating maps of indoor environments beyond the line of sight finds applications in several areas such as planning, navigation, and security. While researchers have previously explored the use of RF signals to generate maps, prior work has two important limitations: (i) it requires moving the mapping setup along the entire lengths of the sides of the building, and (ii) it generates maps that are not fully connected, rather are scatter plots of locations from where some obstacles reflected the signals. Thus, prior approaches require human interpretation to locate the walls and determine how they merge. In this paper, we address these limitations and propose RFMap, which generates fully connected maps, and does not require the measurement setup to be moved along the sides of the buildings. To generate the map, RFMap first transmits RF signals in many different directions by rotating the antennas while keeping them at the same location and then measures the distances of different reflectors inside the building. Next, it identifies these reflectors and classifies them into various types based on the properties of the reflections. A key challenge is that RFMap does not receive reflections from all the directions due to the specular nature of the reflectors. Due to this, it only gets sparse data about the objects in the environment. To address this challenge, RFMap trains deep generative adversarial network (GAN) to intelligently predict the missing information. At runtime, it feeds the locations and types of the detected reflectors to the trained GAN and generates the complete and accurate map. We implemented RFMap using software defined radios and extensively evaluated it in several real world environments. Our results show that RFMap generated the maps of all the buildings that we tested it on with high accuracy.

This paper presents an end-to-end deep learning framework using passive WiFi sensing to classify and estimate human respiration activity. A passive radar test-bed is used with two channels where the first channel provides the reference WiFi signal, whereas the other channel provides a surveillance signal that contains reflections from the human target. Adaptive filtering is performed to make the surveillance signal source-data invariant by eliminating the echoes of the direct transmitted signal. We propose a novel convolutional neural network to classify the complex time series data and determine if it corresponds to a breathing activity, followed by a random forest estimator to determine breathing rate. We collect an extensive dataset to train the learning models and develop reference benchmarks for the later studies in the field. Based on the results, we conclude that deep learning techniques coupled with passive radars offer great potential for end-to-end human activity recognition.

This paper presents a two-dimensional phase extraction system using passive WiFi sensing to monitor three basic elderly care activities including breathing rate, essential tremor and falls. Specifically, a WiFi signal is acquired through two channels where the first channel is the reference one, whereas the other signal is acquired by a passive receiver after reflection from the human target. Using signal processing of cross-ambiguity function, various features in the signal are extracted. The entire implementations are performed using software defined radios having directional antennas. We report the accuracy of our system in different conditions and environments and show that breathing rate can be measured with an accuracy of 87% in the absence of obstacles. We also show a 98% accuracy in detecting falls and 93% accuracy in classifying tremor. The results indicate that passive WiFi systems show great promise in replacing typical invasive health devices as standard tools for health care.