The notion of mobile sensing here refers to leveraging wireless signals to mine and infer valuable, real-time contextual information about the user and environment. This can be accomplished with everyday devices (e.g. smartphones) as well as autonomous systems (e.g. terrestrial/aerial robots) that are growing in popularity and maturity, to enable a host of smart, ubiquitous mobile services in challenging environments both indoors and outdoors.

Mobile Autonomous Systems: The growing maturity of mobile autonomous systems (e.g. terrestrial and aerial robots) coupled with the versatility of RF sensing, allows us to deploy these "mobile autonomous sensing" systems in challenging environments and enable several time-critical, on-demand mobile services. We are particularly interested in deploying autonomous aerial vehicles with RF sensing capabilities to aid in first responder situations both indoors and outdoors for services such as high bandwidth, low-latency communications, user tracking, real-time video surveillance and analytics, etc.

• SkyLiTE: One of the first efforts to design and deploy an on-demand, un-tethered, multi-cell LTE network (on UAVs) that can self-configure itself in the sky. SkyLiTE consists of three main components - SkyRAN, SkyCore and SkyHaul that re-architect the various components (RAN, core and backhaul transport respectively) of a wide-area cellular network to make it deployable on challenging UAV platforms in highly dynamic environments.

• TrackIO: An infrastructure-free solution that enables accurate localization and tracking of first responders in hostile indoor environments. TrackIO is a first-of-a-kind real-time system that leverages ultra-wideband (UWB) and small form factor UAV(s), along with algorithmic innovations, to enable accurate (1-2m) tracking of responders even when they are mobile and deep inside buildings without relying on the building's wireless infrastructure.

Smart Spaces: Smart spaces is the notion of turning everyday objects in indoor environments into smart objects that can interact with users and track/infer their gestures. With the growing number of smart devices and a multitude of interfaces/remotes/apps to control these devices, such spaces will allow for seamless, frustration-free integration of smart devices into our everyday lives.

• xSHIFT: A novel, completely passive (battery-less) RF tag that works with commodity WiFi devices. xSHIFT aims to bring the benefits of passive RFID tags to consumer spaces by designing novel RF tags that can operate with our smartphones, WiFi access points, and other smart devices (e.g. smart speakers), without the need for a dedicated RFID-like reader infrastructure. This can open the door to a myriad of physical analytics applications in consumer spaces.

• RIO: A a novel battery-free touch sensing user interface (UI) primitive for future IoT and smart spaces. With RIO, any surface can be turned into a touch-aware surface by simply attaching RFID tags to them. RIO enables UIs to be constructed using off-the-shelf RFID readers and tags, and provides a unique approach to designing smart IoT spaces.

• Konark: A smart shopping cart that leverages RFID-based sensing and machine learning to enhance in-store shopping experience for users. It does so by tracking the (RFID-tagged) products interacted by the user as well as those placed in the shopping cart by the user in real-time without user engagement. This enables a whole host of smart retail services such as seamless checkout, user interest detection, real-time pricing, dynamic coupon delivery, etc.

Smart Infrastructure: Several indoor applications of IoT such as smart retail, connected healthcare, surveillance, etc., will greatly benefit from two primitives: (a) ability to accurately localize devices (eg. sensors and wearables) indoors; and (b) a robust, on-demand infrastructure for plug-n-play connectivity for IoT devices (eg. security cameras, automation devices). We have designed solutions that transform legacy infrastructure into smart infrastructure to realize these two primitives.

• ToneTrack: A system employing WiFi APs through a distributed antenna system coupled with intelligent signal processing techniques to yield a high resolution of sub-meter accuracy in indoor localization of RF capable devices.

• BOLT: A smart infrastructure system combining the pervasive connectivity of powerline communication (PLC) networks with machine learning analytics to enable rapid mobilization of indoor IoT applications like video surveillance.