Dhairya J. Shah

Wireless connectivity challenges hinder the deployment of embedded systems. We introduce AudioCast to address two critical issues: spectrum scarcity-induced contention and high power consumption in transmitters. The widespread availability of broadcast radio receivers (for example, FM radios using the 88-108 MHz spectrum) and access to underutilized lower-frequency spectrum motivate the design of AudioCast. The lower-frequency spectrum offers superior radio-wave propagation characteristics, exhibiting at least 10x lower path loss than the 2.4 GHz and 5 GHz Industrial, Scientific, and Medical (ISM) bands while avoiding congestion and interference. These properties enable reliable and long-distance communication, even for weakly radiated signals. AudioCast builds on these properties and the unique negative resistance of a tunnel diode.AudioCast rethinks the architecture of radio transmitters using a tunnel diode oscillator to generate carrier signals and self-modulate them with baseband signals. This results in frequency-modulated transmissions at an overall power consumption below 200 μW. Unlike related systems based on the backscatter mechanism, AudioCast does not require an externally generated carrier or rely on ambient signals. We argue that AudioCast represents an example of a new class of transmitters which we conceptualize as Beyond-Backscatter transmitters. Through experiments, we demonstrate that AudioCast achieves a transmission range of up to 130 m in line-of-sight and tens of meters in non-line-of-sight conditions respectively. These transmissions are decodable by ubiquitous commodity FM receivers in cars, homes, and phones. We evaluate AudioCast through theoretical analysis, benchtop experiments, and urban/indoor field deployments. Additionally, we prototype and demonstrate novel applications, including low-power voice transmissions and hand gesture communication, enabled by AudioCast’s range and power efficiency.

We have seen rapid deployment of augmented reality devices. However, these devices currently suffer from limited battery life. They require frequent recharge, especially when capturing and streaming information wirelessly. In particular, these devices stream information over the radio spectrum using transceivers that are power-hungry, and wireless communication dominates the energy budget of these devices. We present our ongoing work to design a system called Li-FiAR, which proposes integrating Li-Fi with augmented reality devices. Specifically, we implement a transmitter and receiver and integrate them with an augmented reality device. We argue that Li-Fi receivers may consume less power than their radio counterparts while benefiting from the spatial nature of light propagation. This enables novel application scenarios for augmented reality. We demonstrate the transmission of images and audio through Li-Fi, presented on an augmented reality device in this work.

Mixed-reality headsets offer new ways to perceive our environment. They employ visible spectrum cameras to capture and display the environment on screens in front of the user’s eyes. However, these cameras lead to limitations. Firstly, they capture only a partial view of the environment. They are positioned to capture whatever is in front of the user, thus creating blind spots during complete immersion and failing to detect events outside the restricted field of view. Secondly, they capture only visible light fields, ignoring other fields like acoustics and radio that are also present in the environment. Finally, these power-hungry cameras rapidly deplete the mixed-reality headset’s battery. We introduce PixelGen to rethink embedded cameras for mixed-reality headsets. PixelGen proposes to decouple cameras from the mixed-reality headset and balance resolution and fidelity to minimize the power consumption. It employs low-resolution, monochrome image sensors and environmental sensors to capture the surroundings around the headset. This approach reduces the system’s communication bandwidth and power consumption. A transformer-based language and image model process this information to overcome resolution trade-offs, thus generating a higher-resolution representation of the environment. We present initial experiments that show PixelGen’s viability.