Thursday, August 25, 2016

SIGCOMM 2016 - Session 8 - Wireless - Paper 3: "Braidio: An Integrated Active-Passive Radio for Mobile Devices with Asymmetric Energy Budgets".

Presenter: Pan Hu
Authors: Pan Hu, Pengyu Zhang, Mohammad Rostami, Deepak Ganesan

Summary:
Battery capacities of mobile devices vary by up to three orders of magnitude between laptops and

wearables. When such asymmetric battery lifetime meets the usual symmetric power consumption in wireless radios, the lifetime of constrained portable devices is significantly limited.

The driving question of this paper is that, can we design a power proportional radio to consume power proportional to battery size? Other than the classic active wireless radio which adopts a symmetric architecture at both transmitter and receiver sides, passive radio such as backscatter adopts asymmetric architectures to achieve a low power consumption either at the transmitter side or the receiver side.

This paper presents Braidio, a radically new radio design that is capable of dynamic carrier offload i.e. the ability to dynamically switch the transmission carrier between the transmitter and receiver.
In particular, Braidio adopts a passive receiver with SAW filter, and leverages antenna diversity for better reception. The first component suffers from a reduce sensitivity while the second suffers from a reduced robustness, both of which lead to a reduced reception range. To cope with this, Braidio uses active radio as a safety net, and automatically switch between different modes, i.e., active, backscatter and passive, to perform carrier offloading.

A prototype of Braidio is implemented, and experiment results show that Braidio can support transmitter–receiver power ratios between 1:2546 to 3546:1 and enables a huge dynamic range of asymmetry to suit a wide range of energy budgets between end points, and in the meantime achieve a reasonable maximal communication range, e.g., 3-6 meters.


Q&A
Q1. What is the performance of Braidio on bitrate under different modes?
A: We fix a constant bitrate in the experiment.


No comments:

Post a Comment