'Airshark: Detecting Non-WiFi RF Devices Using Commodity WiFi Hardware'

摘要

Strengths: Though overstated repeatedly, the use of typical hardware is a strength. The study of 20 locations and the interfering objects in each is a bit of a contribution as a spot check (not as contribution #1). Weaknesses: The writing is poor, redundant, and compressed at a low level, making it difficult to extract the key scientific contributions: why this work was technically difficult and what novel method emerged to address that problem. The practical implication is not quite apparent: if performance is poor, perhaps the remedy is the same no matter what the issue is. Comments to Authors: Should APs simply change channels whenever performance is poor to find another? Does it matter whether the channel is simply oversubscribed by someone streaming video over 802.11 or is someone talking on a phone? I suspect not. This leaves much of your work an oversell: it's great that you can do this, and I think it may have implications for backoff parameters and rate selection, and maybe even pop-ups for the user, but it doesn't mean much for channel selection.