VoIP Surprises Seen in Informal Test of 911 Background Sounds
A myth about VoIP’s ability to handle 911 traffic may have been blown away by gunfire at the Ga. Public Safety Training Center. A Monroe County, Ga., test this month for the Assn. of Public Safety Communications Officials International (APCO) indicated to participants’ ears that -- contrary to widespread belief -- VoIP calls can transmit background sounds such as gun shots from callers’ premises well enough to give responders crucial clues about emergencies and how best to address them, CEO James Cavanagh of the Consultant Registry told us. He’s an APCO and National Emergency Numbers Assn. member who prompted the experiment. Cavanagh acknowledged the testing was preliminary and purposely unscientific, but added: “The gauntlet’s been thrown down. Initial results indicate it was not as bad a problem as we thought it was.”
Sign up for a free preview to unlock the rest of this article
Communications Daily is required reading for senior executives at top telecom corporations, law firms, lobbying organizations, associations and government agencies (including the FCC). Join them today!
Cavanagh and many others in the field had presumed that the way VoIP software optimizes transmission of call participants’ voices to conserve bandwidth would have the effect of squeezing out background noise and whispers, to 911’s detriment. The experiment came about when a questioner at a conference asked him the basis for the assertion. Cavanagh realized he hadn’t heard this for himself and wasn’t aware of any research on it, he said. Research turned up no previous tests, Cavanagh said.
VoIP can transmit background sounds about as well as cellphones, listeners judging the test calls found, Cavanagh said. That makes sense because the 2 technologies “have access to the same set of wireless codecs,” Cavanagh said. The quality of transmission of the sounds isn’t as good as on landline -- but it’s a lot better than nothing, and special training dispatchers receive to handle mobile calls should be adaptable to VoIP, Cavanagh said. The initial conclusion is that “they would be able to recognize the sounds that would be meaningful to them,” though they would have to be more tentative and do more questioning to confirm their impressions. The other main conclusion was that the packet nature of VoIP means that when words or syllables drop out of a call, it could be less apparent and therefore more confusing to the recipient than with other calls, he said.
The results are to be presented Feb. 1 at APCO’s Winter Summit in Orlando. Sound samples will be offered through a digital PowerPoint presentation to let those attending judge for themselves, Cavanagh said. The test only began to address one of issues involved in getting more than a few VoIP providers to offer emergency service. Large remaining hurdles involve the mobility of VoIP adapters and numbers, and sector resistance to imposition of 911 regulations and costs.
Cavanagh judged the sound with Cheryl Greathouse, instructional services dir. of the Ga. training center, and the mgr. of the Monroe County Public Safety Answering Point (PSAP). Simultaneous 10-digit emergency calls were placed over a cellphone, landline phone and VoIP line to the PSAP from the training center nearby, Cavanagh said. Some VoIP services offer the 10 digit calls in lieu of 911 service. The caller answered dispatcher questions in full voice and then in a whisper, Cavanagh said. A residential fire alarm was sounded. A gun in the training center was cocked and then fired, both at varying distances from the phones.
The brief, subjective test wasn’t devised to be scientific, Cavanagh said. He said he chose realism to impress emergency officials -- “real world data to address real world questions” -- over the artifice of lab conditions needed to try to generate scientifically reproducible results. But he said he hoped the work would spur such rigorous research, including to determine how representative of all VoIP the service used in the test was.