An issue for IP -- public safety on-scene communication

[David Farber <dave () farber net>]

Begin forwarded message:

From: David Josephson <dlj04 () josephson com>
Date: September 11, 2009 4:10:12 PM EDT
To: David Farber <dave () farber net>
Subject: An issue for IP

Dave,

This is an issue that might interest some of the people on IP. Several  
firefighters have already died as a proximate result of poor decisions  
made by the communications public safety establishment and the FCC,  
and it is slated to get worse rather than better.

Traditionally, public safety on-scene communications have been  
accomplished with FM analog radios that take about 20 kHz of bandwidth  
to operate. In good conditions they achieve intelligibility ratings  
around 90% and people have learned to live with them. The demands for  
data services and trunking protocols (that enable wide-area coverage  
of dozens or hundreds of channels at once through a central  
infrastructure) have grown far more rapidly than has the available  
bandwidth, so the FCC has gone along with proposals from major  
equipment manufacturers to make the existing channels narrower and to  
mandate migration to "narrowband." By 2013, all public and private two- 
way systems in the most commonly used bands will reduce their  
bandwidth by about half. Many have seen this as a mandate for  
transition to digital radios rather than narrowband analog, and the  
industry has adopted a standard for this that uses a lossy speech  
codec for the voice channel. The standard in the USA is called Project  
25 or P25, and Phase I includes a proprietary sole-source multiband  
excitation speech coder, initially developed at MIT, that works well  
in tests despite being given only 4400 bps for the voice channel.  
Speech quality is compromised but intelligibility remains at 85 to  
90%. Unfortunately, in the presence of noise, intelligibility degrades  
severely, and fires and emergency scenes are not often quiet. A  
summary of the tests that have been done, with links to the data, can  
be found at http://www.iafc.org/displaycommon.cfm?an=1&subarticlenbr=719

Phase II of P25 envisions providing only 2400 bps for the speech  
signal using an AMBE codec, also proprietary and sole-source.  
According to the NTIA tests, these codecs fail in the presence of  
noises like fire engine pumps, nozzles and breathing apparatus alarms.

Work is proceeding on cleaning up the audio presented to the codec so  
that it has a chance of being intelligible in the presence of high  
levels of background noise. I am holding a workshop at the Audio  
Engineering Society's 127th convention in New York in October,  
featuring DJ Atkinson of NTIA's Institute for Telecommunication  
Sciences (who did some of the research linked from the IAFC page  
above). We will be focusing on how to improve the microphone  
technology to make the codec's job easier, but we are chipping away at  
only the part of the iceberg that's visible. See http://www.aes.org/events/127 
 for details.

There is a huge push from Homeland Security and other agencies to  
centralize public safety radio systems, which requires lots of new  
bandwidth. Many equipment vendors, under strong pressure from low cost  
foreign suppliers, are very interested in a mandate to use only their  
equipment. All of this is just an economic issue if everything works.  
Unfortunately, in many cases it doesn't work, and emergency responders  
find their radios unusable, and safety-of-life-critical messages get  
lost. The issue needs more public discussion and thinking about other  
compromises that can be made in order to achieve the information  
bandwidth requirements that users want -- without compromising the  
critical usability of the systems.

Hope this is of interest to you and yours...

Regards

David Josephson
Josephson Engineering, Inc.
Audio Engineering Society Technical Council







-------------------------------------------
Archives: https://www.listbox.com/member/archive/247/=now
RSS Feed: https://www.listbox.com/member/archive/rss/247/
Powered by Listbox: http://www.listbox.com