Tuesday, August 8, 2017

The case for encryption and delay

Lincoln is in the process of acquiring a new public safety radio system. The new system is from Motorola, and is a P25 trunked system, which is inherently digital. Since it is a digital radio system, encryption is easy to implement. It is our intention to use encryption to protect law enforcement-sensitive transmissions. We do not intend to encrypt Lincoln Fire & Rescue radio traffic. If we encrypt all of our law enforcement talk groups (colloquially, "channels"), this will essentially put scanners out of business.

People have been using scanners to listen to police radio transmissions, well, for as long as Lincoln has had police radio transmissions-the 1930's. Prior to the mid-1990's, this hobby generally required some relatively expensive equipment and a little expertise to set up. As a result, the number of people using scanners was rather small, and rarely caused problems for the police. This all changed with the advent of streaming audio on the Internet. By the mid-to-late 1990s, a scanner buff could simply publish the audio from a scanner to a URL, and suddenly anyone with an Internet connection could listen.

With the proliferation of smartphones, this became incredibly easy and popular, and scanner applications proliferated. Publishing the audio from police radio traffic to the Internet became a problem for those of us in law enforcement, because instead of a small group of dedicated hobbyists, the number of people listening to the police radio grew dramatically--including the number of people with ulterior motives: criminals. Part of the problem was that publishers of scanner feeds were often streaming audio that had no business out in public: information channels, tactical channels, investigative channels, and so forth. While the primary police dispatch channels were challenging enough, these private channels carrying even more sensitive traffic were especially problematic.

About five years ago, we decided to publish our own Internet feed of primary dispatch channels. Our motive in doing so was to occupy the market. We hoped that if we put out an official audio feed, the amateurs who were publishing their own feeds--sometimes with no regard for the most sensitive traffic--would fade away, as the official feed became the de facto standard. Most people looking for streaming public safety audio from Lincoln would choose the official feed, and since we controlled that content, the chance of someone inadvertently or intentionally streaming the most sensitive radio traffic would be greatly reduced. It worked; the official feed essentially pushed most of the freelancers out of the game.

Despite this, there is still plenty of radio traffic on the primary dispatch channels that is sensitive, especially early in the life-cycle of an emergent incident. Bad actors can, and have, used this information to evade the police, and to further criminal enterprises. We have many examples of this, but since most of these uses are never discovered, I am sure the practice is much more widespread then we know.

Recently, for example, a burglary occurred at Gateway Mall. There was good video of the burglars, and it was apparent that they were monitoring a smartphone scanner application, which allowed the thieves to skedaddle when the alarm was dispatched. We have documented examples of scanners or scanner applications used by criminals during everything from shoplifting, to kidnapping, robbery, burglary, and murder. While it is probably still a fairly small percentage of criminals who are organized enough and smart enough to use police radio transmissions to their advantage, it is easier than ever, and has become more common than ever.

With our new radio system currently under construction, it will be a simple matter to implement encryption, which would defeat the use of scanners and the rebroadcasting of Internet audio feeds from scanners. The technology is baked into the radio system.

Here's the problem: we also recognize that there is some value in the public being able to listen to our radio transmissions. Citizens may become more informed about emergency events, they may be able to avoid places of congestion or danger arising from those events. Citizens can be more informed about the work of their police department. You can't listen to the scanner for very long without coming away with a much greater appreciation of the sheer volume and complexity of work that we ask our police officers to perform. The drama, tragedy, and terror that fills the air waves helps build greater respect for the work of our officers, and greater support for the police department. It also demonstrates transparency, something we value in a democratic society where we entrust our police officers with immense power and discretion.

How, then, can we minimize the chance that evil-doers will use police radio traffic to their advantage, while preserving those benefits we realize from public monitoring of police radio transmissions? I believe the answer to this conundrum is to encrypt all police radio traffic, but to continue to publish an Internet feed consisting of the de-crypted audio from the primary dispatch channels--the same thing we publish today--but with a short delay to give the police officers a slight advantage over the criminals. Ten minutes is a good period because it matches our response time performance target for crimes in progress. We can be there in 10 minutes, 90% of the time.

With a ten minute delay, the audio is still fresh. In most cases, the incident response is still in progress, and citizens can still get a sense of what's going on. Most of the time, a listener would not even realize that the audio is delayed. The audio feed would still provide unfiltered transparency for police operations, but would protect against the burglar, barricaded suspect, suicidal party, or the individual seeking to ambush a police officer getting the jump by monitoring the police response in real time. It's a good compromise between the competing interests. It will not jeopardize transparency or accountability but it will help protect our police officers, and make it more difficult for criminals to use scanner technology to further their aims.