During the course of criminal investigations, law enforcement agencies use many technologies in order to find out who committed the crime and where a suspect was when the crime was committed. These agencies may use whatever methods they wish in order to catch the person, but in order to have that evidence heard by a jury it must be scrutinized and scientifically reliable. Depending on the jurisdiction, the technology must pass either the Daubert or Frye test (Pennsylvania uses Frye), both of which are tests that examine the validity and reliability of a scientific technique. Over the years, various technologies have come and gone, such as voiceprint identification and comparative bullet-lead analysis. One technology that is going the way of these two other pseudosciences is a cell phone tracking technique called “granulization”.
Granulization is the process of examining which cell tower a person’s phone was connected to in order to determine a his or her location when a call was made. The FBI and other agencies use this method routinely. Although granulization may seem legitimate on its face (perhaps simply because it’s used by the FBI), it is fraught with error and uncertainty. While the issues with granulization are numerous, three main problems ensure that its reliability in court will not last long. To find out more, we spoke to Michael Cherry, president of Cherry Biometrics and renowned defense expert in the field of cell phone tracking.
The accuracy of granulization is predicated upon the false assumption that every cell phone automatically connects to the cell tower nearest to it for every call. This, however, is not so. In fact, cell companies use a concept called “load balancing” to determine exactly which cell tower a phone connects to. In order to balance the cell traffic efficiently across multiple towers, a phone will “ping” several towers before the call is connected and will determine which tower to connect to, based on the current load on that tower and other towers. Because of load balancing and the fact that exactly how these cell networks run are trade secrets, it is impossible to say, “Because John connected to tower X, he therefore must be closest to tower X.” Other factors including weather and topography influence which tower a phone will connect to, but load balancing is the most determinate factor.
Even with the knowledge that a phone connected to a particular tower, that knowledge is practically useless. Most cell towers have a range of approximately 21 miles, creating a massive area that the phone could potentially be in. Applying the concept of load balancing to this fact, a person could theoretically travel 42 miles from one edge of the tower’s coverage area to the other, passing by many more closer towers, all while being connected to the same tower.
A final problem that can’t be argued in court is the simple fact that the FBI experts are, in the words of Mr. Cherry, “better-trained presenters” than local law enforcement officials and “great sellers”. Even when discussing easily disprovable pseudoscience, juries are more likely to believe FBI agents or employees than civilian experts.
Eradicating granulization from courtrooms is a slow process, but it is making progress. Last September, there was a major ruling on this issue from the United States District Court in the Northern District of Illinois – Eastern Division. Judge Joan H. Lefkow ruled that granulization does not pass the Daubert test, and as such, was not admissible in the kidnapping case United States v. Evans. This ruling was a radical departure from how the issue was handled previously. Whereas granulization used to be blindly accepted as reliable, the ruling in Evans calls into question how long this technique will be admissible in other courts.