Bluetooth & WiFi Detectors – A valuable data source for traffic engineers

Written by: Michael McGuire , Vice President & General Manager of Stinson ITS

Travel time sensors have been available in the traffic industry for nearly 10 years and have proven themselves to be a very effective technology to monitor congestion and the average speed between two points.?The technology works by detecting either WiFi or Bluetooth devices and tracking them as they travel from sensor A to sensor B; thereby, measuring the time it took to do so.?By doing this for hundreds or thousands of vehicles per hour you can very accurately measure the average travel time between two points.?Detecting devices is possible because WiFi and Bluetooth devices have a unique attribute called a MAC address which is continuously transmitted whenever they are not connected to a device or network.?This is how the music on your phone starts playing when you start your car, your car knows the MAC address of your phone, and recognizes it when it sees it, and automatically accepts the connection and starts communicating. The WiFi & Bluetooth devices in your smartphone & the Bluetooth in your car stereo are the most commonly tracked devices for traffic applications.?

The notion that third party devices can track your smartphones never sounds very comforting, so it’s important to mention that it is critical to ensure privacy protection mechanisms are in place.?The key element of privacy protection in these systems is primarily the anonymization of data at the edge (onboard the sensor), this ensures that there is no way to discover or view actual MAC addresses. Privacy is of top priority for these types of systems. Given these systems track unique identifying numbers it is very important to ensure that these numbers are constantly scrambled and encrypted so that the data cannot be traced back to individual users/drivers. Most systems transform/scramble unique ID's immediately into altered numbers and then use those altered numbers for calculations throughout the day. Typically, once per day the encryption algorithm used to scramble the numbers is changed to a new random value such that no day to day tracking/comparisons of unique ID's is possible.

Stinson AIS

Stinson ITS has been manufacturing their own travel time sensor for the past seven years.?We call it the Stinson AIS and it was one of the first products I designed when joining Stinson in 2015.?This product is the result of a collaboration with one of our oldest partner companies, Transnomis Solutions Inc. ?Their ATMS software called ITS Central is what turns our raw sensor data into useful travel time information.?While our sensor was designed with an open communications protocol so any software can ingest our raw data, ITS Central was our first software partner and has worked with us to develop a rich set of features and reporting tools over the years. ?

At the end of the day a travel time sensor is not that sophisticated, however the AIS has a number of thoughtful features which make it unique within the market:

• Dual Detection Types – The AIS detects both Bluetooth & WiFi devices.?This ensures detection of the maximum number of devices, ie the largest possible dataset.?I’m not sure why sensor manufacturers choose only one or the other, but we both.?The cost difference is negligible so it seemed like common sense.????
• Open Protocol – Travel time sensor manufacturers have always focused on proprietary protocols in order to force you to use their “special” software.?Coming from the traffic industry and the world of NTCIP, this was not the route we wanted to take.?Our device uses an open protocol and as such its detection data can be fed into any third party software platform.?
• Intersection Delay Metrics – While travel time sensors are often focused on highway applications they are also very valuable in arterial signal networks.?Our sensor generates data on intersection delay by monitoring how long each device is in it’s detection area. ?When installed at intersections, this translates to simple vehicle delay.?What we’re working on now is tracking movement specific delay metrics.
• GPS Data Time Sync & Data Tagging – We recently added a GPS module option for our sensor which allows each detection to be geo-tagged with the sensor location as well as to sync the device’s internal clock.?This is especially useful for mobile deployments where the sensors are used for short term studies and keeping track of what data came from where can be tough.
• Built-in LTE – We also offer a built in LTE option for the sensor which allows for very simple and easy system deployments.?Just plug it in and the data starts streaming, making deployment a breeze.?This is useful for permanent and temporary deployments.????
• NTP Support, Passive POE support, etc
• While other sensors have some of those features, I believe we are the only ones with all of them.?We have done our best to make this a robust and feature rich sensor.?However no matter how good your sensor is, you need good software as well to make proper use of the data.?

Central Software

Bluetooth & WiFi data is only as good as the software you pair with it so by de-linking the hardware and software, our clients can choose their preferred vendor. ?There are still only a few software vendors on the market which offer an open architecture for their travel time features but the list is growing and we continue to advocate for open protocols & interoperability.?As I mentioned earlier, ITS Central (link) was our first software partner and it continues to be widely used by us and our clients.?In addition to ITS Central, Genetec Security Centre (link) is another innovative software vendor advocating for open architectures.?They have also integrated our sensor and we have a number of large deployments with them as well, especially in Quebec.?As many municipal & provincial clients already use Genetec for their Video Management System (VMS), incorporating travel time into the same application only makes sense, and makes product adoption that much easier since they’re already familiar with the platform.?In addition, Genetec has released a number of traffic specific modules allowing it to perform both VMS & ATMS functions.??

Applications

Using software allows you to turn detection data into a meaningful solution/application.???These are the most common applications for these types of systems:

Travel Time Advisory – Providing travel time between two sensor nodes, or across multiple sensors as a “route”.?This data is used on highways and workzones in conjunction with Variable Message Signs (VMS) to provide travel time information to the public. For example: “Time to 401, 18 Mins"

Congestion Mapping – In recent years municipal signal network deployments have become more popular as these sensors allow you to build your own congestion maps.?Midblock count stations measure speed/occupancy at that specific point, but delays can occur right before or right after that. ?In contrast, travel time data is captured across an entire segment, so you can monitor any delays or issues across that entire area.

Origin Destination – As we can track specific devices from sensor to sensor, with a network-wide deployment we can determine which path devices took to get from point A to point B.?For example, using these sensors to determine what route fans take to get to a sports arena, and what are the popular routes for their departure.?

Intersection Delay – This is a newer feature but with the metrics generated by our sensor, simple delay at intersections can be measured.?With a network-wide deployment we can start making additional insights, like movement-specific delay, which we’re interested in developing.?

Learn More – Case Studies

In a few weeks, on Dec 7th @ 1pm, I’ll be hosting a webinar on this topic in which I’ll provide details on some of our key clients & projects as well as be available to answer any questions you have.?I hope to see you there. ?