TrafInfo Communications, Inc.

Internet-based Wireless Telemetry

[ Home ]

Real-Time Traffic Signal Monitoring

The graphic above shows a TS2 Type 1 controller cabinet and the interface between the various existing equipment for real-time traffic signal monitoring.  To extract traffic counts from the existing loops, the Global Traffic Technologies LLC (GTT) Canoga™ 822 detector cards are utilized.  The Canoga 822 has a serial port on the front panel.  Using the serial communication protocol provided by GTT, traffic counts can be obtained for any user defined time interval. 

Signal timing information is obtained from existing EDI MMU-16E Memory Malfunction Unit.  Again, the MMU has a serial port on the front panel.  Using EDI proprietary serial communication protocol, the signal display status of all signal housings are obtained once every second.

 Central to the real-time monitoring system is the Trafmate 6 wireless transceiver.  Firmware within the Trafmate 6 monitors the signal display status by continuously communicating with the MMU twice every second.  Changes in the signal display status are used to determine the green, yellow and red intervals for each signal display.  The mainline phase with recall is used to determine the cycle length from the beginning of yellow to beginning of yellow.  At the end of each cycle, the Trafmate 6 extracts traffic counts from the Canoga 822 cards for the duration of the cycle.

The traffic volume and signal timing information collected by the Trafmate 6 for each signal cycle is transmitted wirelessly to TrafInfo’s central server.  At the server, the data was logged into a MySQL® database.

The Highway Capacity Manual (HCM) equations are used to compute the control delay for each lane group.  It is comprised for uniform delay (d1), and incremental delay (d2).  The main factors to compute the delay terms d1 and d2 are effective green to cycle ratio (g/C), volume to saturation flow ratio (V/S), as well as others including the saturation flow rate, progression factor (PF), adjustment for actuated control (k) and adjustment for upstream filtering (I).  Factors such as PF and I were assumed to remain constant from cycle to cycle.  These factors are obtained from tables in the HCM.  The saturation flow rate is obtained from the HCM as well and is also assumed to remain constant.  The actuated control adjustment factor (k) is computed as per equations in the HCM.  The g/C and v/S which vary for each cycle are computed from field data.  Using both the static and variable factors, the control delay is computed as a sum of uniform and incremental delays.

 The HCM control delay equations are not intended for use on a cycle-by-cycle basis.  They are typically meant to be used for peak 15-minute period.  For this reason, a moving average is used to compute the control delay for each cycle.  In other words, the total volume and average signal timing for the past 15 minutes are used to compute the delay for the most current cycle.

There are several benefits to real-time monitoring of signalized intersections.  First of all, it allows for monitoring the control delay for each lane group and/or approach.  The system provides a comprehensive profile of the delay, g/C and v/c variations throughout the day.  This information will be valuable to traffic engineers for signal retiming or for the need to create additional time-of-day or demand responsive plans.  The system collects this information on an on-going basis, which means it provides a great tool for conducting “before-and-after” assessment of signal timing changes.

Download white paper on real-time traffic signal monitoring.