Step Counter


Description of Technology

A team of researchers from the University of Calgary’s Department of Geomatics Engineering led by Dr. Naser El-Sheimy has developed a new step detection technique that successfully detects steps under varying motion speeds and device use cases with an average performance of 99.6%, and outperforms some of the state of the art techniques that rely on classifiers and commercial wristband products. It adaptively tunes the filters and thresholds used without the need for presets while accomplishing the task in a real-time operation manner.   The algorithm is based on conclusions drawn from extensive analysis of the three signals used for the step detection which are the acceleration norm, angular rates vector, and magnetic vector.

Table 1. Average accuracy of the proposed algorithm

Areas of Application

The software can be easily implemented in a wide range of applications in the field of navigation and health monitoring where intertial navigation systems are often used.

  • Fitness tracking/quality of life-style assessment
  • Pedestrian dead reckoning
  • Search and rescue
IP Status
  • Patent Pending
Competitive Advantages

The algorithm can be used to complement Global Navigational Satellite Systems when there is degraded service such as when you’re located next to large buildings or indoors. The use of the algorithm allows the device to move freely since it does not have to be strapped down through the use of ankle or wristbands. By eliminating the need for the device to be strapped in a set position it allows the software to be run on consumer devices such as a smartphone. Whether the phone is in the texting, talking, or in the pocket position, it provides higher accuracy when measuring the number of steps than competing methods. This software creates high-grade Inertial Measuring Units that cost less, and require less energy when compared to competing products. A further benefit is that it can be installed on smartphones.

  • Higher accuracy
  • Lower power requirement and real time computation
  • Cost effective and uses existing consumer hardware
  • Does not require classifier or presets
Stage Of Development

Prototype tested for accuracy with positive results. The proposed algorithm outperformed the accuracy of two of fitness bands available in the market.

Table 2. Performance evaluation in comparison to Flex2 and MiBand2