• Home
• Program
• Social Events
• Program at a Glance
• Keynote Speakers
• Presentation Guidelines
• Author Centre
• Important Dates
• Committees
• Sponsors and Exhibitors
• Registration Information
• Conference Venue
• Travel Information
• Contact

10:20   Swimming; performance measurements and athlete feedback Chair: Anton Sabo 10:20 20 mins MEDIATION: AN EMBEDDED SYSTEM FOR AUDITORY FEEDBACK OF HAND-WATER INTERACTION WHILE SWIMMING Daniel Cesarini, Chiara Farnesi, Diego Taddei, Davide Calvaresi, Stefano Frediani, Bodo E. Ungerechts, Thomas Hermann Abstract: In swimming sport, the proper perception of moving water masses is a key factor. This paper presents an embedded system for the acquisition of values of pressure on swimmers’ hands and their transformation into sound. The sound, obtained using sonification, is used as an auditive representation of hand-water interactions while swimming in water. The sound obtained is used as an auditive feedback for the swimmer and as an augmented communication channel between the swimming trainer and the athlete. The developed system is self-contained, battery powered and able to work continuously for over eight hours, thus, representing a viable solution for daily usage in swimmers’ training. Preliminary results from in-pool experiments with both novel and experienced swimmers demonstrate the high acceptability of this technology and its promising future evolution and usage possibilities. 10:40 20 mins TRACKING ELITE SWIMMERS IN REAL TIME WITH WEARABLE LOW-POWER WIRELESS SENSOR NETWORKS Jeroen Lecoutere, Robert Puers Abstract: Monitoring elite athletes in real-time remains challenging. This paper presents research towards the performance assessment of elite swimmers using wearable low-power sensor networks. This work focuses on the reduction of power consumption by optimizing the power cost of signal processing and sensor use. An algorithm exploits the sensor data in order to predict which information is valuable to assess the motion of the athlete in real-time. Moreover, the algorithm is designed to run on a low-power microcontroller unit and uses lightweight computational techniques that further reduce the power requirements Automating performance assessment enables the coaching staff to focus more on crucial aspects in the training process, such as technical or mental skills. In contrast to recreational sports or daily life tracking, elite athletes focus on every possible small increase in performance. Hence these athletes demand wearable trackers that should be as unobtrusive as possible. Swimmers basically only wear a swimsuit, goggles and a swim cap, so wearing any additional device already poses a significant obstruction. Our algorithm reduces the raw data of six sensors down to lap time and stroke type information, which corresponds to a 99.97% reduction in data compared to a 15 second lap raw data. Furthermore, our proposed method does not only reduce data through processing but also completely turns off specific sensors that contain no additional information depending on the detected activity. So in short, reducing wireless communication by reducing data, and reducing sensor overhead through smart algorithms, increases power efficiency. Nine elite athletes (National Championships level, 8 male, 1 female) were asked to perform two laps of backstroke, two laps of breaststroke and two laps of front crawl in a 25-meter pool at an easy pace with a sensor node placed on the back of their head. The proposed online algorithm was able to reduce the amount of used sensors to maximum three. A reduction in sensor power consumption of 12 mW (96%) is estimated over minimally 80% of the swimming time, depending on the swimmer’s lap times. Stroke types and turns were detected with 100% accuracy for each tested swimmer, despite different stroke techniques. 11:00 20 mins THE USE OF A CAP-MOUNTED TRI-AXIAL ACCELEROMETER FOR MEASUREMENT OF DISTANCE, LAP TIMES AND STROKE RATES IN SWIM TRAINING Scott Michaels, Stephen Turnock, Ben Holliss, Dominic Hudson, Alexander Forrester, Dominic Taunton Abstract: With the availability and prevalence of technical knowledge in sports increasing exponentially, along with the reduction in costs of technological equipment required for implementing training plans based on such technical knowledge, it is becoming increasingly important for athletes and coaches to implement data acquisition and analysis into daily training if the athlete is to achieve and maintain elite status. The principles of physiological training remain relatively constant from sport to sport (adjusted for intensity and duration); however methods of performance measurement vary due to limitations of each sport. Swimming is an example of a sport in which it is particularly challenging to take direct performance measurements. Currently, performance measurement in swimming during training is done through stopwatch lap-time measurement by coaches, combined with distance measurement and athlete-reported ratings of perceived exertion. But despite the limitations inherent to the activity, performance measurement in swimming has room to become more technical, due to swimming taking place in a predictable, controlled environment with a consistent, repeatable technique. More technical measurement methods will result in more accurate, consistent and comprehensive data collection, which in turn can lead to more advanced analysis, allowing faster and better identification of critical features in an athlete’s training progression. This paper will report the findings of a study currently underway recording accelerometer data of six elite level swimmers (four female and two male, varying primary event stroke and distance) over the course of a regular 15 week training block. Accelerometer measurement is used to determine when the athlete is swimming, marking of turning points (and therefore distance and lap-time measurements), and is processed by frequency analysis to determine stroke-rate. Comparison with video where available, and with training plans and literature where not, have proven this method to be accurate and reliable for determining these performance metrics, and feedback from elite coaches has indicated that developing this system as a simple, low cost method for estimating long-term athlete workload could be an extremely useful addition to their training regime.