Thomas Holleczek

Collecting complex activity datasets in highly rich networked sensor environments

We deployed 72 sensors of 10 modalities in 15 wireless and wired networked sensor systems in the environment, in objects, and on the body to create a sensor-rich environment for the machine recognition of human activities. We acquired data from 12 subjects performing morning activities, yielding over 25 hours of sensor data. We report the number of activity occurrences observed during post-processing, and estimate that over 13000 and 14000 object and environment interactions occurred. We describe the networked sensor setup and the methodology for data acquisition, synchronization and curation. We report on the challenges and outline lessons learned and best practice for similar large scale deployments of heterogeneous networked sensor systems. We evaluate data acquisition quality for on-body and object integrated wireless sensors; there is less than 2.5% packet loss after tuning. We outline our use of the dataset to develop new sensor network self-organization principles and machine learning techniques for activity recognition in opportunistic sensor configurations. Eventually this dataset will be made public.

OPPORTUNITY: Towards opportunistic activity and context recognition systems

Opportunistic sensing allows to efficiently collect information about the physical world and the persons behaving in it. This may mainstream human context and activity recognition in wearable and pervasive computing by removing requirements for a specific deployed infrastructure. In this paper we introduce the newly started European research project OPPORTUNITY within which we develop mobile opportunistic activity and context recognition systems. We outline the projects objective, the approach we follow along opportunistic sensing, data processing and interpretation, and autonomous adaptation and evolution to environmental and user changes, and we outline preliminary results.

Textile pressure sensors for sports applications

Wearable sports trainers are built upon sensor systems recognizing the activities performed by its users. In snowboarding, one of the fastest growing sports in the world, traditional activity recognition approaches make use of pressure insoles with force-sensitive resistors, which, however, are particularly uncomfortable to wear. To make these measurements more convenient, we have developed textile pressure sensors using the principle of a variable capacitor. Electrodes of conductive textiles coated with silver arranged on both sides of compressible spacers made from Croslite™ form a capacitor, whose capacitance indicates the applied pressure. We integrated three sensors into a snowboarding sock at relevant positions under the heel and the ball of the foot. Outdoor experiments on a ski slope in the Matterhorn Glacier Paradise (Zermatt, Switzerland) show that the machine learning algorithm NCC can detect turns, the basic activity of snowboarding, from the sensor data with an accuracy of 84 percent. Moreover, indoor experiments reveal that NCC can clearly distinguish whether a person wearing our sensor socks is standing on the ball of the foot, flat or on the heel. These results suggest the socks might also be used for gait analysis or the monitoring of the in-shoe pressure distribution of runners.

An Educational and Research Kit for Activity and Context Recognition from On-body Sensors

We present an educational and research kit to support hands-on teaching and experience of real-time activity or gesture recognition from on-body sensors. The kit is comprised of: wireless wearable sensor nodes for motion and ECG sensing; software infrastructure for synchronized data acquisition from multiple sensors, data visualization, signal alignment, and synchronized signal/video exploration; algorithm demonstration and education software with a hidden Markov model-based activity recognition system and a low-latency gesture recognition for a platform game; support hardware for the annotation of user activities from a wireless keypad, and for prototyping of other wireless sensor nodes. All hardware and software is open-source.

Recognizing turns and other snowboarding activities with a gyroscope

Wearable sports trainers are built upon sensor systems recognizing the activities performed by its users. In snowboarding, one of the fastest growing sports in the world, traditional activity recognition approaches make use of insoles with integrated force-sensitive resistors. However, such insoles are usually uncomfortable to wear, tedious to calibrate, error-prone and their functionality is limited to the detection of turns. We have therefore developed an alternative snowboarding activity recognition system, which overcomes these downsides. It consists of a mobile computing device with a GPS receiver, and a gyroscope attached to the center of the snowboard. The system is easy to set up, as the gyroscope can be simply taped onto the snowboard without any user-specific calibrations. Experiments with seven riders in three Swiss ski resorts show that our activity recognition system is capable of withstanding harsh conditions on outdoor slopes. It detects turns, the basic elements of snowboarding, independently of user, snowboard, slope and snow characteristics. In addition to the mere detection of turns, our system can reveal whether the snowboarder is riding on the frontside or the backside of the board, whether she is going forwards or backwards and whether she is carving or skidding. Finally, first studies suggest that gyroscope-based activity recognition can also be applied to snowboarding-like sports such as skiing and skateboarding.

Towards an Interactive Snowboarding Assistance System

Snowboarding is one of the fastest growing sports in the world. However, it is rather difficult to learn. Snowboarders, who have reached an average level of expertise, often find it difficult to improve their style without taking expensive private lessons. Moreover, some of their performed actions would not even be visible for the instructors, as they happen inside the shoes. We are therefore developing an interactive integrated sensor system on the snowboard and on the body, which is capable of assisting snowboarders on the piste and helps them improve their riding style.

RidersGuide: The First Real-Time Navigation System for Ski Slopes

Although being great fun to exercise, skiing and snowboarding entail the risk of accidents such as any other high-energy participation sport. A factor increasing the risk of accidents is foggy weather, when the view limited to a few meters and safety-critical areas such as crossings, steep areas, obstacles and the edges of the slope cannot be made out. In this paper, we present a prototype of a navigation system supporting the wayfinding of skiers and snowboarders on ski slopes. It consists of a mobile computing device with a GPS receiver and an audio and visual feedback system. In its basic functionality, the prototype recognizes when skiers and snowboarders approach safety-critical areas and generates corresponding warnings. It is moreover capable of guiding skiers and snowboarders down the slope on safe paths to minimize risk with the help of routing instructions. We evaluated our prototype system on ski slopes in Grindelwald and Savognin, Switzerland, in a user study with 20 participants. According to the study, safety warnings, which have to be perceived instantly, are recognized best through spoken audio messages. Contrariwise, video feedback outperforms natural speech regarding routing instructions.

Statistical Analysis of IP Delay Measurements as a Basis for Network Alert Systems

Measuring packet delay and packet loss through dedicated test packet streams in computer networks allows for the assessment of Quality of Service on the path the probes traverse. For this reason, active IP performance measurements have been carried out in X-WiN, Germanys National Research and Educational Network, and GEANT2, its European counterpart, by the WiN-Labor Group of the University of Erlangen-Nuremberg for many years. This paper provides an overview of how network measurements can be interpreted using a variety of statistical concepts in order to gain insight into the state of monitored connections. The focus here is set on the analysis of One-Way Delay (OWD); the approach demonstrates how OWD can be used for the identification of network congestions and can serve as a basis for network alert systems by statistically categorizing changes in the performance metric.