Showing papers by "D. K. Arvind published in 2015"

Tracking and monitoring horses in the wild using wireless sensor networks

Abstract: The Retuerta is one of the oldest breed of horses in Europe, which roams wild in the Donana National Park, Andalusia, Spain. Thirty-two of these horses were marked with wireless sensors to gather spatio-temporal data on their behaviour over a period of several months. This paper describes our experiences of tracking and monitoring these wild horses attached with body-worn sensors and operating in a harsh and challenging environment. Analysis of this data for the first two months has revealed rare insights into the horses' social behaviour, such as the group dynamics (group sizes and memberships), dispersal and home ranges which are of interest to both animal ethologists and practitioners managing the ecology of their wild habitats. The paper introduces the Virtual Beacon — Time Division Multiple Access (VB-TDMA) protocol for orchestrating the data collection, and describes the choices that were made for addressing the many technical challenges for an extended deployment, such as in the design of the sensor platform, wireless data collection and battery lifetime issues. Our experiences point to the virtue of simplicity in design of wireless sensor networks to support core functionalities for achieving good average case performances.

Muscle strength testing using wearable wireless sensors

Abstract: Manual muscle testing and its variants have a long history of use for classifying muscle strengths. For the first time, inexpensive wearable wireless sensors combined with machine learning techniques are used to classify different levels of muscle strength, which addresses some limitations of the manual method. A mean accuracy of 93% was obtained across ten subjects using gyroscope and accelerometer data in classifying four distinct levels of strengths of the biceps brachii muscle when performing muscle contraction under appropriate load. This was reduced by 2% for accelerometer-only data, thus offering a potentially inexpensive and viable solution for testing muscle strength.

Performance Evaluation of the VB-TDMA Protocol for Long-term Tracking and Monitoring of Mobile Entities in the Outdoors

Abstract: The Virtual Beacon-Time Division Multiple Access (VB-TDMA) communication protocol has been proposed in [12] for a growing class of applications which require GPS tracking of autonomous mobile entities in the outdoors, and the long-term continuous monitoring of their contextual information using wireless sensors. Examples include monitoring animal behaviour in their natural habitat over the annual cycle, tracking shipping containers during their life-cycle of transit, loading/unloading and storage, and the handling of high-value packages during transportation. This paper employs simulations to evaluate the network performance of the VB-TDMA communication protocol in a representative scenario involving wild horses attached with collars, each containing a custom-designed platform with a three-axis accelerometer, a GPS module and ancillary electronics and battery, which uploads wirelessly to static base-stations, its position (sensed thrice an hour) and a summary of its activities between uploads. The simulations benefited from movement models derived from real data obtained from a long-term deployment of the collars on wild horses in the Donana National Park in south-west Spain. Comparisons with other MAC protocols have demonstrated the superior performance of the VB-TDMA protocol over a range of metrics for the representative example. An enhanced version of the VB-TDMA protocol - a multi-hop variant - is introduced for low latency requirements and was simulated for an urban scenario of bicycles fitted with sensors for crowd-sourcing spatio-temporal air quality information along the route of travel which is uploaded to the server when within range of static base-stations, for cases where low latency data upload is a requirement to enable access to the latest air quality information.

The Speckled Cellist : Classification of Cello Bowing Techniques using the Orient Specks

Abstract: Cello bowing techniques are classified by applying supervised machine learning methods to sensor data from two inertial sensors called the Orient specks -- one worn on the playing wrist and the other attached to the frog of the bow. Twelve different bowing techniques were considered, including variants on a single string and across multiple strings. Results are presented for the classification of these twelve techniques when played singly, and in combination during improvisational play. The results demonstrated that even when limited to two sensors, classification accuracy in excess of 95% was obtained for the individual bowing styles, with the added advantages of a minimalist approach.

The speckled cellist: classification of cello bowing techniques using the orient specks

Abstract: Cello bowing techniques are classified by applying supervised machine learning methods to sensor data from two inertial sensors called the Orient specks -- one worn on the playing wrist and the other attached to the frog of the bow. Twelve different bowing techniques were considered, including variants on a single string and across multiple strings. Results are presented for the classification of these twelve techniques when played singly, and in combination during improvisational play. The results demonstrated that even when limited to two sensors, classification accuracy in excess of 95% was obtained for the individual bowing styles, with the added advantages of a minimalist approach.