The integration of machine learning (ML) approaches in sensor-based applications in the field of pervasive computing is becoming increasingly prominent due to the increasing number of sensor-based applications in general and the continuous adaption of ML approaches to new domains. Several ML models are used within a processing pipeline that operates on the same sensor data. Still, the cloud computing approach is a straightforward solution where all sensor data is sent to the cloud before processing, which is inefficient according to resource utilization. Appropriate management of the different processing tasks for ML models enhances resource utilization.This paper proposes an architecture for resource-aware classification empowered by an ML model management (MLMM) framework and a distributed data stream management system (DDSMS). First, the classification pipeline is decomposed and implemented as data stream operators. Second, ML models are retrieved from an MLMM framework considering preprocessing, segmentation, and feature alignment to enable an effective redundancy elimination. Finally, the classification pipeline is deployed using resource-aware operator placement optimization. The evaluation results on a real-world scenario of a sensor-based activity classification pipeline for dairy cows show that our approach can reduce network utilization up to 98.9%.

The increasing availability of indoor and outdoor location sensors raises the interest for understanding the mobility in different places, e.g., in museums, train stations, or cities. One of the highly used approaches for understanding the mobility of individuals is to detect staypoints from trajectories. Although offline detection of staypoints fits best for long-term planning, some applications require online staypoint detection from trajectories, like online recommendation systems. Latency plays an important role here. In this paper, we propose a stream-based approach for staypoint detection, which can be realized by applying a Data Stream Management System. We claim that the proposed approach can detect staypoints with low latency from the high-frequency location update streams. To evaluate our approach, we compare it with a batch-based approach on real data from an indoor tracking system and Geolife dataset. The results demonstrate that the online approach detects staypoints with much lower latency compared to traditional approaches like offline and batch processing. Moreover, we prove that the accuracy of the stream-based approach is similar to the batch-based approach.

Machine Condition Monitoring (MCM) is an important topic for the reliability of industrial machines in increasingly interconnected production facilities. The analysis of a huge amount of data to get information about the machine's condition is a difficult challenge. Current solutions for these analyses are often very specific, need a lot of manual configuration or are difficult to apply. In this paper, we present a system that uses anomaly detection in data streams to find hints for faulty machines in the data. The basis of this system is a Data stream management system (DSMS), which can handle huge amounts of streaming data and simplifies the definition of analyses. Due to the anomaly detection algorithms, the approach can be applied to a variety of data and scenarios. The outcome is a system that allows live analysis of machine data for MCM.

The complexity of critical systems such as traffic management has dramatically increased over the last decades, since they involve more and more sensors to derive distributed situational awareness. Most existing systems require an a-priori configuration of sensors or need human intervention to adapt to changes. Furthermore, analysis of data quality or query plan reliability is often not possible and management of recorded data is done by hand. Our goal is to support the research, development, evaluation and demonstration of such systems. Within this study we analyze requirements and challenges for the data management of sensor based research environments and present a data stream based architecture which fulfills these requirements.

One of the main challenges in the development of traffic systems is to assure safety for all road users. Hence, especially expensive vehicles are equipped with advanced driver assistance systems (ADAS) that use data about the vehicle and information about objects in the proximity of the vehicle to execute the assistance function. These objects have to be detected by sensors and they have to be tracked over multiple scans to keep the object's state up-to-date. Usually, such ADAS are developed as proprietary systems that are tailored for the specific assistance function and the specific sensors in use. Indeed, that leads to a very efficient system. However, changing system properties, e. g. an exchange of sensors, is very expensive. In this case, very often at least some parts of the system code have to be reimplemented. To solve this problem of bad maintainability which arises especially during the development of new assistance functions in this work a new architecture for ADAS is presented. The relevant information for the assistance function is no longer provided by hard coded, predefined processes, but by flexible continuous operator plans in a datastream management system. These operator plans build up a dynamic context model of the vehicle's environment. The context model is kept up-to-date by object tracking operators in these operator plans and is then used as a data source to extract information for different assistance functions. This extraction is also done by operator plans that produce only relevant information and discard other information.