Enhancing Mobile Data Collection Applications with Sensing Capabilities

Over the past years, using smart mobile devices for data collection purposes has become ubiquitous in many application domains, replacing traditional pen-and-paper based data collection approaches. However, in many cases, modern approaches only aim to replicate traditional data collection instruments (e.g., paper-based questionnaires) in a digital from (e.g., smartphone surveys). Thereby, the full potential of smart mobile devices is often not fully exploited. Most modern smart mobile devices comprise a variety of sensing capabilities, which may provide valuable data, and thus insights. In addition, external sensors and devices may be easily connected to become part of the overall data collection process. In order to integrate sensing functionality into existing data collection applications, one has to address each desired sensor manually from within the application, which may cause severe development effort. Alternatively one can fall back on dedicated sensing frameworks to perform sensing operations. However, the latter are often targeted towards one specific mobile platform (e.g., iOS or Android) or lack required functionality, which may also lead to unnecessary development overhead when implementing mobile data collection applications. To cope with these issues, a cross-platform mobile sensing framework that can be used within large-scale mobile data collection scenarios was developed in the context of this thesis. Thereby, an in-depth look at existing mobile sensing frameworks as well as common use case scenarios is taken. Further, requirements derived from the latter are explicitly stated and were taken into consideration in the course of the overall development process. The latter is documented and discussed in detail in the course of this thesis, including the design of a framework architecture, implementation details and the integration of the framework into mobile data collection applications.