Advances in content transformation in heterogeneous tactical wireless networks

Content transformation is a viable technique for managing information delivered to tactical MANET nodes interconnected via a quasi-static network (QSN), especially during the period when available bandwidth is limited due to mobility, terrain, weather, jamming, etc. In previous work, we proposed Heterogeneous Intelligent Filtering (HIF) in multi-domain heterogeneous networks, for intelligent active data filtering and transformation to match network capacity and end-user capability. The need for data filtering was determined solely based on network conditions, for example, bandwidth available to the destined MANET nodes at the time. In this work², we report on enhancing HIF to automatically sense and take into consideration end-user intent and interests in data filtering. We illustrate how a speech-to-text engine can be used to deliver text in place of original speech, and save bandwidth usage, when a circumstance (e.g., a noisy environment) warrants such filtering. We also present several topic-of-interest- and location-based filtering features, and describe how one can combine them to create user intent filtering rules for selectively delivering and/or rerouting messages of interests in applications such as C2MINCS and XMPP-based Chat. We then describe additional HIF extensions and enhancements for deployment and operation support, including the use of HIF as a data dissemination service for individual applications, a distributed primary-backup HIF server mechanism for fault tolerance, and a policy management framework for managing the configuration of HIF components.