Field Studies at Industrial and RF-Harsh Environments

  1. Design and execution of systematic field studies at RF-harsh environments, wherein scalable Distributed Sensor Networks are the core component of heterogeneous network platforms for guaranteeing the bidirectional data flow between the sensors and the end-users. A three-phase procedure is employed, incorporating: (a) benchmark studies for selecting the appropriate hardware platform, designing the system’s architecture, and studying the effect of the industrial environment, (b) implementation and on-site evaluation of the proposed solution, (c) final deployment for 24/7 operation.deployment
  2. Unsupervised Feature Selection framework for characterizing the end-to-end performance of links established over multi-hop WSN topologies. Opposed to the well-studied point-to-point links that are formulated at the Physical layer, and are capable of link quality estimation between 1-hop neighbors, end-to-end links expand towards two different directions: (a) across different sides of the network, exceeding the constrained limits of 1-hop neighborhoods, (b) across different layers of a fully functional protocol stack, ranging from the Physical to the Transport and Application layers. As such, end-to-end links convey a larger volume of information than the one captured by point-to-point, low-level links. Our framework considers: (a) the collection of diverse network parameters in a passive fashion, which introduces to the network neither additional, nor dedicated traffic, (b) the combination of network metrics collected from different sides of the network, and corresponding to different layers of the protocol stack to a feature-level fusion mechanism for delivering high-level inference on the dominant network features, (c) the synthesis of a thorough learning model for characterizing the performance of a multi-hop WSN, that covers the formulation of the classification problem and the engineering of network features.

 

Related Projects: HYDROBIONETS (FP7-287613), 2011-2014

 

Autonomic Networking Techniques for Wireless and Body Sensor Networks

  1. Synthesis, implementation, and evaluation on limited-capacity platforms of distributed and localized algorithms for addressing topology control aspects in massive scales. Our approach exploits the concept of Delaunay triangulations in order to establish a scalable framework for optimizing the transmission power needed for establishing end-to-end connectivity in polynomial time and a localized fashion that, against current state-of-art, eliminates the necessity of additional network traffic. Implementation on Contiki OS is available here.
  2. Theoretical modelling and applied studies for enabling the on-node and in-network feature level-fusion on data-intensive sensor networks with critical real-time characteristics, e.g., Body Sensor Networks. Our approach takes into account the hardware and network imperfections of low-power sensors for crafting novel unsupersived techniques for on-line learning, emphasizing on graph-based feature selection. Our aim is to provide a set of middleware services for self-learning, power- and conditions-aware networks that can adapt their main operational characteristics in real-time, without sacrificing the quality of sensing
  3.  

    Related Projects: HYDROBIONETS (FP7-287613), 2011-2014, SENSE (GSRT), 2013-2015