Usage of data mules to the critical task of network reliability
Abstract
A wireless ad-hoc network consists of several transceivers (nodes) located in the plane, communicating by radio. Unlike wired networks, in which the link topology is fixed at the time the network is deployed, wireless ad-hoc networks have no fixed underlying topology. The physical topology of the network is determined by the distribution of the wireless nodes, as well as the transmission range of each node. The ranges determine a directed communication graph, in which the nodes correspond to the transceivers and the edges correspond to the communication links. The massive data streams and diverse modalities of digital information sources being captured at incredible rates have enriched the confluence of wireless ad-hoc nodes and intelligent sensors. The intelligent sensor ideally offers sensing and actuating capabilities combined with on-node computation and communication functions, while implemented in extremely miniaturized form factor and operating in an energy-autonomous manner. As sensor networks can now provide a rich and varied source of data streams, the need for reliable sensor information, as well as for sensing infrastructure and analytical sense-making tools that are able to make informed decisions, creates many new challenges. One of the most interesting and practical class of applications for sensors networks is battlefield monitoring, where wireless nodes are scattered over a geographical area and form a dynamic, infrastructure-less ad-hoc network. Typical monitoring scenarios consist of two stages. In the first step, the sensors periodically sense their surroundings, collect data and process it if needed. In the second step, the collected data is delivered to a base station using either local or global data collection process. In local data collection the nodes use multi-hop communication over a static topology or dynamic topology to relay their message to the base station. In global data collection, the nodes are visited by a traveling mule, which uses short-range wireless communication to collect data from nearby nodes. In this proposal we are suggesting to investigate the usage of data mules to the critical task of network reliability. That is, how to use the advantages of mobility capabilities to prevent losing crucial information while taking into consideration the additional operational costs. The data mule approach is most suitable when the network topology is sparse, the distance to the nearest node or base station is too large, or when the communication infrastructure between nodes is unstable. In addition, this approach reduces the responsibility for message routing from the nodes thereby minimizing their processing power. The added knowledge from visiting one node in the area might contribute significantly to the overall understanding of the environment. However, the knowledge obtained by visiting additional nodes in nearby locations may not be worth the time and transmission cost of the visit. Thus, the value from visiting a node is not a constant and depends on the nodes that were visited before it. Moreover, we have to take into account the proximity of data mules to the sensor nodes, while taking care not to reduce the Signal-to-Interference-plus-Noise Ratio (SINR) at mules too much so as to prevent reception. We will study the use of mules which have the ability to reach failed sensors, repair them and temporarily replace their role in the task of data collection and transfer. The mules will be used to maintain and improve the network s resiliency and reliability. We aim to optimize the mules deployment as to minimize the sensors down-time and the mules travel distance.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Feb 14, 2019
- Source ID
- W911NF1810399
Entities
People
- Michael Segal
Organizations
- Army Contracting Command
- Ben-Gurion University of the Negev
- United States Army