Using the GNU Octave system to process data coming in over the network

Рассмотрен подход к построению в среде OMNeT++ с использованием фреймворка INET расширенных моделей беспроводных сетей, позволяющих проводить исследование затрат энергии на каждом из узлов сети в процессе их сетевой работы и учитывать особенности сетевого взаимодействия узлов в условиях их мобильности. Показана возможность организации связи между двумя беспроводными узлами, находящимися вне зоны прямой радиодоступности, через ряд промежуточных узлов. An approach to building in the OMNeT ++ environment using the INET framework of advanced models of wireless networks is considered, which makes it possible to study both the energy costs of each of the network nodes during their network operation and take into account the features of network interaction of nodes in the conditions of their mobility. The possibility of organizing communication between two wireless nodes located outside the direct radio accessibility zone through a series of intermediate nodes is shown.

Wireless Sensor Networks Fault-Tolerance Based on Graph Domination with Parallel Scatter Search

In wireless sensor/ad hoc networks, all wireless nodes frequently flood the network channel by transmitting control messages causing “broadcast storm problem”. Thus, inspired by the physical backbone in wired networks, a Virtual Backbone (VB) in wireless sensor/ad hoc networks can help achieve efficient broadcasting. A well-known and well-researched approach for constructing virtual backbone is solving the Connected Dominating Set (CDS) problem. Furthermore, minimizing the size of the CDS is a significant research issue. We propose a new parallel scatter search algorithm with elite and featured cores for constructing a wireless sensor/ad hoc network virtual backbones based on finding minimum connected dominating sets of wireless nodes. Also, we addressed the problem of VB node/nodes failure by either deploying a previously computed VBs provided by the main pSSEF algorithm that does not contain the failed node/nodes, or by using our proposed FT-pSSEF algorithm repairing the broken VBs. Finally, as nodes in a VB incur extra load of communication and computation, this leads to faster power consumption compared to other nodes in the network. Consequently, we propose the virtual backbone scheduling algorithm SC-pSSEF which aims to find multiple VBs using the VBs provided by the pSSEF algorithm and switch between them periodically to prolong the network life time.