Impacto de técnicas com múltiplas antenas em uma rede de sensores sem fio baseada em cooperação probabilística

Abstract

The present work evaluates the effects on the performance of a wireless sensor network by adopting multiple antennas in the sensors in a cooperative network. The main idea is to combine two techniques that exploit spatial diversity: cooperation and MIMO. In the analyzes two MIMO schemes were used: Antenna selection (AS) and beamforming transmission using the singular value decomposition technique (SVD). The SISO transmission is also considered for comparison purposes. For the cooperative scheme, the probability of ON-OFF was modeled for networks in which the nodes have multiple antennas, extending analyzes of the literature that considered only nodes with a single antenna. In this way it is intended to reduce the energy consumption by extending the useful life of the sensors, increasing the amount of information effectively transmitted by the network, always maintaining a certain probability of outage as a restriction. It was verified that when both techniques are used there is a significant gain in the amount of information transmitted by the network with emphasis on the AS at short distances. By increasing the number of antennas, it was found that a lower ON-OFF probability of cooperation is required, providing a considerable gain for the AS and degradation in the SVD. It was also found that for each network size (distance between nodes) there is a transmission power at which a greater amount of information can be transmitted. Finally, an exhaustive search algorithm was proposed to find the best transmission power for each transmission scheme.