Sidelobe Control in Collaborative Beamforming via Node Selection

TL;DR

This paper proposes a distributed, low-complexity node selection method to control sidelobes in collaborative beamforming for wireless sensor networks, significantly reducing interference and increasing network capacity.

Contribution

It introduces a scalable node selection algorithm with low-rate feedback for sidelobe control in WSNs, addressing limitations of traditional centralized techniques.

Findings

Interference is significantly reduced with node selection.

The algorithm requires a low number of trials for node selection.

Simulation results confirm theoretical analysis.

Abstract

Collaborative beamforming (CB) is a power efficient method for data communications in wireless sensor networks (WSNs) which aims at increasing the transmission range in the network by radiating the power from a cluster of sensor nodes in the directions of the intended base station(s) or access point(s) (BSs/APs). The CB average beampattern expresses a deterministic behavior and can be used for characterizing/controling the transmission at intended direction(s), since the mainlobe of the CB beampattern is independent on the particular random node locations. However, the CB for a cluster formed by a limited number of collaborative nodes results in a sample beampattern with sidelobes that severely depend on the particular node locations. High level sidelobes can cause unacceptable interference when they occur at directions of unintended BSs/APs. Therefore, sidelobe control in CB has a potential to increase the network capacity and wireless channel availability by decreasing the interference. Traditional sidelobe control techniques are proposed for centralized antenna arrays and, therefore, are not suitable for WSNs. In this paper, we show that distributed, scalable, and low-complexity sidelobe control techniques suitable for CB in WSNs can be developed based on node selection technique which make use of the randomness of the node locations. A node selection algorithm with low-rate feedback is developed to search over different node combinations. The performance of the proposed algorithm is analyzed in terms of the average number of trials required to select the collaborative nodes and the resulting interference. Our simulation results approve the theoretical analysis and show that the interference is significantly reduced when node selection is used with CB.