Beampatten optimization in distributed beamforming using multiobjective and metaheuristic method

Distributed beamforming is a communication method in wireless sensor networks (WSNs) where the sensor nodes collaboratively create a virtual antenna to direct their radiating power towards the direction of an intended destination. This method could increase the transmission range of the network and save the sensors' energy. However, due to the random locations of the sensor nodes, the beampattern for a finite number of nodes usually has asymmetrical sidelobes with high sidelobe levels. Higher sidelobe levels cause undesirable interferences at directions other than the intended destination. Conventional sidelobe reduction methods proposed for centralized antenna array cannot be used for distributed beamforming networks. This paper proposes a distributed network compliant, multi-objective weight optimization technique to produce a beampattern with lower sidelobe levels, higher directivity and minimal energy. Exhaustive search for the most favorable weight solutions is time-consuming when the number of sensor nodes is large. Therefore, this paper analyses the use of nature-inspired metaheuristic algorithms to solve for the best weight values at each sensor node. Three algorithms were analysed, namely, genetic algorithm (GA), particle swarm optimization (PSO) and gravitational search algorithm (GSA). Simulation results show that the proposed multi-objective weight optimization using nature inspired algorithm can provide improved beampattern with lower sidelobes, higher directivity and better energy savings