This technology is a wireless communication control device that selects k nodes to participate in analog cooperative beamforming from n sensor nodes distributed on a 2D plane within a circular boundary. It uses a weight matrix trained on the 2D coordinates of the receiver to direct cooperative transmission and receives feedback signals indicating transmission success from the selected nodes.
In wireless sensor networks, transmitting data from individual nodes separately is vulnerable to eavesdropping by third parties and results in low diversity gain at the receiver. While cooperative beamforming has been proposed as an alternative, there has been no efficient method for selecting which nodes among many scattered on a plane should participate in cooperative transmission.
This technology converts the receiver's 2D coordinates into a 1×2 matrix, applies a weight matrix and a softmax activation function to calculate probabilities for each node, and uses machine learning to ensure that nodes showing maximum SNR in pre-tests have the highest probability. When a transmission event occurs, it instantly selects the optimal node combination based solely on coordinates and verifies transmission results via feedback, simultaneously blocking eavesdropping and improving diversity gain. It can be applied to planar sensor networks where nodes are spread across the ground, such as in farmland, industrial complexes, or flat smart cities, enabling the construction of secure and reliable data collection links without complex channel estimation.
This invention was developed with research support from the Ministry of Science and ICT for magnetic resonance-based beamforming technology for selective wireless power transmission systems.
N/A
