Network reciprocity in modeling and analysis of phased array antennas

Abstract Achieving robust and fast two-dimensional adaptive beamforming of phased array antennas is a challenging problem due to its high-computational complexity. To address this problem, a deep-learning-based beamforming method is presented in this paper. In particular, the optimum weight vector is computed by modeling the problem as a convolutional neural network (CNN), which is trained with I/O pairs obtained from the optimum Wiener solution. In order to exhibit the robustness of the new technique, it is applied on an 8 × 8 phased array antenna and compared with a shallow (non-deep) neural network namely, radial basis function neural network. The results reveal that the CNN leads to nearly optimal Wiener weights even in the presence of array imperfections.

Download Full-text

Phased-array antennas using novel PSoC-controlled phase shifters for wireless applications

International Journal of Microwave and Wireless Technologies ◽

10.1017/s1759078721001094 ◽

2021 ◽

pp. 1-13

Author(s):

Aparna B. Barbadekar ◽

Pradeep M. Patil

Keyword(s):

Insertion Loss ◽

Phase Shifter ◽

Phased Array ◽

Phase Shifters ◽

Control Circuit ◽

Power Divider ◽

Phased Array Antenna ◽

Phased Array Antennas ◽

Array Antennas ◽

Low Insertion Loss

Abstract The paper proposes a system consisting of novel programmable system on chip (PSoC)-controlled phase shifters which in turn guides the beam of an antenna array attached to it. Four antennae forming an array receive individual inputs from the programmable phase shifters (IC 2484). The input to the PSoC-based phase shifter is provided from an optimized 1:4 Wilkinson power divider. The antenna consists of an inverted L-shaped dipole on the front and two mirrored inverted L-shaped dipoles mounted on a rectangular conductive structure on the back which resonates in the ISM/Wi-Fi band (2.40–2.48 GHz). The power divider is designed to provide the feed to the phase shifter using a beamforming network while ensuring good isolation among the ports. The power divider has measured S11, S21, S31, S41, and S51 to be −14, −6.25, −6.31, −6.28, and −6.31 dB, respectively at a frequency of 2.45 GHz. The ingenious controller is designed in-house using a PSoC microcontroller to regulate the control voltage of individual phase shifter IC and generate progressive phase shifts. To validate the calibration of the in-house designed control circuit, the phased array is simulated using $s_p^2$ touchstone file of IC 2484. This designed control circuit exhibits low insertion loss close to −8.5 dB, voltage standing wave ratio of 1.58:1, and reflection coefficient (S11) is −14.36 dB at 2.45 GHz. Low insertion loss variations confirm that the phased-array antenna gives equal amplitude and phase. The beamforming radiation patterns for different scan angles (30, 60, and 90°) for experimental and simulated phased-array antenna are matched accurately showing the accuracy of the control circuit designed. The average experimental and simulated gain is 13.03 and 13.48 dBi respectively. The in-house designed controller overcomes the primary limitations associated with the present electromechanical phased array such as cost weight, size, power consumption, and complexity in design which limits the use of a phased array to military applications only. The current study with novel design and enhanced performance makes the system worthy of the practical use of phased-array antennas for common society at large.

Download Full-text

Design of multi-step amplitude quantized weights for solid-state planar phased array antennas

Proceedings of the IEEE 1995 National Aerospace and Electronics Conference. NAECON 1995 ◽

10.1109/naecon.1995.521915 ◽

2002 ◽

Author(s):

Li-Jun Qin ◽

Jiang-Da Shao

Keyword(s):

Solid State ◽

Phased Array ◽

Phased Array Antennas ◽

Array Antennas ◽

Step Amplitude

Download Full-text