Multi-pattern synthesis in fourth-dimensional antenna arrays using BGM-based quasi-Newton memetic optimization method

• Published in: International journal of microwave and wireless technologies 2024-03, Vol.16 (2), p.284-294
• Main Authors: Chakraborty, Avishek, Saxena, Ravi Shankar, Verma, Anshoo, Juyal, Ashima, Gupta, Sumit, Singh, Indrasen, Ram, Gopi, Mandal, Durbadal
• Format: Article
• Language: English
• ISSN: 1759-0787; 1759-0795
• DOI: 10.1017/S1759078723000910

The advancement in wireless communication is fueling the growth of innovative antenna array designs toward cost-effective and performance-oriented solutions. This paper proposed unconventional methods to design antenna arrays for multi-pattern synthesis without using attenuators or phase shifters. A low-cost alternative is proposed with “Time-modulation”-based antenna array capable of electronic scanning and beam steering. Here, “Time” is utilized as a fourth-dimensional (4D) array parameter, and that is why “Time-modulated” arrays are also called as 4D antenna arrays. The idea is to control the high-speed switch attached with each antenna periodically to produce desirable current and phase tapering. This article expanded the “Time-modulation” concept to synthesize multiple radiation patterns like monopulse patterns, scanned beam patterns, shaped beam patterns, and cosecant-squared beam patterns for multifunctional radar systems. Suitable time schemes are developed to generate the narrowband sum–difference patterns useful for monopulse radars. Simultaneous scanned beam patterns are also proposed for narrowband communication. Furthermore, to address the wideband applications, shaped flat-top beam patterns and cosecant-squared beam patterns are also proposed. In this regard, 20- and 16-element “Time-modulated” linear array antennas are developed, and the parameters of the arrays are controlled by suitably designed objective functions with quasi-Newton method (QNM)-based memetic optimization method. For this purpose, first a well-known genetic algorithm is adopted to search the potential trust regions in the exploration stage and QNM is used for fine-tuning. Furthermore, the Broyden’s good method-based direction-updating equation is used with QNM to improve the performance.