Design Of High-Performance Small Array Antennas And Novel Convex Optimization Synthesis/Near-Field Measurement Methods For Arrays

Benefit from their flexible and diverse beam control capabilities as well as several features that help to improve system performance indices,antenna arrays are widely used in various modern wireless systems.With the rapid development of modern wireless electronic systems,the performance requirements for antenna arrays have also become more urgent and more diverse.Due to their simple structure,flexible and diverse performance,as well as low cost,small array antennas can be widely used in various application scenarios,which have received extensive attention and developed in recent years.On the other hand,the array optimization synthesis method based on convex optimization can be widely used in the efficient solution of various array optimization synthesis problems due to its excellent algorithm versatility,flexibility and optimization efficiency.In addition,flexible and diverse measurement methods help to deepen the understanding of array performance and provide more convenient and efficient means for the array performance verification,which is also one of the active research areas in the array antenna area.Due to the aforementioned three aspects,in this dissertation several novel high-performance small array antennas based on new design methods are designed and implemented,several efficient optimization synthesis algorithms for antenna arrays based on the convex optimization method are proposed,as well as several antenna array near-field measurement and verification methods based on the computational electromagnetic methods are addressed.Main contributions of this dissertation can be summarized as following.1.A circularly polarized(CP)aperture antenna element with the multi-mode broadband operation is proposed.By combining with the machine learning optimization method,performance indices of the element can be optimized globally.By combining with a specially designed wideband feeding network,two CP antenna arrays with stable radiations are further proposed.First,based on the guidance of modal analysis and multi-mode design concept,a CP aperture antenna element with the multi-mode broadband operating characteristic is designed.Then,by loading a reflecting cavity and a parasitic square ring,the element can achieve the broadband unidirectional CP radiation.By using the global optimization algorithm based on the machine learning to optimize the element structural parameters,a significant improvement on the broadband CP performance is also achieved.Finally,by combining the high-performance element with the proposed broadband accurate-out-of-phase simplified sequential rotation feeding network,both a broadband two-element array and a broadband four-element array are realized.The designed arrays have some attractive advantages on stable broadband CP radiations,which validate effectiveness of the implemented design and optimization methods.2.Two broadband CP arrays based on the modularized design concept with a full substrate integrated waveguide(SIW)structure and a simplified sequential rotation feeding are proposed.By adopting a polarization grid structure based on anisotropic artificial equivalent medium,a broadband CP antenna element with the SIW feeding is realized.By using symmetry of the element structure and phase distribution characteristics of the SIW feeding component,a simplified broadband four-element CP array with the full SIW sequential rotation feeding is also designed and implemented.Finally,to solve the large sidelobe problem in the first array,another array with a three-dimensional SIW feeding cavity is proposed to reduce the array element spacing.The modularized design concept is adopted in design process of the array,by which the design efficiency is greatly improved,and the feeding networks with simplified structures can be implemented.Thanks to this design concept,a high-performance CP array antenna can be obtained finally.3.Based on the convex optimization method,some array optimization synthesis methods for maximizing the array directivity considering the sidelobe constraint are proposed.By transforming the problems of maximizing directivity of the array pencil beam without and with the sidelobe constraint into standard convex optimization problems,array directivities of the array pencil beams can be optimized globally,leading to better array directivity results.Then,in order to deal with the non-convexity in the wide-beam directivity maximization problem,a new iterative directivity optimization method based on Taylor expansion of the quadratic function is proposed.With the convex optimization method,the original non-convex problem can be effectively solved in an iterative manner.Finally,a two-stage iterative method is further proposed to deal with the sidelobe constraints in the wide-beam directivity maximization problem.Two initial schemes for these two iterative algorithms and their influence on the optimized results are also studied in detail.The proposed convex-optimization-based methods can be applied to optimization of the array directivities considering the sidelobe constraints with arbitrary main-beam requirements and arbitrary array configurations,which have very good versatility and robustness.4.Based on the convex optimization method,an efficient time-modulated array(TMA)directivity maximization method and an efficient particle-swarm-optimization-based TMA multi-objective optimization method are proposed.First,by introducing appropriate transformations and approximations,the TMA directivity maximization problem considering the sidelobe constraints can be effectively solved by a two-stage iterative method.Many TMA directivity maximization problems in several TMA scenarios can be efficiently solved with better directivity results.Then,by using solution of the convex optimization problem as initial prior information of the multi-objective particle swarm algorithm,the multi-objective optimization algorithm can be located around the most effective area,which can be used to efficiently solve the TMA multi-objective optimization problems.These two TMA optimization methods could improve efficiency and quality of the problem solution by appropriately using the problem convexity,further leading to better TMA optimization results.5.Three different types of antenna array near-field measurement methods are studied and discussed.A fast simulation method for the algorithm verification in the near-field measurement is established,and a compensation method for improving the amplitude errors in the single-cut near-field linear array measurement is proposed.First,three different methods such as the wave spectrum expansion,the equivalent source and the multipole expansion,as well as their applications in the three-dimensional near-field measurement problem are addressed and compared.Numerical simulation examples are adopted to reveal their pros and cons.Influences of the measurement errors on these three methods and a fast simulation method fully considering the probe influence for the algorithm verification are also discussed.Based on the above studies,three fast measurement methods for measuring the linear-array single-cut patterns are further studied,and a compensation method for improving the amplitude errors is proposed.The established framework of near-field measurement methods for the antenna arrays has a wide range of versatility and practicability,which can be widely applied to various measurement and verification tasks for antenna arrays.