Design and Optimization of Microwave Circuits and Devices with the Particle Swarm Optimizer

Introduction

The Particle Swarm Optimizer (Kennedy, Eberhart, & Shi, 2001; Robinson & Rahmat-Samii, 2004; Clerc, & Kennedy, 2002) has been successfully adopted as a powerful optimization tools in several areas of applied electromagnetic (Grimaccia, Mussetta, & Zich, 2007; Mikki, & Kishk, 2006; Robinson & Rahmat-Samii, 2004; Mussetta, M., Selleri, S., Pirinoli, P., Zich, R.E., & Matekovits, L. 2008; Nanbo Jin, & Rahmat-Samii, Y. 2010; Yilmaz, A.E. 2010) such as microwave imaging (Donelli & Massa, 2005; Caorsi, Donelli, Lommi, & Massa, 2004; Huang, & Mohan, 2007; Huang, C.-H., Chen, C.-H., Chiu, C.-C., & Li C. L. 2010; Genovesi, S., Salerno, E., Monorchio, & A., Manara, G. 2009), antenna design (Robinson, Sinton, & Rahmat-Samii, 2002; Jin, & Rahmat-Samii, 2005; Migliore, Pinchera, & Schettino, 2005, Azaro, De Natale, Donelli, Massa, & Zeni, 2006a; Azaro, De Natale, Donelli, & Massa, 2006b; Azaro, Boato, Donelli, Massa, & Zeni, 2006c; Azaro, Donelli, Franceschini, Zeni, & Massa, 2006c; Fimognari, Donelli, Massa, & Azaro, 2007; Jin, & Rahmat-Samii, 2007; Donelli, Martini, & Massa, 2009; Chamaani, S., Mirtaheri, S.A., & Abrishamian, M.S. 2011; Ismail, T.H., & Hamici, Z.M. 2010; Bevelacqua, P.J., & Balanis, C.A., 2009; Lin, C., Zhang, F.-S., Zhao, G., Zhang, F., Jiao, Y.-C. 2010) and control (Boeringer, & Werner, 2004; Khodier, & Christodoulou, 2005; Boeringer, & Werner, 2005; Donelli, Azaro, De Natale, & Massa, 2006), and other interesting practical applications (Adly, & Abd-EL-Haftiz, 2006; Ho, Yang, Ni, & Wong, 2006; Genovesi, Monorchio, Mittra, & Manara, 2007; Cui, & Weile, 2006; Azaro, Donelli, Benedetti, Rocca, & Massa, 2008; Martini, Donelli, Franceschetti, & Massa, 2008). Among them, the development of CAD tools, aimed at providing the designer with an environment where a given circuit or microwave device can be characterized, investigated and also modified to obtain desired requirements, represents a significant example. As far as the modification of a given component is concerned, a key point is the development of microwave CAD environments (Azaro, De Natale, Donelli, & Massa, 2006) where the device or the circuit is modified automatically in an unsupervised way by the microwave tool itself with an improvement of performance. Such kind of CAD tools can be used to reduce the time to market of devices, and microwave circuits, and keep the commercial predominance. The unsupervised CAD tool do not require expert microwave engineers and they are more efficient with respect to standard trial errors methodologies. Such useful automatic design tools have been the object of research since some years (Azaro, De Natale, Donelli, & Massa, 2006; Caorsi, Donelli, Massa, & Raffetto, 2002). The importance of the subject is widely recognized. In fact, it is well known that the design of new microwave devices and circuit are needed in several important areas, e.g., civil and military telecommunication systems, industrial and medical equipments. Standard microwave synthesis techniques are quite effective for the design of standard devices such as microwave filter, combiners, broad band coupler. Unfortunately, such approaches sometimes led to the development of impracticable, very expensive networks or unrealizable devices especially for very high frequencies. Moreover these techniques are not suitable when new technologies or materials such as meta-materials are considered. For these reasons, the design problem is usually recast as a global optimization problem. Formulated in such a way, the problem can be efficiently handled by the Particle Swarm Optimizer by defining a suitable cost function that represent the distance between the requirements and the obtained solution. In this framework and in recent years, many efforts have been devoted to the developments of unsupervised CAD tools for the control of smart antenna, the design and optimization of non-conventional antennas (Azaro, Donelli, Francheschini, Zeni, & Massa, 2006), and the development of passive microwave devices (Azaro, De Natale, Donelli, & Massa, 2006; Caorsi, Donelli, Massa, & Raffetto, 2002). Despite the successful results reported in scientific literature, the availability of effective CAD tools able to automatically develop microwave circuits and devices is still a challenge. In the following, Section 2, presents the solution methodology based on the use of PSO optimizer for the optimization and synthesis of microwave circuits and devices. A set of selected and representative synthesis results concerned microwave passive devices such as Hybrid ring, filters and frequency doublers are presented in Section 3. Section 4 presents a brief discussion concerning the use of parallel computation. Finally some conclusions are drawn (Section 5).