Antennas constitute the front ends of various electromagnetic (EM) communication and sensor systems. The design of antennas typically poses many challenges as their needs continue to broaden for applications in many current and new technologies (some of which may also require the use of new materials). The performance of antennas commonly has to not only meet the required EM specifications (on gain, bandwidth, polarization, etc.) but also satisfy cost, size, weight and other constraints. Electrically small and electrically large antennas (e.g., reflector systems and lenses), respectively, pose different analysis/design challenges. Papers dealing with such challenges in various antenna applications at microwave and millimeter wave frequencies are of significant interest within this special session, along with papers on more fundamental topics related to antenna theory and design. Additionally, theoretical developments in the analysis and design of large phased arrays continue to be of growing interest for a variety of modern applications in sensor technologies. Issues of scan blindness, grating lobes, mutual coupling effects, etc. are always of interest from both a theoretical and practical standpoint. Frequency scanned arrays and sparse arrays are also of interest. Papers addressing the above fairly general topics on arrays will fit especially well within this special session. Finally, antennas and arrays need to be placed on practical platforms thus raising questions of how the complex platforms can affect the performance of such antennas and antenna arrays. Papers addressing antenna/array placement issues, and in particular conformal antennas/arrays, will also be a focus of this session.

**Abstract:** In modern wireless communication system, the antenna array is an integral part to achieve the spatial diversity. The demand for large array antennas exhibiting increased capabilities and reduced cost and complexity is growing day by day. In view of this increasing demand of antenna arrays in radar and communication systems, the amount of research expended on the different aspects of array technology has grown many folds, in the recent past. Few of the major topics of this technology that have been focussed in by the research community, in recent days, are (i) design of massive MIMO arrays for 5G communication and its related signal processing, (ii) development of self recoverable arrays, (iii) Innovative array structure designs. This special session will include papers in all aspects of array antennas with special emphasis on its present and future applications.

**New and Efficient Techniques to Predict Radar Cross Section of Ship-like objects**

**Abstract**

In this tutorial, we present a new, ultrafast, and accurate method of moments (MOM) algorithm for the prediction of radar cross section (RCS) of Navy ships, subsystems, ship-board antennas, and other large objects.

The tutorial begins by exploring conventional MOM algorithm for arbitrarily shaped perfectly conducting (PEC). The procedure starts by approximating the perfectly conducting (PEC) target surface via planar triangular patches. Then, Rao-Wilton-Glisson (RWG) functions, defined over the triangular domains, are used in the MOM scheme to obtain a square matrix of size N that is more amenable to numerical processing. Note that N represents the total number of basis/testing functions in the numerical solution scheme and must be chosen such that there are 300-400 functions per square wavelength surface area. Also, note that the computer storage and processing time for the MOM solution are proportional to N2 and N3 respectively. For electrically large objects, N can be over a million and the storage and solution times are beyond the capabilities of present day computers. Several examples with detailed mathematical steps are provided for easy understanding.

Next, we develop a new procedure wherein the computer storage and CPU times are drastically reduced without sacrificing the advantages of MOM. In the new procedure, the first step involves approximating the given structure via the triangular patch scheme and using RWG functions as the basis and testing functions as in the conventional MOM solution. The next step involves gathering the basis functions into a small number of groups with size equal to M basis functions; thereby, casting the moment matrix into a collection of sub-matrices representing self- and mutual- interaction between the groups. In this scheme M is much smaller than N. Then, the procedure involves eliminating the interaction of two immediate neighbors in any selected group by generating a set of linear equations of size 2M X 2M and solving using a standard linear equation solver. This process results in a diagonally dominant moment matrix, assuming the group size is sufficiently large. It also sets the matrix blocks residing on either side of the diagonal block to zero. Although the new matrix equation can be solved efficiently in many ways, it is solved very efficiently using a power series approach.

The new approach is simple, efficient, and highly amenable to parallel processing while retaining all the advantages of the conventional MOM algorithm including multiple incident fields trivially. The procedure can be easily coupled with an existing MOM code very easily thereby increasing the capabilities to handle much larger problems. The storage required for this new algorithm is 3N*M as opposed to N2 for conventional MOM (MN). Parallel processing reduces the computation time drastically and near linear efficiency is achieved as more processors are used. With serial processing, the solution time varies as N2, versus N3 with conventional MOM. Several numerical results, along with detailed steps to easily generate the algorithm, will be presented.

**Biographical Sketch: **Dr. Rao is well known in the electromagnetic engineering community and included in the Thomson Scientifics’ **Highly Cited Researchers List.** This is a rare honor bestowed on the 250 most cited researchers in the world and is considered the most significant award given by any non-partisan group for contributions to a field of research (For details, see the website http://isihighlycited.com/isi_copy/howweidentify.htm). Furthermore, he received the prestigious Best Paper research award from the **SUMMA** foundation, awarded only once every three years for published research. He is the first individual to develop the triangular patch modeling technique that allows for the very accurate numerical solutions of several difficult electromagnetic scattering problems for the first time. These problems include the electromagnetic scattering from arbitrarily shaped conducting, dielectric and composite structures in the frequency domain and time domain.

Dr. Rao is the lead author in the classical 1982 paper on triangular patch modeling of arbitrary bodies. In this work, he developed a set of special basis functions to solve electrodynamics problem, which are popularly known throughout the world as RWG (Rao- Wilton-Glisson, the authors of the paper) functions. A casual Google search of RWG functions result in around 250,000 hits as of now. This paper established the now-standard method for calculating the radar cross section of complex conducting objects. Over the years, he continued to work with RWG functions in conjunction with triangular patch modeling techniques to solve frequency-domain dielectric body problems (1986), frequency-domain composite body problems (1991), time-domain conducting body problems (1991, 1992, 1998), time-domain dielectric body problems (1994, 1999), and cavity-backed aperture problems (1998). His most recent research has focused on using RWG basis functions and Genetic Algorithms to solve electromagnetic optimization problems (2007) and developing alternate techniques to address very large complex electromagnetic problems (2011, 2012).

The impact of Dr. Rao's research on electromagnetic community and industry has been tremendous. Because of his efforts, the methodology of design and analysis of several critical systems, both in defense and commercial sectors, and in many areas of research have completely changed. In a recent compilation, RWG functions have been used in more than 30 different and diverse fields. His algorithms enable the working engineer to design, evaluate, and test the product on the computer before fabrication. Furthermore, recent triangular patch modeling techniques are finding new applications. The well-known and troublesome instability problem associated with the time domain integral equation solutions have now been solved using an implicit solution method in conjunction with RWG functions.

Ramakrishna Janaswamy is a Professor in the Department of Electrical & Computer Engineering, University of Massachusetts, Amherst. His research interests include deterministic and stochastic radio wave propagation modeling, analytical and computational electromagnetics, antenna theory and design, and wireless communications. He is Fellow of IEEE and an elected member of U.S. National Committee of International Union of Radio Science, Commissions B and F. He is a recipient of the R. W. P. King Prize Paper Award of the IEEE Transactions on Antennas and Propagation and the IEEE 3rd Millennium Medal. He served as an Associate Editor of Radio Science, the IEEE Transactions on Vehicular Technology, the IEEE Transactions of Antennas and Propagation and the IET Electronics Letters. He is an IEEE Standards Activity member representing the IEEE Antennas and Wave Propagation Standards. He is the author of the book Radiowave Propagation and Smart Antennas for Wireless Communications, Kluwer Academic Publishers, November 2000, and a contributing author in Handbook of Antennas in Wireless Communications, L. Godara (Ed.), CRC Press, August 2001 and Encyclopedia of RF and Microwave Engineering, John Wiley and Sons, 2005.

Stochastic method, wherein the solution of a boundary value problem in electrostatics or electrodynamics is represented as an ensemble average of a stochastic process generated by the underlying partial differential equation, is very attractive in electromagnetics because (i) it permits the solution in any subregion of a computational domain without having to determine the field everywhere, (ii) the solution is amenable to complete parallelization, and (iii) the solution can be generated without an explicit mesh. We discuss here the basics of the stochastic formulation and apply the method to the solution of Poisson’s equation and the Helmholtz equation. The latter will involve examples below the first resonance for normal media and for any frequency in plasmonic media.