0018-926X (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information. This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TAP.2018.2851296, IEEE Transactions on Antennas and Propagation > REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 1 Abstract—We present the accurate modelling and analysis, followed by experimental validation, of a 1024 elements (64-by- 16) antenna array. This fixed-beam array radiates linear polarization in Ka-band (19.7-20.2 GHz). It acts as a first step in the design and modelling of future antenna arrays for Satcom- on-the-Move applications. Accurate prediction of the behavior of such a large array is a challenging task since full-wave simulation of the entire structure cannot be considered. By taking advantage of existing formalisms on periodic arrays and by using appropriated methods to efficiently exploit such formulations, it is possible to accurately define the performances of all building- blocks, from the feeding circuits to the radiating elements, over a frequency range. Such a detailed design also allows an accurate physical analysis. It has been successfully used to guarantee the measured performances. The paper is intended to detail the different steps to antenna designers. Index Terms—Ka-band, large antenna array, modelling accuracy, periodic boundaries, Satcom-on-the-Move. I. INTRODUCTION raditional satellites cannot manage the required data rates anymore [1]. As explained in [2], an increasing number of Ka-band High Throughput Satellites (HTS) is launched or is planned to be launched by many satellite operators to significantly improve the capacity [3]-[5]. Broadband multimedia satellite services were initially proposed in the Ku- band which uses wide beams to cover large territories. On the contrary, Ka-band HTS use multi-spot solutions which increase the available capacity through high-order frequency and polarization reuse [6]-[9]. Smaller beams imply higher satellite antenna gains and thus allow increasing the Equivalent Isotropic Radiated Power (EIRP) per beam [10]. Therefore, Ka-band is drawing more and more attention, especially for “on-the-move” communications such as Manuscript received June 2, 2017. A. Maati, M. Thevenot, C. Menudier, E. Arnaud, and T. Monediere are with the XLIM Research Institute of the French National Center for Scientific Research (CNRS), 123 rue Albert Thomas, 87060 Limoges Cedex, FRANCE (e-mail: marc.thevenot@xlim.fr) B. Lesur, C. Melle, D. Chaimbault, and A. Karas are with Zodiac Data Systems, Aerodrome d’Arcachon, 33260 La Teste de Buch, FRANCE (e-mail: benoit.lesur@zodiacaerospace.com). airplanes. On the aircraft side, a relatively high-gain antenna is needed to communicate with the satellite. Moreover, the antenna needs beam steering capabilities to compensate the moving of the aircraft. Consequently, an increasing effort was made to develop compact antenna terminals complying with these needs [11]. The first systems were used to work in the Ku- band with the previous generation of satellites [12], [13]. A few systems used reflector-based antennas, which are bulky and heavy [14]. Passive antenna arrays combined with a mechanical positioner are also widely used since they allow reducing the size of the system compared to reflector antennas [15]. Nevertheless, the footprint of that kind of system is still high and the lack of reactivity of mechanical beam steering is an important drawback. Therefore, a significant effort was made to develop phased array antennas systems which are low-profile and allow a fast electronic beam steering [16]- [18]. For this purpose, the company Zodiac Data Systems wishes to develop such systems for aircraft satellite communications. As a first step, it was decided to develop a high gain (30 dBi) passive printed antenna array working in the Rx sub-band of the Ka-band (19.7-20.2 GHz) with E- and H- plane half- power beamwidths of around 4° and 1°, respectively. Achieving such a high gain with a printed array is a challenging task. Indeed, the losses caused by the feeding network often prevent the gain increase when the size of the array becomes important [19]. Unlike waveguide-based arrays, a printed array is considered as a high gain antenna when reaching 15-25 dBi gain [20]. In this contribution, a multi-layer array of 1024 (64-by-16) printed radiating elements is designed to radiate a broadside beam in linear polarization. The technology used is not the most performant but it will be possible to upgrade it later. It represents a first step before the reconfigurable design, with the objective of validating accurately the modelling process of all building-blocks and their interactions, often neglected or approximated in classical designs. The accurate analysis of this kind of array is not a trivial task. The mutual couplings between the radiating elements need to be taken into account in order to achieve a proper modelling of the array [21]. It is indeed necessary to obtain the full scattering matrix of the radiating panel. As full-wave A Large Antenna Array for Ka-band Satcom-on- the-Move Applications – Accurate Modelling and Experimental Characterization Benoit Lesur, Amel Maati, Marc Thevenot, Cyrille Menudier, Member, IEEE, Eric Arnaud, Thierry Monediere, Christophe Melle, Member, IEEE, David Chaimbault, and Alain Karas T