Surface wave (SW) excitation is typically regarded as one of undesirable mechanisms in antenna design. Techniques are available to estimate the SW contribution on unwanted coupling between antenna elements, diffraction from the edges of the substrate and scan blindness in phased arrays. Trade-off between the efficiency and bandwidth in some applications forces designers to use thicker substrate to increase bandwidth. However, in some applications, the main goal is to take advantages of SW with least possible leakage or radiated power. For example, bound SW power can be efficiently coupled into theTM_0W mode of a grounded dielectric slab (GDS) as a feeding mechanism in leaky-wave antennas.This concept is recently examined in SW based devise such as SW lenses, SW power combiners and SW-fed antennas. In previous works, the SW excitation problem on a GDS by means of slots in the ground plane is considered as a two-dimensional (2D) model to extract closed forms for the power launched as SWs, power leaked as radiation, and input admittance of a slot source. Design of surface-wave launchers (SWLs), in 2D model is based on choosing optimum slab thickness to maximize the surface-wave power excitation and to optimize impedance matching simultaneously. In this framework, However, 2D model only considers first moment of excitation and gives no vision or guarantee in terms of SW power confinement and SW power transmission. Theoretical considerations in this thesis reveals that, regardless of source position and polarization, analyzing artificial impedance surface intrinsic properties such as reflection phase and surface impedance are sufficient to identify the optimum excitation and power transmission in a SW waveguide (SWG). A systematic design strategy is presented which describes the implementation of SWGs using SWLs. This design strategy is based on a hard surface concept that simultaneously guarantees mode purity, launching efficiency, SW power confinement and preventing backward SW propagation. In compare with full-wave evaluating SW excitation performance of a slot antenna near the GDS, this approach is computationally more efficient. In addition to confirming the 2D model’s precise results, the proposed technique paves the physical interpretations and the way for the optimum design of SWGs. By interpretation of intrinsic properties of GDS, we show that, due to improper boundary conditions of GDS, optimum SW confinement is not possible. We modify the boundary conditions to hard surface and demonstrate a SWG based on bow-tie type self-complementary metasurface. Main idea in the proposed approach is that a SWL is intrinsically an antenna. Generally, the hard surface provides maximum mutual coupling between arbitrarily polarized antennas located on the same ground plane. So for maximum power transmission between two SWLs, ground plane must be hard surface. Also, challenge in pure mode excitation comes from the fact that cross polarization of SWL is not zero and in practice, the excitation of pure mode is impossible. For a specific$TM_0$WL, one can see that although TE radiation/excitation due to a slot antenna in main beam is negligible, but depend of natural properties of SWL, in some angles even it could be dominant. Hard surface by definition against TE waves acts like a high impedance surface, so there is no possibility for excitation and propagation of TE waves. The third challenge is the transverse and longitudinal components of non-TEM, TE, and TM modes in the slab which have sinusoidal characteristics. This dependency in the third dimension questions the validity of traditional 2D models such as Rotman lens equations for SW based lenses. In hard waveguides, due to the reduction of thickness and excitation of TEM mode transmission, sinusoidal dependency is eliminated. According to the proposed concept, a novel compact broad-band SWL based on quasi Yagi–Uda slot antenna is presented. In the second step, a novel SW Rotman lens based on the proposed hard surface is designed and fabricated. In comparison to a conventional microstrip Rotman lens with similar size, the proposed hard SW lens shows higher scanning capability with the same SLL performance. The new lens is expected to show lower mutual coupling than the conventional lenses at millimeter-wave frequencies. Keywords: Artificial impedance surfaces, Rotman lens, Reflection phase, Hard surface, Surface-wave launcher