Leakage Properties of Stub-Loaded Ridge-Rectangular Waveguides

1. Introduction A leaky-wave line-source antenna is basically an open waveguide that supports a leaky mode (instead of a bound mode) and that therefore radiates all along its length. For use at millimeter wavelengths, various novel leaky-wave antennas were proposed, analyzed, and measured based on some low loss open waveguide designed specifically. Since these waveguides are already open, a physical cut to produce leakage is not meaningful and new mechanisms were tried for permitting the initially bound dominant mode to leak, the most common of which was the introduction of asymmetry. One of the most widely used of these open waveguides for antenna purposes is the stub-loaded rectangular waveguide, proposed by us[1-4] and consisting of a rectangular waveguide with an asymmetrically located stub guide. This guide can be fed by an ordinary rectangular waveguide source, radiate in a single polarization (with negligible cross polarization at all scan angles), possess flexibility in beamwidth and permit one to change the beamwidth without affecting the angle of maximum radiation. The only problem of this guide is that one must take more than a half wavelength in the width of a rectangular waveguide. This means that for use of a scanning array antenna, extra main beam due to a grating lobe is not avoided. Therefore, the width of the guide needs to be reduced to less than a half wavelength and then we have done it by using a ridge waveguide(see Fig. 1) , as D. Y. Kim and R. S. Elliott had introduced it for planar array of slots[5]. In this paper, we have investigated the effect of ridging stub-loaded rectangular waveguides, theoretically and experimentally. The waveguide structure proposed here is very complicated and includes many conductor edges at which the fields diverge. So it is hard to analysis it by the ordinary mode-matching technique due to slow convergence. To overcome this difficulty, we have proposed a new analytical method based on the efficient mode-matching technique developed by us[6][7], in which the singular aperture fields are assumed.