Abstract: Harris Hawks optimization (HHO) is a novel swarm intelligence optimization algorithm, which simulates the cooperative behavior of birds in nature. Dielectric waveguide filters are an urgently needed device for current 5G mobile communication equipment. Therefore, it is important to design and optimize dielectric waveguide filters efficiently and accurately by using the new algorithm optization. In this paper, HHO algorithm is first described, and a general framework combined with filter optimization problems is proposed. Then, the updating equations of HHO is derived based on the steady-state assumption. According to these equations, both dynamic characteristics and convergence behavior of HHO are analyzed. Finally, the HHO algorithm is used to design two dielectric waveguide filters to improve the convergence speed, efficiency and accuracy of filter design. To demonstrate the efficiency, HHO is compared with other three famous optimization techniques. The experimental results show that the HHO algorithm are significantly better in the convergence speed, efficiency and accuracy than the self-adaptive differential evolution (SADE), flower pollination algorithm (FPA) and gray wolf optimizer (GWO).