| At this stage,wireless communication resources are becoming more and more tense.It is becoming more and more urgent for the bandwidth and spectrum channel requirements.L ow-cost,miniaturized filters are gaining more and more popularity.In the context of 5G technology application,a folded cavity bandpass filter is designed,and the simulation and test results are basically consistent.Based on the basic theory of microwave filters,this paper discusses the classification of filters.The low-pass filter is used as a prototype.After frequency conversion and admittance transformation,it is converted into a filter that meets the design requirements.Then,the characteristics of the waveguide filter are introduced.The influence on the size factor of different shape cavities are used on the zero-pole.The coupling between the adjacent cavities and the input-output coupling are obtained.According to the above theory,based on the rectangular cavity,a four-cavity waveguide bandpass filter is designed,a new rectangular cavity is added at the input port to facilitate the access of the coaxial connector.Then a six-cavity waveguide bandpass filter is designed by using the folding technique.The filter has a center frequency of 3.5 GHz,a passband range of 3.28 GHz to 3.74 GHz,and a return loss of approximately-60 dB at the center frequency.It achieves a relative bandwidth of 13.1% compared to the four cavities before unfolding.And the five-cavity filter,with a wider bandwidth,basically meets the design requirements.Designing this filter has the following advantages:1.Without adding tuning screws or coupling screws,just adjust the size of the cavity to affect the distribution of zero and pole.2.By adding the ridge waveguide,it can reduce the size of each part.The ridge waveguide can be regarded as a high-pass filter,which realizes a wide bandwidth and effectively improves the coupling ability.3.The folding technology is adopted to greatly reduce the filter size and save processing costs.Finally,according to the simulation model,it is found that the results are not much different and basically coincide comparing the simulation results with the measured results. |
