An Ultra Low-Loss Silicon-Micromachined Waveguide Filter for D-Band Telecommunication Applications

Ahstract-A very low-loss micromachined waveguide bandpass filter for use in D-band (110-170 GHz) telecommunication applications is presented. The 134-146 GHz filter is implemented in a silicon micromachined technology which utilises a double H-plane split, resulting in significantly lower insertion...

Full description

Saved in:
Bibliographic Details
Main Authors: Campion, James, Glubokov, Oleksandr, Gomez, Adrian, Krivovitca, Aleksandr, Shah, Umer, Bolander, Lars, Li, Yinggang, Oberhammer, Joachim
Format: Conference Proceeding
Language:English
Subjects:
bandpass filter
D-band
Frequency measurement
Insertion loss
Loss measurement
micromachined waveguide
Microwave filters
Silicon
telecommunication
waveguide filter
Wireless communication
wireless link
Citations: Items that cite this one
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Ahstract-A very low-loss micromachined waveguide bandpass filter for use in D-band (110-170 GHz) telecommunication applications is presented. The 134-146 GHz filter is implemented in a silicon micromachined technology which utilises a double H-plane split, resulting in significantly lower insertion loss than conventional micromachined waveguide devices. Custom split-blocks are designed and implemented to interface with the micromachined component. Compact micromachined E-plane bends connect the split-blocks and DUT. The measured insertion loss per unit length of the waveguide technology (0.008-0.016dB/mm) is the lowest reported to date for any micromachined waveguide at D-band. The fabricated 6-pole filter, with a bandwidth of 11.8 GHz (8.4 %), has a minimum insertion loss of 0.41 dB, averaging 0.5 dB across its 1 dB bandwidth, making it the lowest-loss D-band filter reported to date in any technology. Its return loss is better than 20 dB across 85 % of the same bandwidth. The unloaded quality factor of a single cavity resonator implemented in this technology is estimated to be 1600.
ISSN:2576-7216
DOI:10.1109/MWSYM.2018.8439601