Filing is concentrated at the top of the ranking: the five leading assignees together account for 30.7% of the 1,605 records in scope, and the top ten account for 44.2%. The single leader holds 168 records, well ahead of the fifth-ranked assignee at 58 and the tenth at 22. Beyond that point the ranking thins into a long tail of smaller filers, so due diligence should weight the leading names heavily but not assume the field is a two-company race.

In this dataset both fall under the same search scope because they solve the same problem — selective frequency filtering with controlled insertion loss and quality factor — using different physical resonating structures. Cavity filters rely on metal cavity geometry, while dielectric resonator filters load that cavity with a ceramic puck to raise unloaded quality factor and shrink size. Claim language in the corpus frequently addresses both together, particularly around tuning screws, spurious mode suppression and temperature drift compensation, which is why the search string spans all three filter types.

Filings rose from 71 in 2017 to a peak of 103 in 2021, then declined to 23 by 2024 — a complete-year drop of -78% over that three-year span. That is the most recent period that can be read reliably; 2025 and 2026 figures are still incomplete because publication typically lags filing by about 18 months. Treat the recent years as provisional rather than evidence of a further slowdown.

Semiconductor integration (H01L, 1.6% of the 1,605 records) and laser or stimulated-emission crossover (H01S, 1.4%) are the thinnest classes in the dataset, far below the 86.4% concentrated in core waveguide and microwave-element claims (H01P). Printed-circuit integration (H05K, 2.4%) is similarly under-filed. These branches represent the more open filing space relative to the dense core geometry claims.

US9705171B2, assigned to Lanteris Space LLC and filed 2017-07-11, covers a dielectric resonator filter where two cavities share a common wall carrying both a centrally located coupling iris and a second, larger peripheral aperture whose effective coupling area is adjusted by an externally accessible tuning screw. It blocks designs matching that specific dual-aperture, externally-adjustable-screw geometry, but does not extend to single-aperture couplings or to filters using other tuning mechanisms. Anyone designing around it should compare their coupling and adjustment mechanism against that claim language directly rather than assume broad coverage.