Japan HF routes in BR IFIC

Checked against the local database on 2026-09-20: 170 cached extracts, issues 2911–3080, published from 2020-01-07 through 2026-09-15. This note records the Japanese routes, their notification history and the SQL used to examine them. Company connections are in operator identities; physical installations are in the dossiers.

Result

The seven attributed Japanese transmitters produce 15 outbound endpoint pairs, after frequencies and emissions are grouped by receiver. DRW's Minamihara installation is receive-only and has no Japanese transmitter in this selection.

These are the ITU coordinates. BR IFIC generally labels the transmitters CHIBA and often omits their callsigns; the established transmitter-coordinate associations supply the operator names. The source country codes include J, G, MLA, INS and SNG, rather than ISO country codes.

Publication chronology at operator level

The 376 outbound Part I assignments have corresponding favorable Part II REX findings. Of those findings, 238 explicitly have fdg_status=FINAL; the older 138 leave that field empty. The table retains the outcome references for further inspection.

received_date comes from fxm_terra.d_rcv. Publication dates describe the BR IFIC issues; neither date establishes when equipment was built or began operating.

Japanese domestic record compared with ITU

Ryota's 23 August 2026 comparison and the saved MIC responses supply the Japanese HF callsigns and licence destinations:

• Raft: Chicago, London and Singapore in the Japanese comparison; Chicago and London in BR IFIC.
• Tower: Hong Kong, Batam, Johor, Seattle and Singapore in the Japanese comparison; Hong Kong, Batam and Seattle in BR IFIC.
• McKay: Singapore and Hong Kong in BR IFIC, with no full Japanese HF licence or public callsign yet reported in that August account. The Yagi had already been mounted at Kanozan.
• IMC: JS2RA serves the Chicago area and JS2RB the Philippines. This supplies the callsign distinction absent from the modified ITU rows.
• Optiver: the saved MIC response lists Johor Bahru, Indiana and London, while BR IFIC supplies only the Johor receiver. Ryota's June licence copy and August comparison list Johor alone. The amendment date and additional receive coordinates are unknown.
• NLN and FalconION: the Japanese destinations agree with the ITU endpoint sets above.

The September 17 MIC FX/ML data concern domestic microwave licences. Channel and municipality comparisons are in the JOSADA (McKay) and NuVo (DRW), GT Technologies (Optiver), Ermis (AlberBlanc/FalconIon) notes.

Japanese HF licence sources

The operator reference lists the seven MIC detail pages, their saved September 11 responses, and the connections between licensees, callsigns and physical sites.

JB2XI's licence changed: Ryota's June 2025 copy gives Johor alone, 2 kW and 送信所 (transmit installation). The saved September response gives Johor, Indiana and London, 5 kW and 送受信所 (transmit/receive installation). His August comparison still lists Johor alone. The saved response does not give the amendment date or additional receiver coordinates.

FCC endpoint joins

Application numbers, the other U.S. facilities and company relationships are in the FCC correlation.

Reverse/inbound search

The September 11 geographic search found one inbound route in its Chiba/Kanto selection:

Essex County, Ontario → Chikura, Canadian callsign CGA984. The transmitter is 42.0419444444444, -82.9741666666667; its Japanese receiver is 34.9630555555556, 139.936944444444, the same coordinate used for NLN's ITU transmitter, rather than Gotenyama.

Its publication history, rechecked in the current cache:

The outbound query below does not search foreign transmitters. To inspect this inbound record, join a primary Canadian TX to an associated_rx endpoint at the Japanese coordinate, retaining source_issue in both joins.

Reproducing the outbound route slice

The maintained query is sql/queries/japan_routes.sql. It takes $transmitters, a list of {id, latitude, longitude} structs from competitors.itu_tx in japan-network.toml. This identifies the seven transmitters; receiver locations and notification batches come from the database.

The query:

1. Joins hf_records to primary Japanese TX and associated RX rows in hf_endpoints, using (source_issue, terrakey) throughout.
2. Matches the attributed transmitter coordinates with a 1e-8 degree tolerance for their decimal representation.
3. Selects NTFD_RR Part I entries through fxm_pub_hist and pub, requiring pub.circ_no = source_issue. An extract also contains linked older records, so presence in the extract alone does not date a notification.
4. Groups by operator, receiver country and receiver coordinates. Frequency and emission variants collapse into one pair; distinct issue/date/intent batches remain in publications. The receiver label comes from its latest selected appearance.

After the database setup, run from the repository root:

uv run --locked python - <<'PY'
import json
import os
import tomllib
from pathlib import Path

import duckdb

research = tomllib.loads(Path("config/japan-network.toml").read_text())
transmitters = [
    {"id": item["id"], **item["itu_tx"]}
    for item in research["competitors"] if "itu_tx" in item
]
data_dir = Path(os.environ.get("BRIFIC_DATA_DIR", "/data/brific"))
with duckdb.connect(data_dir / "derived/brific.duckdb", read_only=True) as db:
    db.execute("LOAD spatial")
    result = db.execute(
        Path("sql/queries/japan_routes.sql").read_text(),
        {"transmitters": transmitters},
    )
    columns = [column[0] for column in result.description]
    print(json.dumps([dict(zip(columns, row)) for row in result.fetchall()], indent=2))
PY

This is notification history across the cache, not a reconstructed current assignment register. The query does not follow modification targets, remove retired assignments or calculate Part II/III outcomes. Those require the additional source tables below.

Following assignment changes and outcomes

For IMC's 2023 change, issue 2999 uses the older MDB layout: v1.fxm_terra.target names the previous terrakey, also exposed as fxm_terra.target in the combined view. The twelve modified records, 92733375–92733386, target the Kane records for six frequencies: 6.864, 9.081, 10.845, 12.117, 14.673 and 16.119 MHz, each with D1D and G1D emissions. Joining old and new associated_rx endpoints shows Kane changing to the Philippines. Other Kane assignments are outside that modification.

The newer layout moves target references into fxm_target. For example, Tower's issue-3070 notice 93047740 targets 92990746; comparing their receivers shows Seattle added alongside Hong Kong and Batam. Join the target within the same source_issue, where the extract carries the earlier record as context.

For publication outcomes, select Part II or III in fxm_pub_hist, join its publication through pub as above, and inspect fxm_fdg.fdg_type = 'REX', is_favorbl and fdg_status. Match an assignment across publication appearances using terrakey, assgn_id and intent. The source row's intent=ADD can persist through later outcome publications; intent alone does not identify the publication stage.

The data design documents the underlying views. The app README describes how the running map consumes the outbound query.