Remove the remaining legacy rig_id aliases across the HTTP frontend and use remote consistently for scheduler and audio requests.
Disable browser caching for the HTML, CSS, and JS assets so clients pick up the parameter rename immediately instead of running stale frontend code.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Parse the renamed `remote` query parameter on `/audio` while keeping
`rig_id` as a compatibility alias, and use per-rig stream info for
rig-scoped audio subscriptions.
Add tests covering both query names.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Keep global state and spectrum updates scoped to the server group that
owns the selected short name, so other servers cannot overwrite the UI
or clear the active spectrum stream.
Add regression tests for cross-server selection and spectrum ownership.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Merge per-server GetRigs updates into the shared remote rig cache
instead of replacing it, so audio tasks from other servers are not
torn down on each poll cycle.
Add a regression test covering the multi-server cache update path.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Use full audio endpoint URLs for trx-server audio connections while
preserving server-advertised ports and legacy port-based fallback for
backward compatibility.
Add `server_url` and per-remote `rig_urls` config entries, plus
validation and tests for audio URL parsing and address resolution.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Rename HTTP query params, JSON fields, and scheduler payloads to
use remote names consistently while still accepting legacy `rig_id`
inputs through serde aliases.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Use the remote short name for the HTTP frontend default selection and
keep `default_rig_id` as a serde alias for backward compatibility.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Introduce [[remotes]] config array where each entry maps a user-chosen
short name to a (server URL, rig_id) pair. Short names replace raw
rig_ids as the universal key throughout frontends, audio routing, and
state management, allowing rig_ids to safely collide across servers.
Entries sharing the same server URL and token share a single TCP
connection. A request routing dispatcher forwards frontend commands to
the correct per-server channel based on the short name.
Legacy [remote] config and CLI --url are preserved via automatic
fallback to a single-entry remotes list.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Include active rig display name and nearest city in the browser tab
title: "freq - rig name - city, country - trx-rs v<version>".
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Add checkbox to enable/disable NB and number input for threshold (1-100).
Controls are hidden by default and shown when the server reports NB support
via SSE filter state updates.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
IQ-domain impulse noise blanker using exponential-smoothing RMS tracker. Samples exceeding threshold × running RMS are replaced with the last clean sample. Configurable via [sdr.noise_blanker] in TOML config and runtime via POST /set_sdr_noise_blanker API.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
The bookmark color palette is derived from CSS variables (--accent-yellow,
--accent-green, etc.) which change on both dark/light theme toggle AND
palette style change (arctic, lime, etc.). The previous fix only covered
setTheme; extract invalidateBookmarkColors() and call it from setStyle
as well so bookmark chips recolour on any visual change.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
getComputedStyle may return stale CSS variable values if the browser
has not flushed the style recalculation after changing data-theme. Force
a recalc by reading a property value first. Also clear cached bookmark
DOM keys so the next draw pass rebuilds chips from scratch.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Bumping bmRevision and scheduling a spectrum draw was not enough because
the spectrum draw path only runs when spectrum data is present. Instead,
directly update --bm-cat-bg/--bm-cat-fg on all existing bookmark chips
from the new theme palette so colors update immediately.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
scheduleSpectrumDraw references a let-bound variable that hasn't been
initialized yet when setTheme runs at startup. Wrap in try/catch so the
early call is silently skipped.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
send_command only updated the global state_tx watch channel, but SSE
sessions subscribe to per-rig rig_states channels. Per-rig channels were
only updated during the poll cycle (default 750ms). Now send_command also
pushes state to the per-rig channel immediately, eliminating up to 750ms
of latency for confirmed frequency/mode/state changes.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
The vchan setRigFrequency wrapper was awaiting vchanTakeSchedulerControl()
(HTTP PUT to /scheduler-control) and vchanSetChannelFreq() (HTTP PUT to
channel freq endpoint) before calling the original setRigFrequency. This
added a full HTTP round-trip of latency to every frequency change. Make
both fire-and-forget: optimistic local update happens first, network calls
run in background.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
The HTTP round-trip for set_freq blocks on the server processing the
command (mpsc → TCP → rig hardware → response). With optimistic local
updates, CSS overlay, and SSE snap-back guard already in place, there is
no reason to await the network call. All callers (jog, freq input, spectrum
click, RDS AF tune) now return immediately after the optimistic update.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
When the user changes frequency, applyLocalTunedFrequency sets lastFreqHz
optimistically. But the SSE state stream could push back the old server
frequency before set_freq completes, causing the marker to snap back then
forward. Add a sequence-guarded _freqOptimisticHz that suppresses stale
SSE frequency updates while a set_freq is in flight.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Replace synchronous drawSignalOverlay() calls in freq/BW change handlers
with lightweight CSS div elements repositioned via transform: translateX().
This is GPU-composited with zero layout/paint cost, making frequency and
bandwidth changes appear instantaneous. The full WebGL overlay catches up
on the next requestAnimationFrame.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Call drawSignalOverlay() synchronously on frequency and bandwidth changes instead of deferring entirely to requestAnimationFrame. Also make bookmark apply fire-and-forget so the click handler returns immediately after optimistic UI updates.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Replace single-connection relay with per-rig audio manager that spawns independent TCP connections for each rig. Each rig gets its own broadcast channel, stream info, and vchan command routing. Selected rig mirrors to global channels for backward compat. Also fix bookmark apply to update spectrum marker instantly and fire all requests in parallel.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
When ?rig_id= is specified on /audio, don't fall back to the global broadcast (which carries whichever rig is connected). Return 404 for rigs without an active audio connection instead of silently delivering the wrong rig's audio. Also create per-rig audio channels for all known rigs eagerly so connected rigs are instantly subscribable.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Channel SSE events were broadcast to all tabs regardless of rig, causing tabs to display wrong rig's channels. Per-rig audio info_rx override caused WebSocket to hang waiting for stream info that never arrives.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Each browser tab can now subscribe to a specific rig's spectrum and
audio independently via ?rig_id= query params on /spectrum and /audio.
The remote client polls spectrum for all rigs with active subscribers
and routes responses to per-rig watch channels. Virtual channel
commands are routed through per-rig senders with global fallback.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
When a new tab connects and receives the initial channels event,
automatically subscribe to channel 0 (primary) so the session joins
the same tuned channel as other users on that rig. Uses a lightweight
auto-join that skips scheduler control takeover since audio isn't
started yet at this point.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
select_rig no longer mutates remote_active_rig_id — only the per-session
mapping is updated. This eliminates cross-tab leaking entirely: each tab
carries its own rig_id via the session manager, /events?rig_id, and
/status?rig_id query params.
The global remote_active_rig_id now only serves as the startup default
for brand-new sessions that have no rig_id yet.
Also fix the About section to show the per-tab lastActiveRigId instead
of the server's global active_rig_id.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
pollFreshSnapshot() fetches GET /status on every SSE connect/reconnect,
but /status always returned the global selected rig's state, overwriting
the per-tab display with whichever rig was last switched to globally.
Now pollFreshSnapshot passes rig_id as a query param and the /status
endpoint uses the per-rig watch channel when provided, matching the
/events behavior for true per-tab rig isolation.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
The previous commit conditionally skipped updating remote_active_rig_id
when session_id was provided, but the remote client reads the global to
route commands to the correct rig on trx-server. Restore the
unconditional global update; cross-tab SSE isolation is handled by the
rig_id query param on /events and the JS-side guard in applyRigList.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
After select_rig stopped mutating the global remote_active_rig_id for
session-aware clients, SSE reconnects would fall back to the old global
default instead of the newly selected rig. Now connect() passes
lastActiveRigId as a rig_id query param to /events, and the server
prefers it over the global default when present.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
The .audio-active state was using --accent-green which is actually
orange (#c24b1a). Match the regular play button's hardcoded #00d17f
green so the header button visually matches.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
select_rig was unconditionally updating the global remote_active_rig_id,
causing all SSE sessions to see the changed rig. Now only the per-session
mapping is updated when session_id is provided; the global default is
only changed for non-session-aware clients.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Bump selector specificity to .header-bar-btn.header-audio-btn so the
padding: 0 rule wins over the generic .header-bar-btn padding, and
switch the SVG to width/height: 100% so it expands to fill the button.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Add per-rig watch::Sender<RigState> map to FrontendRuntimeContext,
populated by refresh_remote_snapshot for every rig returned by GetRigs.
The SSE /events endpoint now subscribes to the session's rig-specific
watch channel instead of the single global one, allowing different
browser tabs to independently view different rigs. The JS frontend
reconnects SSE on rig switch to pick up the new channel.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Add SessionRigManager to track per-SSE-session rig_id so different
browser tabs can independently select rigs without interfering.
The /events SSE stream filters state updates by session rig (falling
back to the global active rig), and /select_rig accepts an optional
session_id to update the per-session mapping.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Rig switch needs the server call so SSE/audio follow the selected rig. Play button now uses a fixed triangle icon sized to match header controls.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
The /select_rig endpoint sets global server state which affects all tabs. Since postPath() already sends rig_id with every command, the rig picker now just sets the local lastActiveRigId variable.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
postPath() was duplicating rig_id on /select_rig calls, causing deserialization failure and silently dropping the rig switch.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
All POST command endpoints now accept an optional ?rig_id= parameter so each browser tab can independently target a specific rig. The JS frontend tracks the active rig per-tab and auto-appends rig_id to every request.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Add checkbox column to bookmark table with select-all support and a
Delete Selected button for batch removal. New POST /bookmarks/batch_delete
API endpoint accepts an array of IDs and removes them in one request.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Broadcast virtual channel list for newly selected rig from select_rig so SSE
clients receive correct channels immediately. Detect rig changes in render()
and reset stale decoder state (RDS, spectrum, decoder status indicators) from
the previous rig.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Add Maidenhead locator and reverse-geocoded city/country to the header.
Uses Nominatim API to resolve nearest city asynchronously.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
When Device::new(args) fails, enumerate all available SoapySDR devices and
include them in the error message. Also hint that args are case-sensitive.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
When specific SoapySDR device args are provided (e.g. with a serial number),
fail hard instead of silently falling back to Device::new("") which opens
the first available device. This caused multi-device setups to bind both
rig instances to the same physical device.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Add weakest decoded signal panel showing top 5 weakest SNR signals. Make all
stat tiles (longest decode, strongest signal, weakest signal) clickable to
highlight the corresponding locator on the map.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Reflect the common/ft8/ft4/ft2 directory reorganization in the
architecture diagram, file tree, and signal flow description.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Move ft2_encode from ft4/ to ft2/ where it belongs. Remove all
module-level #[allow] suppressions and fix the underlying issues:
- Remove dead code: wf_mag_at, xor_rows, unused Monitor IFFT fields, OsdBox.size
- Gate encode174_to_bits with #[cfg(test)] (only used in tests)
- Convert 40+ C-style index loops to idiomatic iterators
- Add targeted #[allow(clippy::too_many_arguments)] on two OSD functions
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Move ft2/osd.rs, ft2/bitmetrics.rs, ft2/downsample.rs, ft2/sync.rs
out of the ft2/ directory into src/ as top-level modules. Convert
ft2/mod.rs to ft2.rs. Update all imports from super:: to crate::ft2::.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Replace FTX_CRATE.md with README.md documenting upstream origins
(kgoba/ft8_lib for FT8/FT4, iu8lmc/Decodium-3.0 for FT2) and a
Mermaid diagram of the crate architecture.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Remove normalize_llr which was undoing the scalefac=2.83 scaling,
causing LLRs to be 2.83x too small for the BP+OSD decoder. Align
sync thresholds with reference: coarse 0.50->0.40, decode 0.65->0.55,
sync quality 10->9, maxosd 3->4. Revert norm_sqr back to norm in
bitmetrics since the metric difference is nonlinear.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Replace Vec<Vec<f32>> with flat stack arrays in ldpc_decode (~114KB),
convert 19+ Vec allocations to stack arrays in osd174_91, eliminate
per-call temp Vec in nextpat91 via in-place mutation, and replace
norm() with norm_sqr() in bitmetrics hot loop (~5.4M calls/frame).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Eliminate duplicated code between FT2 and FT8/shared modules:
- Share parity8() from encode.rs, remove copies in ft2/mod.rs and osd.rs
- Share pack_bits() from decode.rs, remove pack_bits91() from osd.rs
- Add verify_crc_and_build_message() to decode.rs, used by both FT8 and FT2
- Add normalize_llr() to decode.rs, replacing per-module normalization
- Make encode174() pub(crate), add encode174_to_bits() for bit-array output
- Wire FT2 decode_hit to use full BP+OSD decoder from osd.rs instead of
separate BP + sum-product + OSD-lite flow
- Align LLR scale factor to 2.83 matching reference implementation
Net -178 lines removed.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
add() and lookup() had no wrap-around guard in their linear-probe loops.
Once 256 unique callsigns filled the table, any subsequent add or lookup
for an absent hash would cycle through all 256 slots forever, hanging the
FT8 decoder task permanently inside block_in_place. On a busy band this
could happen within a few minutes of operation.
- add(): evict the probe-start slot when a full cycle completes
- lookup(): return None after a full probe cycle
- reset(): call cleanup(10) each slot boundary to age out stale entries
- Add regression tests for both infinite-loop scenarios
Also includes cargo fmt reformatting of pre-existing style issues.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Reuse FT2 downsample and bitmetric work buffers, speed up\nsync2d_score with precomputed references, and cache peak-search\nFFT state on the pipeline.\n\nCo-authored-by: OpenAI Codex <codex@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Delete the obsolete ft8_lib submodule and update documentation to point at the pure Rust trx-ftx decoder.
Co-authored-by: OpenAI Codex <codex@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Normalize tracked SPDX headers to the 2026 Stan Grams identity.
Co-authored-by: OpenAI Codex <codex@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Quiet compiler and clippy warnings in the translated decoder modules.
Co-authored-by: OpenAI Codex <codex@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Delete trx-ft8 (C wrapper around ft8_lib + ft2_ldpc) and update
trx-server to depend on trx-ftx (pure Rust) directly. Removes
~2,900 lines of C code and all unsafe FFI.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Replace the C FFI-based trx-ft8 with a pure Rust implementation
supporting FT8, FT4, and FT2 protocols. Eliminates cc/libc build
dependencies and all unsafe FFI code while providing the same
Ft8Decoder/Ft8DecodeResult public API.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
The bidirectional check required both A→B and B→A directed messages to
draw a contact line, which was too strict — the receiver may not hear
both sides of a QSO. Now a decode path is drawn whenever a directed
message is decoded and the target's locator is known from any message
in the 24 h history window.
Also rename "Longest QSOs" → "Longest decode paths" and update
related UI labels to better reflect what is actually shown.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
A QSO requires both parties to hear each other. Previously the map drew
contact paths whenever A sent a directed message to B and B's locator
was known from any message. Now paths only appear when both A→B and B→A
directed messages are present in the decoded history.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Add hardware AGC on/off control for SoapySDR backend, wired through the
full stack from RigCommand to the web UI:
- RigCommand::SetSdrAgc(bool) + ClientCommand::SetSdrAgc in protocol
- set_sdr_agc() on RigCat trait (not-supported default)
- SoapySdrRig: agc_enabled field, set_sdr_agc() via pipeline agc_cmd,
sdr_agc_enabled in filter_state(); removes the "not yet implemented"
warning — gain_mode="auto" now properly enables hardware AGC via
SoapySDR set_gain_mode()
- IqSource::set_gain_mode() trait method; RealIqSource implements it
- SdrPipeline: agc_cmd channel, read loop applies it each iteration
- POST /set_sdr_agc endpoint in trx-frontend-http
- New "SDR settings" full-row in index.html with Hardware AGC checkbox
and RF Gain (moved out of WFM controls); row hidden when
show_sdr_gain_control is false
- app.js: AGC checkbox handler, disables RF gain input while AGC is on,
syncs checkbox state from filter.sdr_agc_enabled
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Three issues caused audible distortion on AM reception:
1. DC blocker shared r=0.9999 (τ≈1.25 s at 8 kHz) across all modes.
For AM the envelope detector outputs A_c+m(t) — always positive —
so the blocker needs to track the carrier bias quickly. AM now uses
r=0.999 (τ≈125 ms), 10× faster, while keeping the highpass cutoff
below 2 Hz so speech is unaffected.
2. Audio AGC time constants were inverted relative to good AM AGC design:
attack=200 ms (should be fast to prevent overload) and
release=3500 ms (unreasonably sluggish). Changed to attack=5 ms /
release=200 ms, target=0.5, max_gain=36 dB.
3. No IQ AGC before envelope detection meant carrier amplitude variation
went directly into the audio chain, forcing the slow audio AGC to
handle both RF level and audio level simultaneously. Added an AM IQ
AGC (attack=0.5 ms, release=50 ms, target=0.7, max=30 dB) that
normalizes carrier power before demod_am, so the DC blocker always
sees the same steady-state bias regardless of signal strength.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Apply #ft8-messages container style (border, rounded corners, monospace
font, max-height with scroll) to #ft4-messages, #ft2-messages, and
#wspr-messages which were missing it.
Add #ft4-decode-toggle-btn and #ft2-decode-toggle-btn to the narrow-
screen white-space:nowrap media query rule alongside FT8/WSPR.
Cap DOM rows rendered per history view to 200 (FT8_MAX_DOM_ROWS,
FT4_MAX_DOM_ROWS, FT2_MAX_DOM_ROWS). Full history is retained in
memory; only the DOM representation is bounded. This prevents tab
switching from becoming sluggish after a long decode session where
thousands of rows accumulate in the DOM.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Three bugs caused USB/LSB to sound like AM:
1. The IQ low-pass filter was symmetric (passband ±BW/2), so both
sidebands were passed equally — taking .re then produced DSB-SC
rather than SSB audio.
2. cutoff_hz was computed as bandwidth_hz/2, halving the usable audio
bandwidth (1500 Hz for a 3 kHz USB channel).
3. demod_lsb claimed spectrum inversion was "handled upstream by
negating channel_if_hz", but that negation was never applied; USB
and LSB were functionally identical.
Fix: add a shift_norm parameter to build_fir_kernel / BlockFirFilterPair
that complex-modulates the time-domain FIR coefficients by
e^{j·2π·shift_norm·n}, shifting the passband in the frequency domain.
A new ssb_shift_norm() helper returns +cutoff_norm for USB/CW/DIG
([0, BW] Hz passband) and -cutoff_norm for LSB/CWR ([-BW, 0] Hz
passband); all other modes get 0.0 (symmetric LPF unchanged).
After the one-sided filter, taking .re correctly reconstructs the
selected sideband. No IF negation is needed for LSB.
Also fix two unit tests missing the force_mono_pcm argument introduced
after they were last updated.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
The OSD-3 (triples) path over 5 LLR passes was doing ~11,600 CRC checks
per candidate. With a 14-bit CRC this gives ~0.7 expected false positives
per candidate — far too high.
Remove OSD-3 entirely. Cap max_candidates at 16 for OSD-1/OSD-2, giving
136 CRC checks per pass (680 total). Gate OSD-lite behind a check that
LDPC reached within 6 parity errors of converging, so it only fires when
the LLRs are already trustworthy. Combined false-positive rate drops to
~0.04 per near-miss candidate.
Also remove the now-unused ft2_osd_decode and ft2_codeword_distance
functions.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Increase BP/SP iteration count from 30 to 50 to match WSJT-X reference
and give belief propagation more opportunities to converge near-threshold
candidates.
Replace the parity-based OSD-1/OSD-2 fallback (which required LDPC to
have nearly converged) with ft2_osd_lite_decode applied to all five LLR
combination passes. The CRC-based decoder works directly from raw LLRs
without depending on LDPC convergence, searching the 24 least-reliable
systematic bits for up to three bit errors via OSD-3.
Also increase max_candidates in ft2_osd_lite_decode from 12 to 24 for
broader coverage of likely error positions.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Diagnostic logging showed the FT2 BP/SP decoders consistently reach
1-8 residual parity errors rather than zero — the LLRs are correct
in direction but LDPC belief propagation stalls just short of
convergence.
Add ft2_osd_decode() implementing Ordered Statistics Decoding orders
1 and 2: after the five-pass BP/SP loop fails, sort the 174 codeword
bits by |LLR| ascending and trial-flip single bits (OSD-1, always)
or all pairs of the 50 least-reliable bits (OSD-2, when the remaining
error count is <= 4). Each trial costs one O(83) parity check;
worst-case overhead is ~1300 checks per candidate, negligible next to
the 5 x 30-iteration BP/SP passes already performed.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Add err=N/N/N/N/N to the FT2 window diagnostic log line, showing the
minimum number of unsatisfied parity equations across all candidates
for each of the five LLR passes. This makes it possible to distinguish
between a signal-quality-limited failure (small error count) and a
systematic decoder bug (large error count), which is the key unknown
in diagnosing the current FT2 LDPC non-convergence.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Normalize FT2 log-likelihoods before LDPC and fall back to\nthe standard waterfall candidate decoder when the raw FT2\npath produces no decodes.\n\nCo-authored-by: OpenAI Codex <codex@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
sjgandOpenAI Codex