@neilberkman/sidereon 0.23.0 → 0.25.0

package/README.md

@@ -99,9 +99,45 @@ console.log(solution.rxClockS);  // receiver clock bias, seconds
 ```
 
 `solveRtkFloat` / `solveRtkFixed` and `solvePppFloat` / `solvePppFixed` follow
-the same shape: an options object in, a result object with `Float64Array`
+the same pattern: an options object in, a result object with `Float64Array`
 positions and scalar attributes out.
 
+## Example: post-solve integrity
+
+Use `raim` on per-satellite post-fit residuals after a solve. The direct result
+has `faultDetected`, `testStatistic`, `threshold`, `worstSat`,
+`reducedChiSquare`, `normalizedResiduals`, `rmsM`, and `dof`. RAIM residual
+tests must use per-satellite residual variances; unit weights on metre-scale
+residuals make `faultDetected` saturate near 100%. The JS API takes
+inverse-variance weights, so compute them from your variance model.
+
+```js
+import { RaimWeights, raim } from "@neilberkman/sidereon";
+
+const usedSats = ["G01", "G02", "G03", "G04", "G05", "G06"];
+const residualsM = [0.2, -0.1, 0.3, 0.2, 9.0, -0.2];
+const elevationDeg = [72, 42, 35, 64, 50, 28];
+const sigma0M = 0.8;
+const weights = Float64Array.from(
+  elevationDeg.map((el) => {
+    const sinEl = Math.max(Math.sin((el * Math.PI) / 180), 0.2);
+    const varianceM2 = (sigma0M / sinEl) ** 2;
+    return 1 / varianceM2;
+  }),
+);
+const integrity = raim(
+  { usedSats, residualsM },
+  { pFa: 1e-3, weights: RaimWeights.bySatellite(usedSats, weights) },
+);
+
+console.log(integrity.faultDetected, integrity.testStatistic, integrity.worstSat);
+```
+
+Use `araim(geometry, ism, allocation)` for protection levels from line-of-sight
+geometry and an integrity support message. `araimLpv200Allocation()` provides the
+default LPV-200 budget. The direct result has `hplM`, `vplM`, `sigmaAccHM`, and
+`sigmaAccVM`, plus the detailed monitor fields.
+
 ## Capabilities
 
 The wasm surface mirrors the full breadth of the engine:
@@ -115,21 +151,32 @@ The wasm surface mirrors the full breadth of the engine:
   batch least-squares orbit fitting against precise ephemerides (including
   terrestrial-frame SP3 through the Earth-orientation chain) with a
   per-satellite residual ledger.
-- **GNSS positioning:** SPP, RTK (float/fixed), PPP (float/fixed), DGNSS,
-  moving-baseline RTK, DOP, velocity, and a robust SPP driver that runs fault
-  detection and exclusion (RAIM/FDE) with iterative reweighting.
-- **Integrity and error bounds:** multi-constellation ARAIM protection levels,
+- **GNSS positioning:** SPP, public `solveStatic` multi-epoch static
+  positioning with covariance, leave-one-out redundancy diagnostics, and
+  robust weighting, RINEX observation to SPP helpers
+  (`sppInputsFromRinexObs` and `solveSppFromRinexObs`), RTK (float/fixed,
+  sequential/static arcs, wide-lane fixed), PPP (float/fixed, including
+  SPP-seeded auto-init), static PPP temporal-correlation covariance with
+  calibrated day-length bounds, optional elevation cutoff, optional
+  tropospheric-gradient estimation, DGNSS, moving-baseline RTK, DOP, velocity,
+  RAIM over existing SPP solutions, broadcast-ephemeris FDE, and a
+  Huber-reweighted SPP driver that runs fault detection and exclusion
+  (RAIM/FDE) with iterative reweighting.
+- **Integrity and error bounds:** direct post-solve RAIM fault detection,
+  multi-constellation ARAIM protection levels,
   SBAS protection levels (DO-229), per-observation reliability (minimal
   detectable bias, internal/external), observability classification of every
   solution (rank, redundancy, conditioning), and covariance-derived error
   metrics (CEP, R95, SEP, error ellipse) that report wide or flagged bounds
   for weak geometry rather than fabricated confidence.
 - **GNSS corrections and biases:** SBAS message decoding with SBAS-corrected
-  solves, RTCM SSR orbit and clock correction streams, Bias-SINEX code and
-  phase biases (DCB/OSB).
+  solves, RTCM SSR orbit and clock correction streams, RTCM 3 broadcast
+  ephemeris decode for GPS (1019), GLONASS (1020), Galileo (1045/1046),
+  BeiDou (1042), and QZSS (1044), each real-data validated, Bias-SINEX code
+  and phase biases (DCB/OSB).
 - **Timing, estimation, and geodesy:** Allan-family clock stability with
   power-law noise identification (IEEE 1139), scalar Kalman and alpha-beta
-  trackers, CFAR detection thresholds, source localization (ToA/TDOA), robust
+  trackers, CFAR detection thresholds, source localization (ToA/TDOA),
   station velocity (MIDAS) with trajectory fitting and step detection,
   repeating-geometry (sidereal) filtering, geodesic direct and inverse
   problems (Karney), an epoch-aware terrestrial frame catalog (ITRF/ETRF
@@ -154,11 +201,12 @@ The wasm surface mirrors the full breadth of the engine:
   conjunctions and oppositions, Sun/Moon/planet meridian transits.
 - **Atmosphere and Earth models:** Klobuchar and NeQuick-G ionosphere, IONEX
   slant delay, troposphere models, geoid undulation (EGM96), solid Earth and
-  pole tides, ocean tide loading, DTED terrain elevation lookup.
+  pole tides, ocean tide loading, DTED terrain elevation lookup with batch
+  queries, and memory-mappable terrain stores.
 - **RF link budget:** free-space path loss, EIRP, C/N0, antenna gain, Doppler
   shift and range rate.
 - **GNSS/INS fusion:** strapdown mechanization with an error-state EKF (UKF
-  option), loose and tight coupling, robust loose updates, an RTS smoother,
+  option), loose and tight coupling, IGG-III loose updates, an RTS smoother,
   a serializable filter state, and field mode (zero-velocity and
   zero-angular-rate updates, non-holonomic constraints, per-fix-status
   weighting, IMU-to-body mounting matrix), all off by default.
@@ -169,7 +217,7 @@ The wasm surface mirrors the full breadth of the engine:
   same scenario and seed.
 - **Signal analysis:** closed-form BPSK/BOC spectra, spectral separation
   coefficients, DLL jitter, and multipath error envelopes against published
-  constants.
+  constants, plus GPS C/A correlation helpers.
 - **Format parsing and serialization:** TLE/OMM, CCSDS (OEM/OPM/CDM/TDM),
   RINEX observation/navigation/clock, CRINEX (Hatanaka), SP3, IONEX, ANTEX,
   Bias-SINEX, RTCM.
@@ -179,9 +227,10 @@ computes. Failures surface as the JS exception you would expect (`Error` for
 engine rejections such as parse failures, non-converging solves, and SGP4 error
 codes; `TypeError` for malformed input; `RangeError` for out-of-domain numbers).
 Full signatures live in the bundled TypeScript declarations (`sidereon.d.ts`),
-with the plain-object request shapes in `@neilberkman/sidereon/types`.
+with the plain-object request types in `@neilberkman/sidereon/types`, including
+typed ARAIM, RTK/PPP, fusion, signal-analysis, and terrain protocols.
 
-A few conventions worth knowing: positions and state arrays cross as
+A few conventions to know: positions and state arrays cross as
 `Float64Array` (multi-epoch arrays are flat row-major, `3 * epochCount`); SGP4
 epoch grids are `BigInt64Array` of unix microseconds; SP3 query epochs are plain
 numbers in seconds since J2000.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@neilberkman/sidereon",
-  "version": "0.23.0",
+  "version": "0.25.0",
   "description": "WebAssembly / JavaScript interface over the sidereon GNSS + astrodynamics engine: SP3 loading and multi-product merge, SPP / RTK (float + fixed) / PPP (float + fixed) positioning, IONEX slant delay, SGP4 and numerical orbit propagation, RINEX observation/navigation/clock parsing, CRINEX (Hatanaka) decoding, observable-domain math (carrier combinations, cycle slips, Hatch smoothing, weighting, velocity, GPS C/A signal), and the astro domain (scale-tagged time, frame transforms, Sun/Moon ephemerides, eclipse / body angles, DOP, passes / visibility, conjunction + collision probability, CDM / OMM / ANTEX parsing, RF link budget)",
   "type": "module",
   "license": "MIT",