test: real-data PVT integration test (Fraunhofer L125, GPS L1 C/A)

[zsoerenm]

Summary

Adds an opt-in real-data integration test that computes a full PVT fix end to end from a real-sky GNSS recording — the real-signal counterpart to the synthetic pvt.jl / pvt_iono_tropo.jl tests. It exercises the whole JuliaGNSS stack the way a receiver does: Acquisition.jl acquires GPS L1 C/A, Tracking.jl tracks, GNSSDecoder.jl decodes the navigation message, and calc_pvt produces a position that is asserted to land at the recording site.

Modelled on the GNSSDecoder.jl / Tracking.jl Flexiband integration tests, but taken all the way to a position fix.

Capture & why GPS L1 C/A only

• Fraunhofer IIS "L125" static roof recording (Nuremberg, 2014-09-23, 210 s, CC BY-NC), L1-band .bin (zero-IF complex baseband, fs = 20 MHz, interleaved int8 I/Q — parameters from Fraunhofer's own GNSS-SDR config).
• calc_pvt needs ≥4 healthy satellites of the same GNSS. GPS L1 C/A is the only constellation here with ≥4 satellites that decode a complete ephemeris. (The III-7a capture that Tracking/GNSSDecoder use is only ~15.7 s — too short to decode a full ephemeris, so it can't yield a fix.)
• The first 100 s is used: a GPS LNAV frame is 30 s and a cold start can need ~45 s to capture a consecutive subframe 1-2-3 set, so 100 s clears the ≥4-satellite bar with margin.

Streaming demux (constant memory)

Mirrors JuliaGNSS/Tracking.jl#158: the full archive is ~3.45 GB, so we range-download only a bounded compressed prefix (~2 GB) and cache that (never the decoded signal). During tracking it raw-inflates the prefix and demuxes one 0.2 s window of int8 I/Q → ComplexF32 at a time, feeding each straight into track!. Peak memory is a single window (~16 MB), not the ~16 GB a whole-span decode would need.

Assertions

Run against the real capture, the pipeline decodes 4 satellites (PRNs 13, 17, 20, 23) and fixes at 49.4873°N, 11.1255°E, 421 m (the Fraunhofer roof). The fix is fully deterministic (bit-identical single- vs multi-threaded), and the satellite set is pinned so the geometry is fixed too, so the test asserts:

• ≥4 satellites used; the 4 known PRNs decode a healthy ephemeris,
• position within 100 m of the decoded reference,
• finite, positive GDOP,
• decoded navigation time is GPS week 1811 (2014-09-23),
• a non-null solution time.

Opt-in & CI

Gated on ENV["PVT_RUN_INTEGRATION_TEST"] — a plain ]test skips it (no download). A dedicated CI job enables it and caches the compressed prefix (actions/cache, fixed key) so repeat runs skip the download.

Verified locally: default suite passes with the integration test skipped; the integration test passes 8/8 with the fix above.

🤖 Generated with Claude Code

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 92.77%. Comparing base (e031f5a) to head (b07b0a1).

Additional details and impacted files

@@           Coverage Diff           @@
##           master      #44   +/-   ##
=======================================
  Coverage   92.77%   92.77%           
=======================================
  Files           7        7           
  Lines         360      360           
=======================================
  Hits          334      334           
  Misses         26       26           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[github-actions]

Benchmark Results (Julia v1)

Time benchmarks

	master	b07b0a1...	master / b07b0a1...
calc_pvt/GPSL1/4sats/cold	0.0325 ± 0.0029 ms	0.0328 ± 0.0027 ms	0.99 ± 0.12
calc_pvt/GPSL1/4sats/warm	28.2 ± 4 μs	28.9 ± 4.2 μs	0.974 ± 0.2
calc_pvt/GPSL1/9sats/cold	0.0415 ± 0.00075 ms	0.0415 ± 0.00076 ms	1 ± 0.026
calc_pvt/GPSL1/9sats/warm	19.1 ± 0.56 μs	19.1 ± 0.48 μs	1 ± 0.039
calc_pvt/GalileoE1B/4sats/cold	30.8 ± 5.3 μs	31.3 ± 5.5 μs	0.987 ± 0.24
calc_pvt/GalileoE1B/4sats/warm	11.9 ± 0.36 μs	12 ± 0.38 μs	0.988 ± 0.043
calc_pvt/GalileoE1B/5sats/cold	0.032 ± 0.0045 ms	0.0323 ± 0.0047 ms	0.992 ± 0.2
calc_pvt/GalileoE1B/5sats/warm	13.3 ± 0.44 μs	13.4 ± 0.38 μs	0.993 ± 0.043
time_to_load	2.27 ± 0.018 s	2.26 ± 0.0023 s	1.01 ± 0.0078

Memory benchmarks

	master	b07b0a1...	master / b07b0a1...
calc_pvt/GPSL1/4sats/cold	0.312 k allocs: 16.9 kB	0.312 k allocs: 16.9 kB	1
calc_pvt/GPSL1/4sats/warm	0.257 k allocs: 14 kB	0.257 k allocs: 14 kB	1
calc_pvt/GPSL1/9sats/cold	0.295 k allocs: 18.3 kB	0.295 k allocs: 18.3 kB	1
calc_pvt/GPSL1/9sats/warm	0.141 k allocs: 9.02 kB	0.141 k allocs: 9.02 kB	1
calc_pvt/GalileoE1B/4sats/cold	0.301 k allocs: 16.3 kB	0.301 k allocs: 16.3 kB	1
calc_pvt/GalileoE1B/4sats/warm	0.136 k allocs: 7.09 kB	0.136 k allocs: 7.09 kB	1
calc_pvt/GalileoE1B/5sats/cold	0.291 k allocs: 16 kB	0.291 k allocs: 16 kB	1
calc_pvt/GalileoE1B/5sats/warm	0.137 k allocs: 7.34 kB	0.137 k allocs: 7.34 kB	1
time_to_load	0.149 k allocs: 11.2 kB	0.149 k allocs: 11.2 kB	1