v1.4.1

1.4.1 — 2026-03-13

• Hardened antenna and receiver compatibility checks: pole-count validation, loka/polarity metadata matching, and keyed frequency gating now live directly in the device pipeline.
• Integrated SigmaGuard into receiver-side purification flows and aligned secure_transmission / property_transfer_chain with the explicit O2 cleanup path.
• Expanded theory-facing coverage for odd-N cascades, translation-gap branches, probability_tensor(), and clean-state witness/noise validation.
• Reworded tetradic/pseudo-quantum helper metadata so complex arithmetic, phase operators, and direct composition are clearly marked as projection/reference vocabulary rather than primary multipolar law.

v1.4.0

Added weighted linear-sum (“linear form”) Σ support via linear_coeffs in sigma.py and pseudomultipolar_timeseries.py.
Added complex sigma_residual() (and extended sigma_norm()) for richer Σ telemetry.
Extended SigmaGuard to store/apply linear_coeffs across N/NX stages.
Added theory-friendly conjugacy aliases on waves: conjugacy_density() / conjugacy_norm().
Updated demos and docs (README/AGENTS) to expose linear-form cleaning and residual traces.

Full Changelog: v1.3.0...v1.4.0

v1.3.0

Added NumPy time-series pseudomultipolar cascade engine (M→N→NX) with Σ diagnostics + demo artifacts (NPZ/JSON).
Added “Witness Pack” reference utilities: CHSH (d=2) / CGLMP (d>2) + Σ-consistent vs generic unitary noise + demo summary JSON.
Added Translation Gap demo (n→2 projection) with visibility/loss metrics + artifacts.
Updated docs and bumped version to 1.3.0.

Full Changelog: v1.2.0...v1.3.0

v1.2.0

v1.2.0 (2026-02-03)
Algebra/theory made consistent for experiments: added table-driven GenericRelationalLoka, a strict create_loka() factory with name normalization, and doc-derived relational lokas (RelHexaSym, RelHeptaTPL3); RelationalLoka.inverse() is now conservative (only returns a true inverse to neutral).
Superposition wiring clarified: SuperpositionalLoka/Harloka enforce shared neutral by default; direct-sum compositions (e.g. compose_two_triads_to_c6) explicitly opt out.
Σ-cascade tools improved: nx_stage() now supports real per-section “tap” strengths (0 < tap ≤ 1) and SigmaGuard docs/behavior updated.
Wave + metadata standardized: Σ is the sum of complex amplitudes; WaveMetadata now includes frequency_hz and a from_amplitudes() helper; metadata is refreshed after transformations.
Pseudo-quantum consistency: measurement/visualization and MultiConjugateFunction normalization now use |ψ|^k with k = n_conjugates.
Device chain improvements: transmitter now modulates an oscillator-formed carrier (message_gain added); antennas refresh metadata after gain/loss; receiver frequency checks use metadata.frequency_hz.
Tests and README examples verified; scenarios made deterministic by default (pseudo_quantum_hadamard_phase uses seed=123).

Full Changelog: v1.1.0...v1.2.0

v1.1.0

Release notes — v1.1.0

Align Σ with theory: WaveMetadata.sigma_residual is now Σ (sum of pole amplitudes) and sigma_norm is |Σ|; frequency_hz is an explicit field (physics/multipolar_wave.py (line 15)).
Align n‑conjugate modulus: MultiConjugateFunction.probability_density() returns Σ|ψ|^k (k=n_conjugates) and normalize() enforces Σ|ψ|^k = 1 (physics/multipolar_wave.py (line 108)).
Pseudo‑quantum measurement now uses |ψ|^k (instead of |ψ|²) consistently in gates + summaries (simulation/multipolar_pseudo_quantum/gates.py (line 1)).
Volumetric chain: MultipolarTransmitter now forms a carrier via MultipolarOscillator.generate_wave() and modulates it per‑pole (parameter message_gain) (devices/communication.py (line 31)).
NX is now meaningful as N1…NX: nx_stage supports per‑section taps (0<tap≤1) for gradual Σ removal (physics/sigma.py (line 86)).
Adds pytest coverage for Σ, k‑norm, measurement, NX taps, and carrier modulation (tests/test_theory_alignment.py (line 1)).
Migration notes

If you depended on the old meaning of sigma_residual/sigma_norm or the old probability_density() scaling, update callers accordingly.

v1.0

update readme