38-EFT.WP.EDX.EMI v1.0 | Chapter 10 — Experimental Design & Falsification (M20-EMI)

Home ／ Docs-Technical WhitePaper ／ 38-EFT.WP.EDX.EMI v1.0

Chapter 10 — Experimental Design & Falsification (M20-EMI)

I. Chapter Objectives & Structure

1. Objective: Build EDX.EMI positive/negative criteria and power analysis, provide reproducible protocols, controls, and release gates so that Z_eft(omega), ΔZ_rad(omega), I_CM(omega), and field envelopes E_rad/H_rad close under a unified metrology/recording dialect and are falsifiable.
2. Structure: Test matrix & controls → Criteria & sample size → Observation geometry & path/mode design → Data cleaning & blinding → Reference experiments & workflow → Failure modes & troubleshooting → Compliance templates → Cross-chapter closure.
3. Shared time-of-arrival dialect (equivalent; explicit gamma(ell) & d ell; record delta_form):
   • Constant-factored: T_arr = ( 1 / c_ref ) * ( ∫ n_eff d ell )
   • General: T_arr = ( ∫ ( n_eff / c_ref ) d ell )

II. Test Matrix & Controls (Overview)

1. Primary endpoints: Re{ΔZ_rad(ω)} (positive-real / decreases after sealing), I_CM(ω), E_rad/H_rad envelopes, ΔW, slope of arg Z_eft vs ΔT_arr, and S↔Z/mixed-mode normalization consistency.
2. Experiment types:
   • Seam/aperture sealing (verify Re{ΔZ_rad} and fields drop together).
   • Return bridging / stub back-drill (verify w_side and Re{ΔZ_rad} co-vary and drop after mitigation).
   • CDN/BCI/external-field injections (increasing injection yields monotonic rises in I_CM, Re{ΔZ_rad}, |E|/|H|).
   • Mixed-mode consistency (S_se→S_mm→Z_mm→Z_se preserves power/reciprocity).
3. Controls: anti-radiation baseline Z_base = Z_ref + ΔZ_T; zero-level injection / post-sealing as negative controls.

III. M20-EMI-1 Criteria & Sample Size (Criteria & Power)

1. P1 Positive-real & sealing criterion
   Over the band, Re{ΔZ_rad(ω)} ≥ 0; after sealing/meshing/bridging, mean(Re{ΔZ_rad}) and mean(|E_rad|), mean(|I_CM|) decrease together (95% confidence).
2. P2 Power–field consistency
   Port-estimated P_rad(ω) = ½·Re{ΔZ_rad}|I_port|^2 vs |E_rad|/|H_rad| correlation ρ ≥ ρ_gate (suggest ρ_gate ≥ 0.8).
3. P3 Mixed-mode consistency
   After S_se→S_mm→Z_mm→Z_se, power and reciprocity preserved; mapped data pass passivity/KK; KPIs not degraded.
4. P4 Path–weight coupling
   Pre/post mitigation (sealing/back-drill/bridging), changes in ΔW and Re{ΔZ_rad} are same-direction, deviation ≤ 3·u(ΔW).
5. P5 Two-dialect agreement
   |T_arr^{(n_over_c)} − T_arr^{(n_over_c_times_n)}| ≤ u(T_arr).
6. Power analysis (indicative)
   • For mean-difference δ = mean_before − mean_after, variance σ^2, significance α, power 1−β:
     N ≥ 2 · ( ( z_{1−α/2} + z_{1−β} ) · σ / |δ| )^2 .
   • Suggested: engineering 1−β ≥ 0.8, α ≤ 0.05; release 1−β ≥ 0.9, α ≤ 0.01.

IV. Observation Geometry & Path/Mode Design (Execution Dialect)

• Layout & paths: via binding_ref, annotate {γ_main, γ_side} and per-segment {Δell_i, n_eff(i)}; record geometry: sigma_seam/L_seam/pitch_seam/mesh_ratio, stub_len/backdrill, bridge locations.
• Port normalization & mixed-mode: unify Znorm(ω) and T_mm/Z0_mm (if used); normalize in mixed-mode before back-projecting to SE.
• Window & frequency grid: set in-band window Ω for phase/arrival fits; align field-scan and port frequency grids.
• Records: arrival{form,gamma,measure,c_ref,Tarr,u_Tarr,delta_form}, qa_gates, probe/antenna factors and distance R.

V. Data Cleaning & Blinding

• Pre-registration: freeze frequency grid, IF bandwidth/time constants, fitting methods, criterion gates, and baseline_id/Znorm/T_mm/Z0_mm/binding_ref/delta_form.
• Blinding: scramble geometry/injection configurations; conceal labels until analysis.
• Cleaning flow: de-embed → renorm → S↔Z → sync correction → path correction → radiation equivalence → field conversion → KPI/correlation.
• Hard QA: Re{Z_eft} ≥ 0, KK = pass, two-dialect T_arr agreement, and (if enabled) Re{ΔZ_rad} ≥ 0.

VI. Reference Experiments & Workflow (Reproducible)
A. Seam sealing (validates P1/P2)

• Fabricate a sealable seam fixture;
• Measure S(ω), capture I_port/I_CM/E/H;
• Compute Z_eft/ΔZ_rad/P_rad and E/H; compare means pre/post sealing and correlation ρ;
• Decision: positive if positive-real and joint decreases pass; otherwise negative.

B. Return bridging & stub back-drill (validates P1/P4)

• Add via bridges/back-drill stubs;
• Compare ΔW vs Re{ΔZ_rad} trends;
• Decision: pass if same-direction decreases within 3·u.

C. CDN/BCI/external-field injection (validates P2/P3)

• Increase injection level;
• Verify monotonic rises in I_CM, Re{ΔZ_rad}, |E|/|H| and S↔Z loop consistency;
• Decision: ρ ≥ ρ_gate and passivity/KK pass.

D. Mixed-mode consistency (validates P3/P5)

• Run S_se→S_mm→Z_mm→Z_se loop;
• Check power/reciprocity and two-dialect T_arr;
• Decision: any failure is negative.

VII. Failure Modes & Troubleshooting

• Normalization/mapping mismatch: inconsistent Znorm/T_mm/Z0_mm → unify and recompute.
• Out-of-window coherence: linear-phase fit used outside Ω → confine to Ω or use energy composition.
• Sync mismatch: missing Δt_sync correction → correct timebase before fitting.
• Missing paths: unlabelled γ_side → update binding_ref and geometry records.
• Probe/antenna/site errors: inconsistent factors/distance/coordinates → recalibrate and recheck site.
• Missing injection records: incomplete type/level/detector/IF_BW/time_const → fill before judgment.

VIII. Compliance Templates (copy-ready)

• Experiment protocol card (YAML)
• protocol_emi:
• prereg:
• baseline_id: "BLSN-EMI-001"
• Znorm_ohm: [50.0, 50.0]
• mixed_mode: {enabled:true, T_mm:"/cfg/T_mm.yaml", Z0_mm_ohm:[100,25]}
• binding_ref: "LAY2PATH-EMI-0001"
• delta_form: "n_over_c"
• criteria: ["P1","P2","P3","P4","P5"]
• alpha: 0.05
• power: 0.8
• configs:
• - {id:"before_seal", seam:"open"}
• - {id:"after_seal", seam:"sealed"}
• acquisition:
• band_GHz: [f_min, f_max]
• N_ω: 1024
• repeats: 5
• detector: "qpk"
• if_bw_kHz: 120
• time_const_ms: 1.0
• site: {distance_m: 3.0, coords: "..."}
• alignment:
• deemb: {method:"TRL", version:"1.2", artifact:"/artifacts/deemb.json"}
• sync: {dt_sync_s: 2.0e-12}
• arrival:
• form: "n_over_c"
• gamma: "explicit"
• measure: "d_ell"
• c_ref: 299792458.0
• qa_gates: ["check_dim","passivity(Re{Z}≥0)","KK_consistency"]
• Analysis skeleton (pseudocode)
• # De-embed/renorm → S↔Z
• Z = map_S_to_Z(renorm(deembed(S_raw, fixture), Znorm(ω)), Znorm(ω), T_mm=Tmm, Z0_mm=Z0mm)
• assert min(Re(Z)) >= 0.0 and KK_consistency(Z)
• # Phase correction & arrival
• phi = unwrap(arg(Z) - ω*Δt_sync)
• Tarr = fit_linear_phase(phi, ω).slope
• assert abs(Tarr_n_over_c - Tarr_one_over_c_times_n) <= u_Tarr
• # Radiation equivalence & fields
• ΔZ_rad = Z - Z_base
• P_rad = 0.5 * Re(ΔZ_rad) * abs(I_port)**2
• ρ_E = corr(P_rad, E_rad_env)
• ρ_H = corr(P_rad, H_rad_env)
• assert ρ_E >= ρ_gate and ρ_H >= ρ_gate
• # Injection consistency (if present)
• assert corr(I_CM, Re(ΔZ_rad)) >= ρ_gate
• # P1: pre/post sealing
• assert mean(Re(ΔZ_rad_after)) < mean(Re(ΔZ_rad_before)) and \
• mean(E_after) < mean(E_before)
• # Emit dataset card & QA results

IX. Cross-Chapter Links & Closure

• Dependencies: Chapter 2 (Terms & Symbols), Chapter 4 (Minimal Equations & Equivalents), Chapter 6 (Cables/Connectors & Mixed-Mode), Chapter 7 (Immunity & Injection), Chapter 8 (S↔Z & Field Mapping), Chapter 9 (Metrology Chain).
• Downstream: Chapter 11 (Engineering Rules: Shielding/Ground/Returns — acceptance via these criteria), Chapter 14 (SimStack & Benchmark Cases — regression), Chapter 15 (Data & Repro — equivalence & audit).