sc_neurocore_engine 3.15.34

High-performance SIMD backend for SC-NeuroCore stochastic neuromorphic computing
Documentation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
// SPDX-License-Identifier: AGPL-3.0-or-later
// Commercial license available
// © Concepts 1996–2026 Miroslav Šotek. All rights reserved.
// © Code 2020–2026 Miroslav Šotek. All rights reserved.
// ORCID: 0009-0009-3560-0851
// Contact: www.anulum.li | protoscience@anulum.li
// SC-NeuroCore — Rust Photonic NoC Acceleration

//! High-performance photonic network-on-chip primitives.
//!
//! Accelerates the hot paths in the Python `photonic_noc` bridge:
//! - **Waveguide routing** — O(N²) Manhattan distance on mesh
//! - **Crosstalk analysis** — O(N²) pairwise channel coupling
//! - **MZI transfer matrix** — 2×2 unitary matrix cascade
//! - **Power budget** — path loss accumulation

use rayon::prelude::*;

// ── Constants ────────────────────────────────────────────────────────

const CROSSING_LOSS_DB: f64 = 0.08;
#[allow(dead_code)] // reserved for future MZI loss accounting
const MZI_INSERTION_LOSS_DB: f64 = 0.5;

// ── Waveguide routing ────────────────────────────────────────────────

/// A routed waveguide segment.
#[derive(Clone, Debug)]
pub struct WaveguideResult {
    pub source: usize,
    pub target: usize,
    pub length_um: f64,
    pub loss_db: f64,
    pub n_crossings: usize,
}

/// Route waveguides on a mesh topology from an adjacency matrix.
///
/// Returns vector of (source, target, length_um, loss_db, n_crossings).
pub fn route_waveguides(
    adjacency: &[f64],
    n: usize,
    pitch_um: f64,
    loss_db_per_cm: f64,
) -> Vec<WaveguideResult> {
    let grid_size = ((n as f64).sqrt().ceil() as usize).max(1);

    let pairs: Vec<(usize, usize)> = (0..n)
        .flat_map(|i| ((i + 1)..n).map(move |j| (i, j)))
        .collect();

    pairs
        .par_iter()
        .filter_map(|&(i, j)| {
            let w = adjacency[i * n + j].abs() + adjacency[j * n + i].abs();
            if w < 1e-12 {
                return None;
            }

            let (ri, ci) = (i / grid_size, i % grid_size);
            let (rj, cj) = (j / grid_size, j % grid_size);
            let manhattan = (ri as isize - rj as isize).unsigned_abs()
                + (ci as isize - cj as isize).unsigned_abs();

            let length_um = manhattan as f64 * pitch_um;
            let mut loss = length_um * 1e-4 * loss_db_per_cm;
            let n_crossings = if manhattan > 0 { manhattan - 1 } else { 0 };
            loss += n_crossings as f64 * CROSSING_LOSS_DB;

            Some(WaveguideResult {
                source: i,
                target: j,
                length_um,
                loss_db: loss,
                n_crossings,
            })
        })
        .collect()
}

// ── MZI transfer matrix ──────────────────────────────────────────────

/// MZI 2×2 transfer matrix elements (complex).
///
/// M = [[cos(φ/2), i·sin(φ/2)], [i·sin(φ/2), cos(φ/2)]]
///
/// Returns (re_00, im_00, re_01, im_01, re_10, im_10, re_11, im_11)
pub fn mzi_transfer_matrix(phase_rad: f64) -> [f64; 8] {
    let half = phase_rad / 2.0;
    let c = half.cos();
    let s = half.sin();
    // [[c, i·s], [i·s, c]]  →  stored as (re, im) pairs
    [c, 0.0, 0.0, s, 0.0, s, c, 0.0]
}

/// Cascade N MZI stages by multiplying transfer matrices.
///
/// Returns the final 2×2 complex matrix as 8 f64s.
pub fn cascade_mzi(phases: &[f64]) -> [f64; 8] {
    let mut result = [1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0]; // identity

    for &phase in phases {
        let m = mzi_transfer_matrix(phase);
        result = complex_mat_mul(&result, &m);
    }

    result
}

/// Multiply two 2×2 complex matrices (each stored as 8 f64s).
fn complex_mat_mul(a: &[f64; 8], b: &[f64; 8]) -> [f64; 8] {
    // a = [[a00, a01], [a10, a11]], b = [[b00, b01], [b10, b11]]
    // Each element is (re, im) at indices (2k, 2k+1)
    let cmul = |ar: f64, ai: f64, br: f64, bi: f64| -> (f64, f64) {
        (ar * br - ai * bi, ar * bi + ai * br)
    };
    let cadd = |a: (f64, f64), b: (f64, f64)| -> (f64, f64) { (a.0 + b.0, a.1 + b.1) };

    // c00 = a00*b00 + a01*b10
    let c00 = cadd(cmul(a[0], a[1], b[0], b[1]), cmul(a[2], a[3], b[4], b[5]));
    // c01 = a00*b01 + a01*b11
    let c01 = cadd(cmul(a[0], a[1], b[2], b[3]), cmul(a[2], a[3], b[6], b[7]));
    // c10 = a10*b00 + a11*b10
    let c10 = cadd(cmul(a[4], a[5], b[0], b[1]), cmul(a[6], a[7], b[4], b[5]));
    // c11 = a10*b01 + a11*b11
    let c11 = cadd(cmul(a[4], a[5], b[2], b[3]), cmul(a[6], a[7], b[6], b[7]));

    [c00.0, c00.1, c01.0, c01.1, c10.0, c10.1, c11.0, c11.1]
}

// ── Crosstalk analysis ───────────────────────────────────────────────

/// Crosstalk result per channel.
#[derive(Clone, Debug)]
pub struct CrosstalkResult {
    pub channel_id: usize,
    pub wavelength_nm: f64,
    pub n_adjacent: usize,
    pub crosstalk_db: f64,
    pub osnr_db: f64,
}

/// Analyze inter-channel crosstalk for WDM channels.
///
/// channels: list of (channel_id, wavelength_nm, bandwidth_nm, power_dbm)
pub fn analyze_crosstalk(
    channels: &[(usize, f64, f64, f64)],
    adjacent_xt_db: f64,
) -> Vec<CrosstalkResult> {
    channels
        .par_iter()
        .map(|&(ch_id, wl, bw, power)| {
            let n_adj = channels
                .iter()
                .filter(|&&(other_id, other_wl, _, _)| {
                    other_id != ch_id && (wl - other_wl).abs() < bw * 3.0
                })
                .count();

            let xt = adjacent_xt_db + 10.0 * (n_adj.max(1) as f64).log10();
            let osnr = power - xt;

            CrosstalkResult {
                channel_id: ch_id,
                wavelength_nm: wl,
                n_adjacent: n_adj,
                crosstalk_db: xt,
                osnr_db: osnr,
            }
        })
        .collect()
}

// ── Power budget ─────────────────────────────────────────────────────

/// Path power budget result.
#[derive(Clone, Debug)]
pub struct PowerBudgetResult {
    pub source: usize,
    pub target: usize,
    pub total_loss_db: f64,
    pub received_power_dbm: f64,
    pub margin_db: f64,
    pub passed: bool,
}

/// Analyze power budget for all waveguide paths.
pub fn analyze_power_budget(
    waveguides: &[(usize, usize, f64)], // (source, target, wg_loss_db)
    mzi_ports: &[(Vec<usize>, usize, f64)], // (input_ports, output_port, insertion_loss)
    laser_power_dbm: f64,
    detector_sensitivity_dbm: f64,
) -> Vec<PowerBudgetResult> {
    waveguides
        .par_iter()
        .map(|&(src, tgt, wg_loss)| {
            let mzi_loss: f64 = mzi_ports
                .iter()
                .filter(|(inputs, output, _)| inputs.contains(&src) || *output == tgt)
                .map(|(_, _, loss)| loss)
                .sum();

            let total_loss = wg_loss + mzi_loss;
            let received = laser_power_dbm - total_loss;
            let margin = received - detector_sensitivity_dbm;

            PowerBudgetResult {
                source: src,
                target: tgt,
                total_loss_db: total_loss,
                received_power_dbm: received,
                margin_db: margin,
                passed: margin >= 0.0,
            }
        })
        .collect()
}

// ── Geometric (evanescent) crosstalk for a waveguide bank ───────────
//
// Physical model: coupled-mode theory with a Marcatili-form transverse
// decay for the coupling coefficient.
//
//   L_decay(λ, n_core, n_clad) = λ / (2π √(n_core² - n_clad²))      [nm]
//   Δn_eff(g)                  = 0.1 · exp(-g / L_decay)             [—]
//   κ(g)                       = π · Δn_eff(g) / (λ [µm])            [µm⁻¹]
//   power coupling ratio       = sin²(κ·L)                           [—]
//   pair isolation             = -10 log₁₀(ratio)                    [dB]
//
// References:
// - Marcatili, Bell Syst. Tech. J. 48(7):2071-2102, 1969
// - Okamoto, *Fundamentals of Optical Waveguides*, 2006, Ch. 4

#[derive(Clone, Debug)]
pub struct CrosstalkPairResult {
    pub index_a: usize,
    pub index_b: usize,
    pub gap_nm: f64,
    pub coupling_length_um: f64,
    pub coupling_coefficient_per_um: f64,
    pub coupling_ratio: f64,
    pub isolation_db: f64,
}

#[derive(Clone, Debug)]
pub struct CrosstalkBankResult {
    pub num_waveguides: usize,
    pub num_near_pairs: usize,
    pub num_far_pairs: usize,
    pub gap_nm: f64,
    pub coupling_length_um: f64,
    pub adjacent_coupling_ratio: f64,
    pub adjacent_isolation_db: f64,
    pub next_nearest_coupling_ratio: f64,
    pub next_nearest_isolation_db: f64,
    pub worst_isolation_db: f64,
    pub mean_coupling_ratio: f64,
    pub max_coupling_ratio: f64,
    pub crosstalk_safe: bool,
}

#[inline]
fn pair_coupling(
    gap_nm: f64,
    coupling_length_um: f64,
    wavelength_nm: f64,
    core_index: f64,
    cladding_index: f64,
) -> (f64, f64, f64) {
    // Transverse evanescent decay length (Marcatili).
    let n2 = (core_index * core_index - cladding_index * cladding_index).max(1e-6);
    let l_decay_nm = wavelength_nm / (2.0 * std::f64::consts::PI * n2.sqrt());
    // Effective-index split at the coupler.
    let dn_eff = 0.1 * (-gap_nm / l_decay_nm).exp();
    // κ in per-µm. λ converted µm.
    let lambda_um = wavelength_nm * 1.0e-3;
    let kappa = std::f64::consts::PI * dn_eff / lambda_um;
    // Power coupling ratio for uniform parallel coupler of length L.
    let kl = kappa * coupling_length_um;
    let ratio = kl.sin().powi(2);
    let iso_db = if ratio > 1.0e-30 {
        -10.0 * ratio.log10()
    } else {
        300.0
    };
    (kappa, ratio, iso_db)
}

/// Analyse crosstalk in a uniform parallel-waveguide bank. Adjacent pairs
/// (gap = g) are the dominant term; next-nearest (gap = 2g) are included
/// as the largest secondary term — Marcatili 1969 predicts that all other
/// pairs are at least `exp(-2·g/L_decay)` smaller still.
pub fn analyze_crosstalk_bank(
    num_waveguides: usize,
    gap_nm: f64,
    coupling_length_um: f64,
    wavelength_nm: f64,
    core_index: f64,
    cladding_index: f64,
) -> CrosstalkBankResult {
    let (_, near_ratio, near_iso) = pair_coupling(
        gap_nm,
        coupling_length_um,
        wavelength_nm,
        core_index,
        cladding_index,
    );
    let (_, far_ratio, far_iso) = pair_coupling(
        2.0 * gap_nm,
        coupling_length_um,
        wavelength_nm,
        core_index,
        cladding_index,
    );

    let num_near = num_waveguides.saturating_sub(1);
    let num_far = num_waveguides.saturating_sub(2);
    let total_pairs = num_near + num_far;
    let (worst_iso, mean_ratio, max_ratio) = if total_pairs == 0 {
        (f64::INFINITY, 0.0, 0.0)
    } else {
        let worst = near_iso.min(far_iso);
        let mean =
            ((num_near as f64) * near_ratio + (num_far as f64) * far_ratio) / (total_pairs as f64);
        let mx = near_ratio.max(far_ratio);
        (worst, mean, mx)
    };

    CrosstalkBankResult {
        num_waveguides,
        num_near_pairs: num_near,
        num_far_pairs: num_far,
        gap_nm,
        coupling_length_um,
        adjacent_coupling_ratio: near_ratio,
        adjacent_isolation_db: near_iso,
        next_nearest_coupling_ratio: far_ratio,
        next_nearest_isolation_db: far_iso,
        worst_isolation_db: worst_iso,
        mean_coupling_ratio: mean_ratio,
        max_coupling_ratio: max_ratio,
        crosstalk_safe: worst_iso > 20.0,
    }
}

/// Per-pair crosstalk for arbitrary waveguide geometry.
/// `pairs` carries `(idx_a, idx_b, gap_nm, coupling_length_um)` per pair.
/// Evaluated in parallel via Rayon — this is the O(N²) path the
/// commercial layout tools call after full pair enumeration.
pub fn analyze_crosstalk_pairs(
    pairs: &[(usize, usize, f64, f64)],
    wavelength_nm: f64,
    core_index: f64,
    cladding_index: f64,
) -> Vec<CrosstalkPairResult> {
    pairs
        .par_iter()
        .map(|&(a, b, gap, len)| {
            let (kappa, ratio, iso) =
                pair_coupling(gap, len, wavelength_nm, core_index, cladding_index);
            CrosstalkPairResult {
                index_a: a,
                index_b: b,
                gap_nm: gap,
                coupling_length_um: len,
                coupling_coefficient_per_um: kappa,
                coupling_ratio: ratio,
                isolation_db: iso,
            }
        })
        .collect()
}

// ── Thermal phase shifter ────────────────────────────────────────────

/// Compute electrical power (mW) needed for a phase shift.
pub fn thermal_power_for_phase(
    phase_rad: f64,
    wavelength_nm: f64,
    heater_length_um: f64,
    dn_dt: f64,
    thermal_resistance_kw: f64,
) -> f64 {
    let wl_m = wavelength_nm * 1e-9;
    let l_m = heater_length_um * 1e-6;
    let delta_t = (phase_rad * wl_m) / (2.0 * std::f64::consts::PI * dn_dt * l_m);
    delta_t.abs() / thermal_resistance_kw
}

// ── Tests ────────────────────────────────────────────────────────────

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn test_route_simple_2x2() {
        // 2-node full adjacency
        let adj = vec![0.0, 1.0, 1.0, 0.0];
        let result = route_waveguides(&adj, 2, 250.0, 2.0);
        assert_eq!(result.len(), 1);
        assert_eq!(result[0].source, 0);
        assert_eq!(result[0].target, 1);
        assert!((result[0].length_um - 250.0).abs() < 1.0);
    }

    #[test]
    fn test_route_sparse() {
        // 3 nodes, only 0↔1 connected
        let adj = vec![0.0, 1.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0];
        let result = route_waveguides(&adj, 3, 250.0, 2.0);
        assert_eq!(result.len(), 1);
    }

    #[test]
    fn test_route_loss_model() {
        let adj = vec![0.0, 1.0, 1.0, 0.0];
        let result = route_waveguides(&adj, 2, 1000.0, 2.0);
        // 1000 µm = 0.1 cm → 0.2 dB propagation loss
        assert!((result[0].loss_db - 0.2).abs() < 0.01);
    }

    #[test]
    fn test_mzi_identity() {
        // phase=0 → identity
        let m = mzi_transfer_matrix(0.0);
        assert!((m[0] - 1.0).abs() < 1e-10); // M00 real
        assert!((m[6] - 1.0).abs() < 1e-10); // M11 real
        assert!(m[3].abs() < 1e-10); // M01 imag = 0
    }

    #[test]
    fn test_mzi_pi_barstate() {
        // phase=π → bar state (swap)
        let m = mzi_transfer_matrix(std::f64::consts::PI);
        // cos(π/2) ≈ 0, sin(π/2) ≈ 1
        assert!(m[0].abs() < 1e-10); // M00 real ≈ 0
        assert!((m[3] - 1.0).abs() < 1e-10); // M01 imag ≈ 1
    }

    #[test]
    fn test_cascade_identity() {
        // Cascade 0 phases → identity
        let result = cascade_mzi(&[]);
        assert!((result[0] - 1.0).abs() < 1e-10);
        assert!((result[6] - 1.0).abs() < 1e-10);
    }

    #[test]
    fn test_cascade_two_stages() {
        let phases = vec![std::f64::consts::PI / 4.0, std::f64::consts::PI / 4.0];
        let result = cascade_mzi(&phases);
        // Should be equivalent to single π/2 cascade (non-trivial)
        let mag = (result[0] * result[0] + result[1] * result[1]).sqrt();
        assert!(mag <= 1.0 + 1e-10);
    }

    #[test]
    fn test_crosstalk_single() {
        let channels = vec![(0, 1550.0, 0.8, 0.0)];
        let result = analyze_crosstalk(&channels, -25.0);
        assert_eq!(result.len(), 1);
    }

    #[test]
    fn test_crosstalk_adjacent() {
        let channels = vec![
            (0, 1550.0, 0.8, 0.0),
            (1, 1550.8, 0.8, 0.0),
            (2, 1551.6, 0.8, 0.0),
        ];
        let result = analyze_crosstalk(&channels, -25.0);
        assert_eq!(result.len(), 3);
        // Middle channel has 2 adjacent
        let mid = &result[1];
        assert_eq!(mid.n_adjacent, 2);
    }

    #[test]
    fn test_power_budget_pass() {
        let wgs = vec![(0, 1, 1.0)]; // 1 dB loss
        let mzis: Vec<(Vec<usize>, usize, f64)> = vec![];
        let result = analyze_power_budget(&wgs, &mzis, 0.0, -20.0);
        assert_eq!(result.len(), 1);
        assert!(result[0].passed); // 0 - 1 = -1 dBm > -20 dBm
        assert!((result[0].margin_db - 19.0).abs() < 0.01);
    }

    #[test]
    fn test_power_budget_fail() {
        let wgs = vec![(0, 1, 25.0)]; // 25 dB loss
        let result = analyze_power_budget(&wgs, &[], 0.0, -20.0);
        assert!(!result[0].passed); // 0 - 25 = -25 dBm < -20 dBm
    }

    #[test]
    fn test_thermal_power() {
        let p = thermal_power_for_phase(
            std::f64::consts::PI, // π phase shift
            1550.0,               // wavelength
            100.0,                // heater length
            1.86e-4,              // dn/dT
            10.0,                 // thermal R
        );
        assert!(p > 0.0);
        assert!(p < 100.0); // reasonable range
    }

    #[test]
    fn crosstalk_bank_isolation_grows_with_gap() {
        let narrow = analyze_crosstalk_bank(4, 100.0, 10.0, 1550.0, 3.48, 1.45);
        let wide = analyze_crosstalk_bank(4, 400.0, 10.0, 1550.0, 3.48, 1.45);
        // Wider gap ⇒ less coupling ⇒ higher isolation (dB).
        assert!(wide.worst_isolation_db > narrow.worst_isolation_db);
        // Nearest-neighbour dominates over next-nearest (smaller gap couples more).
        assert!(narrow.adjacent_coupling_ratio >= narrow.next_nearest_coupling_ratio);
    }

    #[test]
    fn crosstalk_bank_counts_match_bank_size() {
        let r = analyze_crosstalk_bank(5, 200.0, 10.0, 1550.0, 3.48, 1.45);
        assert_eq!(r.num_near_pairs, 4); // N-1 adjacent
        assert_eq!(r.num_far_pairs, 3); // N-2 next-nearest
        assert!(r.crosstalk_safe || r.worst_isolation_db <= 20.0);
    }

    #[test]
    fn crosstalk_bank_single_waveguide_has_no_pairs() {
        let r = analyze_crosstalk_bank(1, 200.0, 10.0, 1550.0, 3.48, 1.45);
        assert_eq!(r.num_near_pairs, 0);
        assert_eq!(r.num_far_pairs, 0);
        assert!(r.worst_isolation_db.is_infinite());
    }

    #[test]
    fn crosstalk_pairs_parallelism_matches_serial_math() {
        let pairs = vec![
            (0, 1, 200.0, 10.0),
            (1, 2, 400.0, 10.0),
            (0, 2, 800.0, 10.0),
        ];
        let out = analyze_crosstalk_pairs(&pairs, 1550.0, 3.48, 1.45);
        assert_eq!(out.len(), 3);
        // Sanity: larger gap ⇒ smaller coupling ratio.
        assert!(out[0].coupling_ratio >= out[1].coupling_ratio);
        assert!(out[1].coupling_ratio >= out[2].coupling_ratio);
        // Isolation monotonically increases (inverse of ratio).
        assert!(out[0].isolation_db <= out[1].isolation_db);
        assert!(out[1].isolation_db <= out[2].isolation_db);
    }
}