Grid: mobility × channel¶

Varies: mobility model (rwp, ssrwp, gaussmarkov) × channel model (tworay, nakagami) — six cells at the paper base scenario.

← Benchmark index · Metrics · Methodology

Provenance — measured at a1daa7a, ReconvHoldCap = 200 ms. This is the v1.5.0 phase-1 re-baseline (campaign plan, #371): the #411 merge flipped the shipped ReconvHoldCap default 1 s → 200 ms — a protocol-behaviour change that superseded the v1.4.0 corpus under the provenance rule — and all six cells were re-measured on main at that merge commit. The baselines are byte-identical to the v1.4.0 grid (0/18 rows moved — see the attribution control below), so every AntHocNet delta on this page is attributable to the flip alone. The previous version of this page (measured at 4cdfb96, 1 s) remains in git history — git show v1.4.0:docs/benchmarks/grid.md — and its numbers stay valid as historical evidence of the 1 s operating point. Reproduce this page at commit a1daa7a.

The fifth arm — the oracle — was measured separately, at 40b434d. Campaign phase 3 (#415, #296) re-ran the same six cells with --protocols=…,oracle on main @ 40b434d, the commit that added the arm. Those cells are a different dispatch at a different commit, so they would normally not be quotable in the same table — except that their 480 baseline ##RUN## rows are byte-identical to the phase-1 corpus (6 cells × 4 protocols × 20 seeds, AntHocNet included). That control is what makes the composition legitimate; it is stated in full in Why the oracle columns compose below, together with the caveat that travels with every oracle number on this page.

What it varies¶

The v1.4.0 exit criteria ask for a headline grid under ≥2 mobility models × ≥2 channel models with a ranking-stability statement. This is that grid, re-established at the v1.5.0 default.

The axes are deliberately built as controlled contrasts rather than as a collection of unrelated models:

Mobility (#61) — rwp is the original evaluation's Random Waypoint; ssrwp is the same model started from its stationary distribution, so rwp vs ssrwp isolates the speed-decay transient alone; gaussmarkov is the qualitatively different model (smooth correlated tracks). pause is inert under Gauss-Markov and is set to 0 there — preflight FAILs any other value.
Channel (#60) — nakagami stacks Nakagami-m fading on the same two-ray path loss used by the tworay arm, so the pair isolates fading alone. The range disk model is deliberately not one of the two: it is a propagation abstraction with no fading and no distance-dependent loss curve, so counting it would satisfy the criterion's letter while leaving its motivation untouched.

How it is produced¶

Six manual paper-benchmark.yml dispatches, one per cell — not the per-merge benchmarks workflow and not scenario-matrix.yml. Each cell runs the paper base scenario (50 nodes, 1500 × 300 m, 900 s, 20 CBR flows) at 20 seeds, with all four protocols on identical realisations inside that cell. Phase 3 repeated the same six dispatches with the oracle control added as a fifth arm — same scenario, same 20 seeds, same #352-pinned RNG streams (the oracle evaluates no propagation model, so it draws from none of them).

Because the cells come from independent dispatches rather than one classified sweep, this page has no generated block — the tables below are hand-written, as on the satellite suite page. The per-run data behind them lives in the six Actions run logs (IDs in the provenance table below; each cell's ##PROV## line pins commit=a1daa7a and its ##CONFIG## pins ReconvHoldCap=+2e+08ns, so every block is self-describing). The committed ../benchmarks/campaign/pooled-grid-mobility-channel-20260808.csv and its per-run sibling remain the 1 s corpus: their aodv/olsr/dsdv rows match this page to the last digit (deterministic baselines, identical #352-pinned seeds), and their anthocnet rows are the superseded 1 s measurement.

Results¶

Delivery — PDR %, mean ± 95 % CI¶

mobility	channel	anthocnet	aodv	olsr	dsdv
rwp	tworay	92.58 ± 0.63	85.92 ± 0.55	90.59 ± 0.46	84.99 ± 0.63
ssrwp	tworay	92.95 ± 0.63	86.56 ± 0.63	91.25 ± 0.62	85.74 ± 0.72
gaussmarkov	tworay	90.08 ± 0.63	83.90 ± 0.81	85.92 ± 0.90	78.70 ± 1.02
rwp	nakagami	89.78 ± 0.86	73.49 ± 1.08	87.78 ± 0.38	71.86 ± 1.03
ssrwp	nakagami	89.54 ± 0.88	73.02 ± 0.71	87.64 ± 0.47	72.37 ± 1.14
gaussmarkov	nakagami	85.87 ± 1.11	67.23 ± 1.26	83.46 ± 0.78	63.99 ± 1.17

Overhead — NRL, mean ± 95 % CI¶

mobility	channel	anthocnet	aodv	olsr	dsdv
rwp	tworay	34.92 ± 0.63	64.64 ± 1.41	4.02 ± 0.04	23.01 ± 0.21
ssrwp	tworay	34.37 ± 0.70	64.37 ± 2.09	3.96 ± 0.04	22.74 ± 0.28
gaussmarkov	tworay	36.65 ± 1.05	63.73 ± 1.81	4.34 ± 0.07	24.90 ± 0.37
rwp	nakagami	46.89 ± 1.09	77.98 ± 1.79	4.35 ± 0.04	27.83 ± 0.41
ssrwp	nakagami	46.88 ± 1.44	76.80 ± 1.53	4.31 ± 0.05	27.41 ± 0.54
gaussmarkov	nakagami	52.44 ± 1.69	85.05 ± 2.40	4.73 ± 0.06	31.43 ± 0.55

Tail — `delay99` ms, mean with bootstrap 95 % interval¶

mobility	channel	anthocnet	aodv	olsr	dsdv
rwp	tworay	420.8 [385.4, 454.6]	584.6 [560.2, 609.2]	23.2 [22.3, 24.2]	419.8 [374.1, 485.7]
ssrwp	tworay	373.3 [331.7, 415.9]	583.5 [558.2, 611.1]	22.2 [21.6, 22.9]	499.1 [411.1, 600.2]
gaussmarkov	tworay	512.8 [463.9, 563.9]	575.5 [551.5, 600.8]	68.7 [24.4, 144.7]	545.9 [421.7, 677.5]
rwp	nakagami	920.0 [868.4, 964.2]	839.1 [807.3, 871.4]	2714.1 [2530.0, 2880.5]	2036.0 [2027.9, 2044.7]
ssrwp	nakagami	890.6 [824.1, 952.6]	817.6 [780.5, 857.5]	2685.2 [2516.3, 2846.7]	2034.0 [2026.5, 2041.8]
gaussmarkov	nakagami	1022.9 [987.5, 1055.6]	1027.5 [958.0, 1103.3]	2961.0 [2867.6, 3009.2]	2271.4 [2137.7, 2428.9]

Bold marks the nominal best in each row, and three of the six rows are now ties rather than wins: rwp-tworay's anthocnet vs dsdv intervals overlap almost completely, gaussmarkov-nakagami's anthocnet-vs-aodv paired difference is −4.65 ms [−83.20, +69.55] (p = 0.7 — a clean statistical tie with anthocnet nominally first), and in the rwp/ssrwp fading cells aodv's paired edge (+80.95 / +72.95 ms, p = 0.012 / 0.033) is marginal against the multiple-comparison expectation below. A t-interval on a per-run p99 is not defensible, so the tail uses a percentile bootstrap (policy).

AntHocNet vs AODV — paired, per seed¶

Both protocols run on identical realisations inside a cell, so this is a paired difference (t-CI for PDR/NRL, bootstrap for delay99, two-sided Wilcoxon). This is the test of record; the per-arm intervals above are for orientation.

mobility	channel	ΔPDR (pp)	ΔNRL	Δ`delay99` (ms)
rwp	tworay	+6.66 [+5.94, +7.39]	−29.73 [−31.36, −28.09]	−163.85 [−208.10, −119.90]
ssrwp	tworay	+6.40 [+5.53, +7.26]	−30.00 [−32.27, −27.74]	−210.20 [−264.25, −157.50]
gaussmarkov	tworay	+6.17 [+5.29, +7.06]	−27.08 [−28.74, −25.42]	−62.65 [−118.50, −8.90]
rwp	nakagami	+16.28 [+15.00, +17.57]	−31.08 [−32.75, −29.42]	+80.95 [+23.50, +134.60]
ssrwp	nakagami	+16.53 [+15.70, +17.36]	−29.93 [−31.56, −28.30]	+72.95 [+10.70, +134.80]
gaussmarkov	nakagami	+18.63 [+17.63, +19.64]	−32.61 [−34.54, −30.68]	−4.65 [−83.20, +69.55]

Every PDR and NRL comparison has p ≤ 9.6 × 10⁻⁵; the smallest of those effects is 7 interval half-widths from zero, so none is marginal. The delay99 column is different in kind from the 1 s corpus, where AntHocNet paid +137…+335 ms against AODV in every cell: at 200 ms the sign flips negative in four of six cells, and the remaining tail comparisons are the marginal ones (p = 0.012–0.7; eighteen comparisons at α = 0.05 expect ~0.9 false positives, so treat the rwp/ssrwp-nakagami values as an aodv-leaning-or-tie, not a settled ordering).

Old → new: what the 200 ms flip cost and bought, per cell¶

The #411 flip's grid-wide price and benefit, AntHocNet only (the baselines did not move — next section). Old = 4cdfb96 at 1 s, new = a1daa7a at 200 ms:

mobility	channel	PDR (Δpp)	`delay99` (Δ%)	NRL Δ
rwp	tworay	97.43 → 92.58 (−4.85)	787.5 → 420.8 (−46.6 %)	−0.22
ssrwp	tworay	97.55 → 92.95 (−4.60)	720.8 → 373.3 (−48.2 %)	+0.10
gaussmarkov	tworay	96.97 → 90.08 (−6.89)	832.3 → 512.8 (−38.4 %)	+0.48
rwp	nakagami	92.03 → 89.78 (−2.25)	1155.2 → 920.0 (−20.4 %)	−2.89
ssrwp	nakagami	92.08 → 89.54 (−2.54)	1152.4 → 890.6 (−22.7 %)	−3.02
gaussmarkov	nakagami	89.33 → 85.87 (−3.46)	1343.2 → 1022.9 (−23.8 %)	−4.66

The trade is clean and grid-wide: tail −20 % to −48 %, delivery −2.3 to −6.9 pp, overhead flat-to-down (the fading cells shed 2.9–4.7 NRL). The #411 pre-merge A/B's −4.38 pp at paper-base/disk sits inside this envelope; the two-ray cells pay more delivery than the fading cells, with the maximum at gaussmarkov-tworay (−6.89 pp). This is the #308 ablation's mechanism at grid scale: the reconvergence hold converts would-be drops into late deliveries, and the cap trades those deliveries back for the tail.

The attribution control — 0/18 baseline rows moved¶

Every aodv/olsr/dsdv pdr/delay99/nrl value matches the v1.4.0 grid (4cdfb96 corpus) to the last printed digit — deterministic baselines on identical #352-pinned seeds, no harness drift between the corpora. Every AntHocNet delta above is therefore attributable to the ReconvHoldCap flip alone. This is the same control the #308 ablation ran (byte-identical AODV blocks across cap arms), now confirmed across a commit gap and all six cells.

The oracle control — how much of the shortfall is routing?¶

This is the question the four-arm grid above cannot answer and the reason phase 3 exists. Every table so far compares protocols to each other; none of them says how much of the distance to perfect is protocol overhead and how much is the channel. The oracle — global-knowledge Dijkstra over the ground-truth topology, replayed as an Ipv4RoutingProtocol, emitting no control traffic whatsoever (#415; framing in methodology.md) — is the arm that makes the split measurable.

Why the oracle columns compose with the tables above¶

The oracle cells are a different dispatch at a different commit (40b434d, which adds contrib/oracle) from the four-arm tables (a1daa7a). Quoting a column measured at one commit inside a table measured at another is exactly the provenance-rule violation this repo re-baselines corpora to avoid — so the composition needs a control, and it has one:

480 ##RUN## rows byte-identical. Every per-seed row of all four original arms — 6 cells × 4 protocols × 20 seeds — is byte-for-byte identical between the phase-1 and phase-3 blocks, as are the ##BENCH##, # stddev, # paths, # drops and # energy lines. AntHocNet included: the subject under test did not move by a single printed digit when the fifth arm was added. Adding the oracle therefore perturbed nothing measurable — same seeds, same realisations, same scheduler order for the arms that were already there — and the oracle column is a measurement of the same six cells, not of a neighbouring configuration. This is the same class of control as the 0/18 attribution control above, applied to an added arm rather than to a changed default, and it is the reason the rest of this section is legitimate rather than merely convenient.

Two structural facts back it up: the oracle module is off unless --protocols names it, and it evaluates no propagation model at all, so it takes no draw from the channel's #352-pinned RNG stream and every other arm sees the identical fading realisation it saw in phase 1.

The oracle arm, per cell¶

Same layout as the tables above — PDR and mean delay with t 95 % CI half-widths, delay99 with a percentile-bootstrap interval, NRL, and the # paths mean hop count. n = 20 seeds.

mobility	channel	PDR %	delay ms	`delay99` ms	hopsMean
rwp	tworay	100.00 ± 0.00	5.16 ± 0.51	23.4 [22.9, 23.7]	2.12
ssrwp	tworay	100.00 ± 0.00	4.88 ± 0.67	23.1 [22.7, 23.6]	2.07
gaussmarkov	tworay	100.00 ± 0.00	6.53 ± 0.63	26.1 [25.8, 26.6]	2.43
rwp	nakagami	99.54 ± 0.13	81.06 ± 3.27	2010.5 [2009.2, 2011.8]	2.04
ssrwp	nakagami	99.55 ± 0.17	80.36 ± 2.69	2010.8 [2009.8, 2011.8]	2.06
gaussmarkov	nakagami	99.41 ± 0.14	110.52 ± 2.59	2019.5 [2018.0, 2021.0]	2.41

PDR ≥ every arm in all six cells, by margins of +7.05 to +35.42 pp (smallest: anthocnet at ssrwp-tworay; largest: dsdv at gaussmarkov-nakagami). NRL is 0.00 in all six — and not merely as a rounded mean: all 120 per-seed oracle rows have nrl min = max = 0.00 and nrl_bytes max = 0.0000. That zero is an asserted invariant (NS_ABORT in-harness plus a scenario_check rule), not a measurement that happened to come out at zero.

These columns are not all readable the same way. PDR is a bound in all six cells; delay/delay99 are a bound only on the three two-ray cells; NRL is an assertion rather than a measurement; and hopsMean is not a bound anywhere on this page. The quoting rule below is where that is settled, with its evidence. Note already that hopsMean is higher than every real arm's in every cell, and that this is not the defect it was once read as. hopsMean averages over each arm's own delivered set, and the oracle delivers 6–33 pp more packets than the arms it bounds — the extra ones being precisely the long-path, hard-to-route packets the others drop. A shortest-path control that delivers the hard packets should read longer here. The survivorship-free instrument is the identity-matched ##COMMON## set below, and on it the ordering reverses: the oracle is below every arm on two-ray and below the reactive arms under fading. An earlier version of this page read the hopsMean ordering as evidence that the adjacency graph was wrong; on the matched set that reading does not hold, and the real approximation error is measured in the caveat section.

Gap decomposition — the headline¶

With an arm whose routing overhead is exactly zero, the shortfall to 100 % splits without modelling assumptions:

channel = 100 − oracle_pdr        (what no routing protocol could have delivered)
routing = oracle_pdr − arm_pdr    (what this protocol lost that perfect routing did not)
total   = 100 − arm_pdr           = channel + routing

mobility	channel	arm	total gap (pp)	channel (pp)	routing (pp)	routing share
rwp	tworay	anthocnet	7.42	0.00	7.42	100.0 %
rwp	tworay	aodv	14.08	0.00	14.08	100.0 %
ssrwp	tworay	anthocnet	7.05	0.00	7.05	100.0 %
ssrwp	tworay	aodv	13.44	0.00	13.44	100.0 %
gaussmarkov	tworay	anthocnet	9.92	0.00	9.92	100.0 %
gaussmarkov	tworay	aodv	16.10	0.00	16.10	100.0 %
rwp	nakagami	anthocnet	10.22	0.46	9.76	95.5 %
rwp	nakagami	aodv	26.51	0.46	26.05	98.3 %
ssrwp	nakagami	anthocnet	10.46	0.45	10.01	95.7 %
ssrwp	nakagami	aodv	26.98	0.45	26.54	98.3 %
gaussmarkov	nakagami	anthocnet	14.13	0.59	13.54	95.8 %
gaussmarkov	nakagami	aodv	32.77	0.59	32.18	98.2 %

Each term is rounded independently from the full-precision per-seed means, so one row's printed components differ from its printed total by 0.01 pp (ssrwp-nakagami/aodv: 0.45 + 26.54 vs 26.98). The values are quoted as the analysis emitted them rather than re-derived to make the row add up.

On two-ray the channel costs nothing at all. The oracle delivers 100.00 % exactly, so every point of AntHocNet's 7.0–9.9 pp shortfall — and of AODV's 13.4–16.1 pp — is protocol overhead: discovery floods, reconvergence holds, stale next hops, packets dropped waiting for a route. Under Nakagami the channel finally costs something, and it costs half a point: 0.45–0.59 pp, leaving 95.5–95.8 % of AntHocNet's gap and 98.2–98.3 % of AODV's on the routing side.

This decomposition is pending a re-derivation, and the direction of the change is known. The table above uses the v1.5.0 oracle, whose adjacency was the scenario's --range — a 300 m geometric disk. #457 replaced that with a radius derived from the installed PHY (423.3 m on two-ray, 373.4 m on Nakagami), and a wider graph routes over marginal links, so the oracle's own delivery falls. Re-measured at 20 seeds on the fixed harness, oracle PDR is 96.00–98.20 % rather than 99.41–100.00 %, which moves the channel term to 1.80–4.00 pp and the routing share to 67.0–88.4 % (AntHocNet 67.0–76.2 %, AODV 79.1–88.4 %). The qualitative reading is unchanged — routing dominates the gap in every cell — but "the channel costs nothing at all on two-ray" and "under 0.6 pp" are specific to the 300 m disk and must not be quoted against the current oracle.

The table is not silently restated here because which oracle belongs in this decomposition is a real analytical choice, not a transcription: the 300 m disk is the more conservative delivery reference (its links are solidly in range, so 100 − oracle_pdr is a tighter floor on what no router could deliver), while the derived radius is the better adjacency model and is what makes the hop bound hold. Picking one for this table is tracked on #460 with the measured numbers for both.

The reading that matters is the one this grid could not previously support: the headroom above AntHocNet is almost entirely addressable in the protocol. A fading channel that intuition blames for a 10–14 pp delivery gap turns out to account for under 0.6 pp of it. What the decomposition does not do is split the routing term further — into discovery cost, suboptimal path choice and reconvergence loss — so it bounds the addressable headroom rather than itemising it.

Note the shape of the two columns: the channel term is a property of the cell, identical for every arm in it, while the routing term is what distinguishes the arms. AntHocNet's routing loss is 2.4× to 2.7× smaller than AODV's on the fading cells (9.76 vs 26.05; 10.01 vs 26.54; 13.54 vs 32.18), which is the same ranking the paired ΔPDR table reports, now expressed against an absolute reference instead of against AODV.

The caveat, stated with the numbers rather than under them¶

All six cells are approximate. Every oracle row on this page is flagged approx=1. A two-ray or Nakagami channel has no crisp adjacency — link viability is a continuous function of distance, and under fading a random one — so the control cannot derive the true graph. Since #457 it is held to a radius derived from the installed PHY rather than to the scenario's --range: mode=decode-approx at 423.3 m on two-ray (where the two-ray power crosses the decode floor) and mode=p50-approx at 373.4 m on Nakagami (the closed-form Gamma law's P = ½ crossing at that same floor). scenario_check.py says so once per seed: "a fading or two-ray channel has no crisp adjacency, so this arm is a reference point, not a proven upper bound." The exact (approx=0) rule exists only where the graph is the wiring — see the satellite ISL suite, which publishes the mode=wired approx=0 cell.

These numbers supersede the phase-3 matched-hop and matched-latency tables, which were measured on a broken instrument. Phase 3 reported that the oracle used more hops than every real arm in all six cells (e.g. rwp-tworay oracle 1.90 against anthocnet 1.56) and that under fading it was 12–30 ms slower in the mean than AntHocNet. Both readings were artifacts. The harness drew flow start times from a cumulative RNG stream counter that the routing helpers had already advanced by a protocol-dependent amount, so every arm started every flow at a different instant and ##COMMON##'s (flow, seq) keys named packets sent at different times in each arm — the identity match was by index, not by transmission (#459). On the paper scenario's 180 s start window the expected separation between two arms' copies of the same key is startWindow / 3 ≈ 60 s, which at 20 m/s is a different topology entirely. The tables below are the same six cells re-measured on the fixed harness at 20 seeds; the superseded values are kept in #431 and the blast radius across the corpus in #460. Per the #352 rule, do not run sweep_summary --vs across that commit.

The approximation is not hypothetical, and it has now been measured twice — the second time with an instrument that works. The instrument is the identity-matched ##COMMON## set (#308) — the exact (flow, seq) packets all five arms delivered, now genuinely the same transmissions — so nothing below is survivorship:

mobility	channel	anthocnet	aodv	olsr	dsdv	oracle	hop bound
rwp	tworay	1.545	1.460	1.325	1.335	1.300	holds 20/20
ssrwp	tworay	1.516	1.441	1.309	1.317	1.285	holds 20/20
gaussmarkov	tworay	1.586	1.476	1.355	1.364	1.322	holds 20/20
rwp	nakagami	1.560	1.248	1.203	1.083	1.231	fails vs olsr, dsdv
ssrwp	nakagami	1.578	1.258	1.209	1.084	1.238	fails vs olsr, dsdv
gaussmarkov	nakagami	1.633	1.268	1.233	1.104	1.271	fails vs olsr, dsdv

On the three two-ray cells the oracle is below every real arm in every one of the 20 seeds. That is the hop bound holding, un-suppressed, for the first time on a wifi channel — and it is the acceptance bar #431 set for a replacement adjacency rule.

On the three fading cells it is not. The failure is narrow, one-directional and identical across all three: the oracle is above olsr by +0.028…+0.038 and above dsdv by +0.149…+0.167 in 20 of 20 seeds, while beating anthocnet 0/20 and aodv in all but the gaussmarkov cell. Because common is the intersection over all arms, dsdv is routing the same packets in fewer hops than the shortest-path control believes possible — which is only possible if the graph is missing links the radios genuinely have. A median radius does exactly that: every link Nakagami delivers on beyond 373.4 m is invisible to the solver, and the arms with full topology knowledge are the ones that exploit them. That is approx=1 appearing directly in the data, at its measured size.

The latency comparison, on the same ##COMMON## basis (anthocnet minus oracle):

mobility	channel	meanC anthocnet	meanC oracle	Δ	p99C anthocnet	p99C oracle	Δ
rwp	tworay	18.4	1.9	+16.5	302.2	14.6	+287.6
ssrwp	tworay	18.2	1.9	+16.3	289.1	14.4	+274.7
gaussmarkov	tworay	16.9	2.1	+14.8	280.1	15.3	+264.8
rwp	nakagami	39.3	30.9	+8.4	563.9	1015.9	−452.0
ssrwp	nakagami	40.6	32.4	+8.2	571.0	1024.1	−453.1
gaussmarkov	nakagami	43.4	40.4	+3.0	626.5	1272.2	−645.6

The oracle is now faster in the mean in all six cells, fading included — the phase-3 mean inversion was the instrument. What survives is the tail inversion under fading: the oracle's common-set p99 is 1.02–1.27 s against AntHocNet's 0.56–0.63 s on identical packets. It is real, and at roughly half the size phase 3 reported.

The mechanism is one graph mismatch with two signs, and the two signs now separate cleanly by channel.

Missing links. The derived radius is a threshold on a continuous law, so links beyond it that nonetheless deliver are invisible to the solver and the control routes around edges that work. This is the only sign active on two-ray, and there it is small enough that the bound still holds in 20/20 seeds. Under fading, with a median radius, it is what produces the olsr/dsdv hop residual above.
Admitted links that are effectively absent. A nominally in-range neighbour can be in a deep Nakagami fade, and the control, which evaluates no propagation model, routes over it anyway. The packet is not lost; it is retried until it arrives very late. That lands in the tail and not in the mean, which is exactly the shape of the surviving p99 inversion.

Under two-ray only the first sign is active and it is cheap: the oracle wins latency by an order of magnitude in the mean and ~19× at the common-set tail, a margin no plausible graph correction closes.

The two acceptance constraints pull in opposite directions, which is the sharpest statement this page can make about the limit. Shrink the radius and the oracle's hop count rises above its subjects — today's fading failure. Grow it and the oracle's PDR falls below them: the refuted link-budget rule recorded in ns3/oracle/README.md made 2440 of 2450 edges adjacent and delivered 30.4 % PDR, a control its own subjects beat. Two-ray has a radius in the feasible band between those failures. Whether a fading channel has one at all is open, and if it does not, the answer is a probability-weighted (ETX-shaped) graph or an accepted-and-scoped limit — not further radius tuning.

How these numbers may be quoted: a delivery bound everywhere, latency and hops on two-ray¶

Delivery — robust in all six cells. Re-verified on the fixed harness: oracle PDR 96.00–98.20 % against the best real arm's 85.58–92.42 %, with zero violations in 120/120 seeds. A different-but-reasonable adjacency rule moves that by a fraction of a point and cannot move the margins, so the gap decomposition and every PDR conclusion on this page stand.
Latency — two-ray only. The mean and tail advantages hold there with the graph error working against the oracle (missing links only, so the true optimum is faster still — the bound is conservative). On the fading cells the error is not signed and the surviving p99 inversion is direct evidence of that, so no latency bound may be quoted from the three Nakagami cells — neither for nor against AntHocNet. The mean figures there are reportable as a measurement but are not a bound.
Hop count — a bound on the two-ray cells, and not on the fading cells. This is the change #457 and #459 bought: the assertion fires and passes at 20 seeds on all three two-ray cells. On the fading cells it fires and fails, and the failure is the measured size of the median-radius approximation rather than a defect in any arm — quote it as that, not as "dsdv beats optimal routing".
delay99 across arms is a separate trap, and the oracle's fading tail is the clearest instance of it in the corpus. It is a general metric rule, not an oracle quirk — see metrics.md.

The ranking-stability statement¶

Scoped, because one ranking is stable and another is not — and the re-baseline moved the boundary.

Stable — delivery and overhead. The delivery ordering is anthocnet > olsr > aodv > dsdv in all six cells, with the first-vs-second gap exceeding the summed per-arm CIs in every cell, and AntHocNet's paired lead over AODV significant in every one. The overhead ordering (olsr < dsdv < anthocnet < aodv) likewise holds in all six. Neither claim depends on the mobility model or the channel. But the magnitude changed: the paired AntHocNet−OLSR delivery lead narrowed from +4.25…+11.06 pp at 1 s to +1.70…+4.16 pp at 200 ms (all still significant, max p = 9.5 × 10⁻⁴; tightest cell ssrwp-tworay at +1.70 pp against a summed per-arm CI of 1.25). The stability statement survives the flip; a claim quoting its old size does not.

Not stable — the tail. At 1 s this section reported a clean inversion: OLSR → dsdv → aodv → anthocnet under two-ray, aodv → anthocnet → dsdv → olsr under fading. At 200 ms the invariant part is OLSR: best tail under two-ray (22–69 ms, a factor of ~25 ahead) and worst under fading (2685–2961 ms). Its jitter moves the same way (9.6–12.6 ms → 164–189 ms), so the two are one effect rather than two. AntHocNet's position, by channel:

channel	`delay99` at 200 ms
two-ray	olsr (22–69 ms) → anthocnet 2nd (ssrwp, gaussmarkov; tied with dsdv in rwp) → dsdv → aodv. The 1 s "…anthocnet last" ordering is obsolete.
Nakagami	aodv-or-tie first: aodv nominally ahead in rwp/ssrwp (paired +81/+73 ms, p = 0.012/0.033 — marginal), anthocnet nominally first in gaussmarkov (paired p = 0.7 — a tie) → dsdv → olsr worst. The 1 s "aodv wins the fading tail" claim degrades to aodv-or-tie.

Consequence: a tail claim that does not name its channel is unsupported. That survives the re-baseline unchanged — OLSR's factor-of-~25 inversion carries it on its own. What the re-baseline retired is the claim that AntHocNet's tail is last anywhere: at 200 ms the #21 deficit against AODV persists only as a marginal edge in two fading cells, and under two-ray AntHocNet's tail now beats AODV's outright.

Mobility is the weaker axis. Across the three mobility models at fixed channel, the paired ΔPDR (vs AODV) moves by ≤ 2.4 pp and no ordering changes anywhere. Steady-state RWP lands essentially on classic RWP (+0.37 pp two-ray / −0.24 pp Nakagami, p = 0.68 / 0.72 — not significant), which is a useful negative: at this scenario the speed-decay transient the steady-state model exists to remove is not what drives the result.

Provenance¶

main @ a1daa7a (the #411 merge commit), image ghcr.io/danieljoppi/ns3:3.42-opt, runs=20, time=900, areaX=1500, speed=20, protocols=anthocnet,aodv,olsr,dsdv, ReconvHoldCap=200 ms (the shipped default — no extraArgs override).

mobility	channel	pause	run ID
rwp	tworay	30	`31618105814`
ssrwp	tworay	30	`31618110426`
gaussmarkov	tworay	0	`31618116286`
rwp	nakagami	30	`31618108070`
ssrwp	nakagami	30	`31618114426`
gaussmarkov	nakagami	0	`31618118283`

The oracle arm (phase 3) was measured on main @ 40b434d — the #419 merge commit that adds contrib/oracle — with the same image, runs=20, time=900, areaX=1500, speed=20, gaussmarkov at pause=0, and protocols=anthocnet,aodv,olsr,dsdv,oracle:

mobility	channel	run ID
rwp	tworay	`31807666381`
ssrwp	tworay	`31807668353`
gaussmarkov	tworay	`31807670848`
rwp	nakagami	`31807672820`
ssrwp	nakagami	`31807676290`
gaussmarkov	nakagami	`31807678924`

All six scenario_check.py results runs exit 0 with zero FAILs, 25 checks per cell, and the oracle's positional ##RUN## column mapping was validated against the harness's own # stddev oracle line. The four baseline arms in these six blocks are byte-identical to the six phase-1 blocks above, which is what licenses reading the two dispatches as one table — see Why the oracle columns compose. Full readout, including the assertion record and the anomalies: #415 (comment).

Every cell self-identifies through its ##CONFIG## row (#369) — cell identity is read from the data, not from dispatch order — and its ##PROV## line pins commit=a1daa7a (#365). bench_parse column-mapping self-checks passed (20 checks) on all six cells. scenario_check.py results found nothing outside the known classes: the standing 3-per-cell #230 path-diversity instrumentation FAILs (aodv/olsr/dsdv, non-blocking), scattered #386 ICMP-re-injection WARNs (one seed each in four cells), and end-of-run-queue drop-cause overshoots ≤ +2.26 pp (WARN class). No anchor, energy, reordering-bounds, or route-quality failures.

What is deliberately not published here¶

Two metric families are omitted from every table above, because they are known to be unreadable in these cells. The columns are left empty rather than filled with wrong values — the same rule ##HOLD##/##AIR## follow.

drop_* — broken on the fading cells. #377: drop_chan_pct is a residual, not a measurement, and it goes to −13.77 with ~20 pp unaccounted on every Nakagami cell and every protocol. The two-ray cells are clean, but publishing the family for half a grid would invite exactly the cross-cell comparison that is invalid.
path_div_* / path_entropy_bits — the standing #230 limit. pathWindowS outlives the route, so route replacement reads as concurrent multipath. Diversity remains readable only from the dedicated cell.

Neither affects the numbers on this page: PDR, delay, delay99, throughput and NRL come from FlowMonitor and never touch the drop counters.