אִם יִרְצֶה הַשֵּׁם

article

Three Layers, Three Physics Regimes

The MAFAT RFI deadline is days away. The fiber-optic FPV is not one threat; it is three problems sharing a munition. Armored maneuver in valleys, static rear-area points, and dismounted infantry each face a different ambient environment, a different reaction-time budget, and a different sensor that actually works. Any single-architecture answer fails at least one of them.

What follows is the case for a three-layer response, with the open engineering questions named on the page rather than buried.

Why one architecture cannot cover all three regimes

Acoustic detection works at a forward operating base. It is deaf next to a running tank. A diesel powerpack and tracks on rocky ground put 90+ dB into the near field of any microphone array bolted to the hull. A 10-inch FPV prop at 100 meters is well below that floor. The ZVOOK-class story is real, but the regime it covers is static, not armored maneuver.

Radar is the other way around. Micro-Doppler AESA can pick a rotor signature out of foliage clutter at vehicle ranges, and it does not care about engine noise. But fielding a thick radar mesh across every static rear point is expensive in a way that a $500 acoustic node is not.

Different physics, different sensor, different layer.

Layer 1: Armored maneuver vanguard

Sensor: micro-Doppler AESA radar, hull-mounted. The Echodyne EchoGuard class is the reference candidate.

Open validation question. Published Echodyne tests are against Group 1 UAS in clean air. Performance against low-RCS fiber-optic FPV in foliage clutter, co-sited with Merkava emitters (Trophy radar, vehicle EW), is the load-bearing assumption of this layer and is not a solved problem. Any serious submission funds a co-site integration test before committing to procurement at scale.

Soft kill: radar-cued multispectral smoke. Within milliseconds of a treeline break, dispense smoke that blinds the operator's video link. Against an optically-piloted FPV with a human in the loop, two seconds of blindness is often a kill in itself, because the operator overflies, stalls, or loses the picture and cannot reacquire. Per-shot cost is trivial. This is the cheapest high-impact effector on the vehicle and the most overlooked.

Hard kill: programmable airburst from a 30mm class remote weapon station. The EOS Slinger pattern (3P airburst, tungsten fragmentation) is the published reference. Per-engagement consumable cost is in the few-hundred dollar range. A 30mm RWS with 3P is not a miniaturized Phalanx; it is a proven class of effector against Group 1 UAS.

Integration caveat. Mounting a 30mm RWS on a Merkava IV/V is not a checkbox. It contests turret real estate with Trophy launchers, the commander's independent thermal viewer, and the loader's hatch. This is a year of mechanical engineering. Start now.

Reaction-time budget. Whether a 50ms deterministic engagement loop is achievable on this platform (radar dwell, track confirmation, classification, slew, time of flight) is the second open engineering question of Layer 1.

Layer 2: Static and logistical area denial

The acoustic environment of a forward staging area is workable, which is why this layer can run cheap.

Sensor: ZVOOK NW0 or equivalent acoustic mesh, hardwired by buried fiber rather than wireless link, so the defensive network itself cannot be jammed. 360-degree coverage, 150 to 450 meter range, 15 to 20 watts a node, around $500 per node.

Cueing: acoustic flags an azimuth and rough range. EO/IR head confirms and classifies. Visual or LIDAR seeker on the interceptor handles terminal.

Hard kill: Novasky CY200 class interceptor, containerized launcher, 300 km/h kinetic ram, no shaped charge, per-kill consumable around $1,500. Endurance is short (about ten minutes), which is fine because this is launch-on-cue, not patrol.

Decoy flooding is the open problem here. A harmonic-discriminating classifier handles a clean signal well. It handles fifty cheap noisy quadcopters launched as cover for the actual munition badly. Adding an elevated EO/IR mast as a tiebreaker trades acoustic-flooding vulnerability for line-of-sight and mast-survivability vulnerability; the right answer depends on terrain and threat density. The submission should name this trade, not finesse it.

Layer 3: Dismounted infantry terminal defense

Effector: SMARTSHOOTER SMASH 3000 class fire control on the squad's existing rifles. AI-aided aimpoint correction against a Group 1 UAS at close range. Fielded technology with a real combat record. Saturate it.

This is the cheapest layer to field and the fastest. There is no good reason any frontline squad lacks it.

Available under appropriate authority: source eradication

The fiber spool is a feature for the operator and a vulnerability. The cable physically points home. A VTOL asset (Hunter Eagle class) can backtrack the spool to the launch nest and service it.

This is not part of the standing architecture. Backtracking a fiber across the Lebanese border is a counter-battery mission with rules-of-engagement, sovereignty, and proportionality questions that the architecture does not adjudicate. Treat it as a capability commanders can authorize on a mission-by-mission basis, not as a default kill-chain element.

The candidate field

Trophy's detection chain was built for RPGs and ATGMs, not low-RCS FPVs arriving at fifteen meters per second from above. Recalibration does not fix a sensor mismatch this fundamental. Iron Beam belongs in the static rear; it does not travel with the maneuver force.

The economic argument

Per-shot consumables are not the whole cost. A more honest accounting:

Consumables per engagement: $450 (Slinger), trivial (smoke), $1,500 (CY200), trivial (SMASH 3000). Against a $500 attacker, only the CY200 loses on a per-shot basis, and only by a factor of three.

Fixed costs amortized. Acoustic mesh, radar fit-out, launcher inventory. For a representative static node with roughly $50K in fixed costs, break-even against a $500 attacker is around 30 engagements at $1,500 per kill. The architecture is sustainable in steady state across a long campaign. It is not sustainable as a short-campaign answer because the fixed costs do not amortize fast enough. The threat profile (large adversary magazine, long horizon) is what makes the math work.

Cope cages: the floor, not the answer

Cope cages detonate the shaped charge above the turret roof. They fail against side and underside attacks. They are what you have until the RFI answer is fielded, not part of the answer itself.

Red team: where each layer breaks

Layer 1 (radar): low-RCS FPV variants, terrain-masking flight profiles below the radar horizon, co-site interference exploitation timed to vehicle emitter cycles. The deterministic loop is the single point of failure.

Layer 2 (acoustic): quieter props, lower-RPM flight, and decoy flooding. Flooding is the immediate worry.

Drone-on-drone interceptors: sacrificial first waves to deplete the magazine, terminal-phase evasion, swarm timing tuned to launcher reload.

Source-eradication missions: decoy spools, hardened nests, and operators who relocate after each launch.

Forecast: the architecture buys 12 to 18 months of asymmetric advantage at the system level, with the radar layer probably adapting first (low-RCS designs already exist in adversary inventories) and source-eradication slowest (operator relocation requires real doctrinal change). Plan procurement on that horizon.

Phasing

Doctrine before hardware. The next 90 days do not depend on procurement.

Phase 1 (0 to 3 months): mandate overhead cover for medevac, recovery, and resupply operations. Saturate dismounted squads with SMASH 3000. Complete the cope-cage retrofit program already in progress. Nothing here requires an RFI award.

Phase 2 (3 to 6 months): prototype radar-cued multispectral smoke on a Merkava platoon. This is the cheapest Layer 1 capability and the fastest to field. Begin acoustic mesh deployment around static logistics nodes under existing procurement authorities.

Phase 3 (6 to 12 months): integrate 30mm RWS on Vanguard armor (long mechanical engineering tail). Roll out the ZVOOK-to-CY200 hardwired network for rear-echelon defense. Fund KAiDEN-class R&D as the 2027-2028 refresh.

What to submit

1. Vanguard: micro-Doppler radar (with co-site validation funded), radar- cued smoke, 30mm programmable airburst (with integration timeline named).
2. Static: hardwired acoustic mesh, EO/IR confirmation, drone-on-drone interceptor (with decoy-flooding mitigation named).
3. Infantry: SMASH 3000 saturation.

Source eradication stays available under commander authority, separate from the standing architecture.

The sensor and effector technologies exist. The bottlenecks are platform integration, doctrine, and honest accounting of what each layer does and does not cover.