All Military Technology is Relative Against a Reactive Enemy

04/30/2026
By Ed Timperlake

A conceptual “white paper” think piece.

The meaning of U.S. “kill web” warfighting capabilities is captured in our book: Maritime Kill Web Force Making. At its core, the American way of war is about linking every sensor to every shooter across all domains—air, sea, land, space, and cyber. That is the integrated tactics and training the U.S. Navy is currently perfecting.

As co-author Robbin Laird has recently observed, foreshadowing his forthcoming book “Forces in Motion: Essays on the Transformation of Western Society” — a Kamikaze drone costs roughly $5,000 to manufacture, while a Patriot interceptor missile costs $4 million. That brutal arithmetic demands better and cheaper countermeasures, and the offensive-defensive technology cycle must now accelerate to meet it.

USN layered weapon engagement zones, beginning with the classic “shoot the archer, not the arrows” approach to targeting launch sites, have been refined over decades. Kinetic close-in systems such as the Phalanx CIWS, directed energy research including lasers at varying power levels, and EMP pulse programs all show genuine promise. The central question is whether the pace of that research is fast enough.

Systems like the F-35 Lightning II can function as precision sensing nodes within a kill web, collecting, fusing, and sharing targeting data at machine speed through a payload utility function (Pu) that drives both target acquisition (TA) and target engagement (TE). The result is resilient, distributed, and highly connected decision-making operating at the speed of light.

Making all of this work with robust redundancy and reliable weapons remains the central challenge. Adding to that challenge is the hard fact that the People’s Liberation Army (PLA) knows a great deal about the American way of war. Yet despite the U.S. not hiding most of its capabilities, there are always unwelcome surprises in store for PLA concept-of-operations planners and, of course, the reverse is also true.

This action-reaction dynamic is captured perfectly in the film Patton, when General Patton, played by George C. Scott, describes his first engagement against German General Rommel: “I read your book”—a reference to Rommel’s Infantry Attacks (Infanterie greift an). It is equally obvious that PLA leadership, across the Army, Navy, Air Force, and Strategic Rocket Forces, are reading emerging U.S. kill web doctrine with the same intent.

The PLA’s “System Destruction Warfare”

Thanks to AI-driven capabilities, an asymmetric opportunity now exists to complicate a major dimension of the PRC’s investment in space warfare against U.S. forces. Negating GPS satellite systems is one of Beijing’s core tactical and strategic objectives. Just as Laird’s brutal cost-benefit asymmetry example illustrates, there may be an opening to go asymmetric against a centerpiece of PLA doctrine: “system destruction warfare.”

PLA doctrine targets the system itself, not just individual platforms. Rather than simply shooting down every aircraft or sinking every ship, the PLA aims to:

  • Disrupt data links and blind ISR (intelligence, surveillance, and reconnaissance) capabilities.
  • Compromise command systems through cyber intrusion.
  • Spoof sensor inputs via the PLA Strategic Support Force.

The strategic logic is clear: if the network fails, the kill web collapses into isolated, uncoordinated units. The PLA has invested heavily in jamming communications and data links. Their ultimate goal is to turn a connected force into a confused, delayed, and fragmented one.

They are especially focused on destroying U.S. space-based enablers, because kill webs rely heavily on satellites for GPS, communications, and missile warning. PLA counters include anti-satellite weapons, co-orbital systems, and ground-based jammers. The prevailing assumption is that without space support, kill web coordination degrades dramatically.

On the weapons side, the PLA continues building large missile inventories and now deploys drone swarms in coordinated multi-axis attacks, designed to generate more targets than any network can process or intercept.

The American Counter: Adaptability

But not so fast. The PLA’s offensive effort has triggered exactly the kind of counter-response that American scientific, military, and industrial ingenuity excels at generating.

In his seminal history of World War II, Inferno, Max Hastings observes that after early setbacks, the U.S. and Royal Navies became the finest unified fighting forces of the war. The Royal Navy entered the conflict as the most experienced and tradition-rich naval force in existence, but was stretched dangerously thin. The United States became the dominant naval power through industrial-scale logistics and, the key word Hastings correctly identifies, adaptability.

That adaptability now finds expression in an emerging component of kill web capability: AI-powered MAGNAV (Magnetic Navigation), enhanced by the mathematical frameworks of Kalman filtering and Bayesian decision theory for precision global positioning without GPS.

MAGNAV, Kalman Filtering, and Bayesian Decision Theory

MAGNAV is a map-based navigation technique that uses variations in the Earth’s magnetic field to determine position—especially when GPS is unavailable, denied, or unreliable. In a sense it is a back-to-the-future concept: think of a compass, but with orders of magnitude greater precision and, critically, entirely passive internal sensing.

The historical precedent is instructive. Before satellite-assisted strategic targeting was available, Kalman filtering was already managing “drift” in the Inertial Navigation Systems (INS) guiding U.S. ICBM deterrence forces. The math has been battle-tested at the strategic nuclear level.

The building blocks, in cascading order, work as follows:

Layer 1 — Prediction

INS projects where a platform should be next, providing continuous motion estimation.

Layer 2 — Measurement

MAGNAV provides an independent environmental “fix” by comparing measured magnetic intensity against a stored map to infer position. Other sensors may contribute additional data.

Layer 3 — Estimation (Kalman Filter)

The Kalman filter fuses INS predictions with MAGNAV measurements and optionally GPS or other sensors—into a continuously updated best estimate of position and velocity. Rather than trusting any single source, it blends them all.

Layer 4 — Decision (Bayesian Logic)

Bayesian reasoning evaluates the consistency of sensor inputs, detects anomalies (such as spoofing or interference), and adjusts sensor weighting accordingly. It updates confidence levels in real time as new data arrives.

Where AI Fits In

AI does not replace these methods: it enhances them. Specifically, AI contributes:

  • Pattern learning: improves magnetic map matching in noisy or geographically sparse regions.
  • Adaptive modeling: dynamically tunes noise parameters within the filter.
  • Anomaly detection: identifies subtle inconsistencies that fixed models or human operators might miss.
  • Sensor management: determines which inputs to prioritize under changing operational conditions.

AI functions as a meta-layer that improves robustness and adaptability, creating a distributed, self-correcting passive navigation node, delivering navigational resilience under denial conditions.

The PLA’s GPS War and Its Potential Undoing

Consider the scenario: GPS becomes unreliable due to PLA jamming or spoofing. The system responds in layered fashion:

  • INS predicts motion but begins to drift.
  • MAGNAV provides a position cue based on geophysical terrain.
  • Kalman filter blends both into a stable estimate.
  • Bayesian logic detects GPS inconsistency and reduces its influence accordingly.
  • AI refines the magnetic map match and maintains performance over time.

Result: Navigation continues without dependence on any single vulnerable source.

This combination, redundancy, resilience, and adaptivity, supports Assured Positioning, Navigation, and Timing (APNT): systems designed to remain reliable even when conditions are contested, degraded, or actively denied.

Proven Precedents and Converging Systems

TERCOM (Terrain Contour Matching) guidance is already battle-tested in cruise missile accuracy, using a radar altimeter to follow an elevation map. The Tomahawk’s lethal precision comes from combining TERCOM with the optical DSMAC (Digital Scene Matching Area Correlator) for visual confirmation at terminal guidance. Adding MAGNAV creates a third, independent layer: an invisible geophysical fingerprint. Over open ocean before feet dry, MAGNAV would dominate navigation, and it is inherently immune to RF spoofing.

Similarly, the AIM-9 Sidewinder, one of the most effective fire-and-forget air-to-air missiles ever built, has been continuously improved over more than half a century and remains a first-choice weapon in air-to-air engagements. The pattern is consistent: existing systems improve through layered technological enhancement, not wholesale replacement.

Conclusion: Time to Mix It Up

Researching non-GPS combat effectiveness for both platform navigation and long-range weapon targeting accuracy, using the conceptual framework outlined here, may render obsolete a massive PLA investment in systems designed to blind U.S. and allied forces. GPS satellite navigation is currently both a strategic vulnerability and a strategic dependency.

The offensive-defensive enterprise, now powered by genuine AI capability, must move.

The time to mix it up is now.