Innovation and advanced research — Time-Fabric, frame-coherence and RF physics
A substantial share of Subpico's (imported from Australia) is a unique network security and defense system with capabilities for military/ intelligence/ law enforcement/ critical infrastructure/ etc. The time and resources of the company goes into advanced physics and engineering research in communications. It is the source of the product, not a sideline. We treat the network not as a closed, deterministic IT system but as an open physical system, and we study — and implement — the external forces that act on it. The programme is published at capability level on the Interlace® research portal and is the core innovation we bring to the EU.
The thesis — cyber is closed, communications is open. Mainstream cyber defence treats the network as a closed, tactile system — an assumption baked into MITRE-style
signatures, fingerprinting and standards, all of which are first-order and qualitative: they attach a label and test it against a fixed notion of "normal." But no TCP/IP system is closed, deterministic or reversible; the upper layers carry too many degrees of freedom (TLS 1.3 PFS, encryption, NAT) for a signature to say why or how a session changed — which is exactly why threats manifest as zero-day. To claim "we protect" as though the future were predictable is to mistake an open system for a closed one. Subpico's position is that you need both quantitative and qualitative measurement: quantitative tells you how high the tree is; qualitative describes the tree — but you must know how high in order to describe it meaningfully.
External forces and higher-order time-derivatives (frame-coherent networking). Recall Newton: a system's internal state changes only when an external force acts on it. Subpico is a pioneer in the study — and the realized implementation — of those external forces on distributed communications. We model a remote system (a device, an IP:port service flow, an API, a fleet of field devices) as a rotating reference frame S′ and relate its time-derivatives back to the local inertial observer frame S — the published mapping (dA/
dt)|S = (dA/dt)|S′ + ω × A . This yields higher-order (2nd/3rd-order) metrics over the live variables common to any silicon system — CPU, FPGA, ASIC, MCU, GPU: given a
clean, working comms/API path, the derivatives of connection state, resource and processing across the estate become a quantitative measure of qualitative change — preemptive, actionable intelligence that lets us act rather than react, and that puts the human investigator back in the loop. The Cartesian x/y/z treatment in the published paper is deliberately simple; the method generalises to any set of controllable dimensions and to an n-manifold design.
Time-Fabric — picosecond time as a network primitive. On the same fabric and silicon (SubpicoCAT® DS-1/DS-2/LF2/SC103 FPGA clocking), the Time-Fabric pillar exposes subnanosecond, picosecond-grade time as a service; makes every packet framecoherent — bound to a measurable physical reference frame, so a packet whose measured
timing/phase/kinematic signature contradicts its claimed origin is dropped at the wire, not analysed after the fact; and turns the global probe fleet into a planet-scale measurement instrument. Frame-coherence makes a whole class of L3 header-spoofing attacks physically impossible at the protected edge, and anchors identity and provenance on physics rather than certificates. Target surfaces span 6G fronthaul, quantumnetworking testbeds, satellite-ground handoff and defence.
RF and physical-layer resonance — engineering the medium, not just the packet. Below the IP layer we do genuine RF and materials physics. Our published work quantifies
the two resonances that bracket every radio link: the ferrimagnetic resonance of rare-earth garnet front-ends (YIG and its gadolinium/dysprosium/ holmium-substituted
variants, tuned by the Kittel relation) and the molecular absorption resonance of the atmosphere (the 22 GHz water-vapour and 60 GHz oxygen lines), applying ITU-R models to
drive climate-aware C/X/Ku/K/Ka band selection for telemetry that cannot drop. For defence and critical communications this means live processing tuned to harmonic resonance, and front-ends whose rare-earth composition is matched to the mission environment — the rare-earth type and quantity map directly to the capability of the system in question. This RF/physical-layer and supply-chain-materials competency sits largely outside the ENISA taxonomy (see Gaps below), yet it is decisive for sovereign,
satellite and defence infrastructure.