This article examines Ukraine’s shift to drone-centred warfare, lessons from Pokrowsk, and the resulting demands on surveillance and EW.
In a December 2025 interview by Katrin Eigendorf (ZDF Heute), Oleksandr Syrskyi, newly appointed as army chief, offered a stark assessment of how the war in Ukraine has transformed. He noted that at the outset, combat was fought in a traditional way — with offensive battalion troops, massed infantry, and tank columns. But today the paradigm has changed rapidly: “decision-making is centred around drones,” he said. According to him, drone missions now account for roughly 60 percent of “Feuerwirkung” — the destructive effect on targets — while artillery’s share has decreased to about 40 percent. In average according to Syrskyi, Ukraine has faced between 130 and 400, and sometimes as many as 700 drone-based attacks, often paired with missile strikes against industrial, infrastructure, or civilian targets (further reading: Australian Army Research Center, Hudson.org).
As a result, air-defence systems — interceptors, radar networks, missile shields — are now indispensable. The conflict around Pokrowsk exposes the harsh consequences of this shift. Military-industrial systems may be robust; but in Pokrowsk, the defenders now struggle with weather, “fog of war,” and limited visibility — complicating reconnaissance of both drone activity and small ground-force infiltration.
This context forces a reconsideration of the old dichotomy between “air war” and “ground war.” Today, both are deeply intertwined. A modern combined-arms engagement demands that drone swarms, missile salvos, artillery fire, and ground manoeuvres are coordinated — or at least carefully separated — in time and space. For engineers and EW technologists, this means rethinking how surveillance, detection and decision-support are organized.
The paradigm shift in military surveillance and target detection
As drone strikes and missile attacks have become the primary vector of destructive force, early and reliable detection has become vital. Two major challenges arise: first, detecting individual threats (drones, missiles) in time; second, detecting their launch bases or “drone infrastructure” — the nodes that perpetuate repeated attacks.
a) Early detection of attacks
Modern conflict regions like Ukraine illustrate that many attacks come from small, inexpensive unmanned aerial vehicles (UAVs) or loitering munitions rather than from traditional aircraft (Australian Army Research Center, Journal of Political Science, Understanding War)
This raises major design-engineering and EW issues:
-
Conventional radar systems often struggle to detect small, low-radar-cross-section UAVs — especially when they fly low, or take advantage of terrain masking or urban clutter (sitejournal)
-
Multi-sensor detection becomes a necessity: combining radar, infrared/EO, RF emission monitoring, acoustic and possibly visual detectors. The challenge is to fuse data from disparate sensors in real time, enabling reliable classification (friend/foe/neutral), tracking, and targeting. Surveys of anti-UAV methods highlight that real-time performance and detection of stealthy or swarming UAVs remain significant open problems (Cornell University).
-
EW-hardened communication lines and low-latency data links become critical, to enable rapid sensor-to-shooter cycles especially when drones serve as forward eyes for artillery or air defense.
Thus, successful defense is no longer just about firepower, but about situational awareness — the ability to detect, classify, and respond before damage is done.
b) Detecting drone-launch infrastructure and “enemy drone architecture”
A further complication arises from the need to locate and degrade drone-launch bases, storage sites, maintenance hubs, and staging areas — the infrastructure that sustains repeated drone strikes. Detecting these “nodes” can be more decisive strategically than focusing only on intercepting individual UAVs.
Several technical approaches are relevant:
-
Passive radar and RF-signal monitoring: By observing changes in ambient electromagnetic signals (cellular, broadcast, Wi-Fi, LTE, other comms), passive detection systems can infer unusual activity, such as drone control uplinks, preparatory emissions, or communications bursts typical for launch operations. Such passive systems may detect drone operators even when drones themselves remain below radar threshold. This approach is discussed in recent UAV detection surveys (Cornell University).
-
Multimodal sensors + sensor fusion: Combining signals intelligence (SIGINT), imagery (EO/IR), radar, acoustic sensors, and perhaps thermal sensors to identify signatures of drone base activity: vehicle traffic to hides, transient heat signatures from recharging, small launch-pad footprints in rural or semi-urban areas.
-
Satellite surveillance, including SAR and pattern-analysis: Synthetic-Aperture Radar (SAR) imagery from space can detect changes in terrain, presence of shelters, temporary hangars, and vehicle movement even at night or through cloud cover. Repeated SAR passes, when analyzed through change-detection algorithms, can reveal the buildup of drone-launch infrastructure. For sustained surveillance, satellite-enabled change detection may be indispensable. Several analysts recommend integrating satellite-based ISR into modern drone-warfare threat-detection architectures (CSIS.org, Understanding War).
-
Data analytics and behavioral pattern recognition: Beyond raw sensing, software tools must flag patterns typical of drone-economy activity: frequent small-vehicle traffic, energy consumption spikes, thermal anomalies, logistic trails. Machine learning and anomaly detection become useful, especially when numbers of potential sites are high and human monitoring impractical.
Relevance for SkyRadar Defence Series 1–5
The shift from battalion-centric manoeuvre to drone-centred decision loops requires training systems that can reflect the complexity of modern detection, classification, and counter-stealth operations. The simulator extensions in SkySim and the FreeScopes Military Detection Feature Set offer a framework that aligns closely with the operational realities described in Pokrowsk. They provide the building blocks for training, experimentation, and engineering analysis in environments where drones, missile threats, and low-observable targets coexist.
Series 1–2: Radar, RCS Awareness, and Multi-Band Operations
The early stages of the Defence Series introduce the fundamentals of radar operation. These modules are now extended by the Military Detection I package, which trains operators to understand how frequency, aspect angle, and target shaping influence radar cross section. Multi-band simulation — switching between L-, S- and X-band — allows users to explore how low-observable targets present different signatures depending on wavelength. This capability mirrors a key requirement of modern drone and counter-drone operations, in which small UAVs or stealth-designed loitering munitions often appear or disappear depending on frequency.
Operators also learn to apply frequency agility and band diversity to expose targets that rely on narrowband stealth optimisation. These elements provide a technological foundation for understanding the reconnaissance challenges Syria and Pokrowsk are facing.
Series 3: Electronic Warfare, Passive Sensing, and Multi-Static Architectures
The Defence Series’ electronic-warfare modules are expanded by Military Detection II, which introduces bi-static and multi-static geometries, passive radar, and ELINT analysis. These capabilities reflect the increasing need to detect drones and small infiltration teams under conditions where classic monostatic radars struggle — for example, fog-covered urban terrain such as Pokrowsk.
The passive-radar and ELINT features allow operators to detect targets using civilian broadcast illuminators, analyse intercepted emissions, and characterise hostile radar or communication systems through pulse descriptor words (PDWs), de-interleaving, and correlation with synthetic threat libraries. Multi-sensor fusion algorithms — including gradient-boosted models — demonstrate how combining radar geometry, passive sensing, and emitter analysis increases detection probability when SNR is low. These tools map directly onto the real-world problem of locating drone launch infrastructure, control links, and short-range C2 emissions.
Series 4: Communications, Swarm Detection, and Data-Fusion Chains
As decision loops accelerate, communication resilience becomes central. The simulator enhancements in Military Detection III introduce operators to the behavior of drone swarms, low-RCS objects, and coordinated multi-platform activity. Micro-Doppler rotor signatures, track-before-detect (TBD) algorithms, and CNN-based spectrogram classification reflect the detection challenges of small UAV masses and FPV-type drone strikes — now common in Ukraine.
Swarm-pattern identification and clutter discrimination help prepare trainees for environments where hundreds of drones may be active simultaneously. These modules reflect the transition from isolated target detection to network-scale sensing, which is essential for the processing of drone-intensive attack patterns.
Series 5: Integrated Simulation of Full-Spectrum Counter-Stealth
The Defence Series concludes in an integrated environment where modern threat profiles can be explored comprehensively. Military Detection IV extends the system into EW-rich, multi-sensor, and counter-stealth scenarios. Operators learn to combine:
-
multi-band radar returns,
-
multi-static geometry,
-
micro-Doppler cues,
-
SAR/ISAR change-detection imagery, and
-
rule-based or AI-driven fusion models.
This integrated approach prepares them for environments where drones, loitering munitions, and stealth UAVs use terrain masking, deceptive jamming, or coordinated swarm dispersion — tactics increasingly visible on the Eastern Front. The ability to simulate deception jamming, classification under low SNR, and dynamic target splitting reflects real battlefield conditions described by military analysts.
Forward-Looking Modules (2027): Polarimetry, Passive Pol Radar, and Advanced Signature Analysis
SkyRadar's roadmap plans future add-ons — such as polarimetric radar, direction-finding networks, interferometric wake detection, and plasma-signature tracking. These modules support training in the next generation of detection problems, including hypersonic threats and EMCON-optimized stealth vehicles. Their relevance to the Pokrowsk scenario lies in the increasing necessity to detect low-observable targets even when classic RF signatures vanish.
Synthesis
The ongoing evolution of SkySim and FreeScopes makes it possible to train operators and engineers for the kinds of surveillance and counter-drone challenges that now define modern warfare. The modular expansions — RCS analysis, passive sensing, ELINT, swarm analysis, and full-spectrum counter-stealth — reflect precisely the multi-layered detection grid required when drones, missiles, and infiltration teams operate concurrently, often in degraded visual conditions such as those found in Pokrowsk.
Stay connected with our ongoing publications on Electronic Warfare and Radar Technology.
