The Architecture of Defense: Sensor Fusion in Modern Counter-Drone (C-UAS) Systems

1The C-UAS Kill Chain: From Detection to Assessment

A modern Counter-Drone System (C-UAS) is not a single device; it's a complex, integrated architecture that governs a sequence of events known as the "kill chain." This chain consists of seven critical stages: Detection, Identification, Tracking, Classification, Decision, Defeat, and Assessment. These stages don't always follow a strict linear order — advanced systems run many functions simultaneously, continuously feeding data back to improve situational awareness and response decisions.

Detection is the stage of determining the presence of an object in the airspace, and it doesn't provide information about the object's identity. Radar detects the reflection of radio waves off a physical object, providing range, bearing, and altitude data. RF detectors passively listen for the communication signals a drone or its controller emits. Electro-optical (EO) and infrared (IR) sensors capture visual and thermal signatures, while acoustic sensors listen for rotor and motor noise. Each method has its own particular strengths and environmental dependencies — which is why no single technology is sufficient to provide reliable coverage in every weather condition.

The Identification stage tries to determine what the object is. For example, an RF sensor can identify the type of signal being emitted and link it to a known drone model or controller brand. An EO/IR camera, meanwhile, lets an operator visually confirm the object's shape and size, distinguishing it from birds or other flying objects.

Once an object is identified as a threat candidate, Tracking begins — the target's position, speed, and trajectory are continuously estimated over time. Active Electronically Scanned Array (AESA) radars like Raytheon's KuRFS stand out for their ability to simultaneously track multiple targets even in crowded environments. Advanced tracking algorithms also analyze "micro-Doppler" signatures — subtle variations in the radar signal caused by rotating parts like propellers — to help distinguish a drone from a bird or other object.

The Classification stage categorizes the threat: "Is this a hostile mini-UAV, a civilian hobby drone, or a friendly asset?" AI-assisted classification algorithms can analyze data from multiple sensors to probabilistically classify objects. ASELSAN's DRONEDEF concept explicitly integrates AI into this process for rapid threat assessment.

After classification comes the Decision stage — which takes place within the system's Command-and-Control (C2) framework. C2 evaluates factors such as proximity to a protected asset, the drone's behavior, and available countermeasures to select the most effective and proportionate effector in the arsenal (jammer, laser, projectile, or another weapon type).

Finally, the Assessment stage determines whether the defeat action was successful — the critical feedback loop that closes the chain. This assessment can be done by re-tracking the change in the object's movement, detecting debris with radar, or analyzing imagery from an EO/IR sensor. Some platforms, like Lockheed Martin's Sanctum system, are designed with modular architectures that naturally support rapid assessment and follow-up actions.

2Sensor Technologies: A Spectrum of Strengths and Limits

Modern C-UAS architectures rely on complementary sensor technologies to build a comprehensive picture of the battlespace. No sensor is universally effective, which is why architectures are designed so that one technology's strength compensates for another's weakness.

Radar is a core component of many C-UAS systems because it can detect physical objects at long range, day or night, even in adverse weather. However, detecting small drones with a low radar cross-section (RCS) is a major challenge; their weak reflections can blend with ground clutter, especially at low altitude. Raytheon's KuRFS radar uses Ku-band frequencies, which offer high resolution suited to identifying small objects. But radar alone cannot positively identify a target's nature or intent — it only confirms the presence of a physical object.

RF Detection offers a complementary approach by passively monitoring the electromagnetic spectrum emitted by drones. Systems like Boğaziçi Defense's ILTER J350 are designed to scan wide frequency bands (400–8000 MHz) to detect and jam these signals. But RF-based architectures have clear limits: they're ineffective against fully autonomous drones flying pre-programmed routes without continuous communication or GNSS updates. The most significant challenge is the emergence of fiber-optic-controlled FPVs, where video and control data travel over a thin physical cable — this renders conventional RF detection and jamming entirely useless.

Electro-Optical (EO) and Infrared (IR) sensors provide visual and thermal data critical for identification and verification. However, their effectiveness depends heavily on atmospheric conditions; fog, rain, smoke, and heavy camouflage can seriously degrade performance. For this reason, EO/IR systems are rarely used as the primary detection sensor; they're more valuable as a secondary sensor that verifies tracks reported by radar or RF systems.

Acoustic Detection uses microphone arrays to detect a drone's rotor and motor noise. This passive technique can provide direction-finding information even if a drone isn't emitting RF signals or is flying below the horizon; it's especially useful in urban environments where other sensors suffer from severe clutter. However, it's extremely sensitive to ambient noise such as traffic, machinery, and wind, with a high potential for false alarms.

Effective C-UAS architectures mitigate these individual weaknesses through sensor fusion — the process of merging data streams from multiple sensors into a single track or Common Operational Picture (COP). For example, a track initiated by radar can be correlated with RF sensor data revealing its signal source, then handed off to an EO/IR turret for visual confirmation. This multi-sensor correlation greatly increases confidence in the target's identity and reduces false alarms.

3The ILTER Family: A Case Study in RF-Centric Architecture

The ILTER family, developed by Boğaziçi Defense Technologies, is the leading case study for an RF-centric C-UAS architectural philosophy. The company's public documentation emphasizes an integrated architecture built around the detection and neutralization of the radio-frequency (RF) emissions from unmanned aerial systems (UAS) and their associated ground control stations.

At the center of ILTER's architecture are RF direction-finding (DF) and electronic warfare (EW) capabilities. The flagship ILTER J250 system is described as a medium-weight system combining RF direction-finding with three-dimensional tracking. Its core function is to monitor and precisely locate emissions across common drone control bands (UHF, S, C-band). With up to 250 watts of RF output power, the J250 can both detect and actively counter threats — this dual capacity places it in the soft-kill category. The system claims to provide 360-degree detection and jamming, protecting fixed installations against attacks from any direction.

The inclusion of spoofing as a primary soft-kill method points to a decision layer capable of choosing between simply disrupting a signal (jamming) and deceiving a drone's navigation system with fake data (spoofing). This matters especially against autonomous drones relying on GNSS for waypoint navigation: by spoofing the GNSS signal, the system can make the drone believe it's in a different location, causing it to abort its mission or land safely.

ILTER MRKAS (Mobile Radar-Based Jamming and Deception System) offers a hybrid approach that combines a radar component with RF jamming and deception capabilities — moving the architecture beyond a purely passive RF listener toward an active, sensor-fused system. Platform-specific variants further demonstrate the family's modularity: ILTER Tank provides dedicated 360-degree RF jamming for armored combat vehicles; ILTER Maritime addresses challenges like sea clutter and moving-platform protection; and ILTER FPV Finder focuses on detecting and jamming the specific analog video frequencies used by FPV drones.

4Integrated Layered Architectures: Raytheon, Lockheed Martin, and ASELSAN

In contrast to the ILTER family's RF-centric approach, the most advanced C-UAS architectures are evolving into complex, multi-layered systems-of-systems. These designs deliberately unite different sensors and effectors under a single Command-and-Control (C2) umbrella to create a flexible, resilient defense.

Raytheon LIDS: A Radar-First Architecture

Raytheon's LIDS (Low, slow, small, unmanned aircraft Integrated Defeat System) architecture is a mature, operational system that pairs a high-performance radar with a versatile family of effectors. At its heart is the Ku-band RF Sensor (KuRFS), an AESA radar system designed for persistent 360-degree detection and tracking. The real integration comes from C2's ability to assign different effectors to targets identified by KuRFS: the kinetic interceptor Coyote Block 2, and the non-kinetic Coyote Block 3, which disables drones using electronic warfare. This weapon-agnostic approach lets the operator select the most appropriate response based on the threat and environment. LIDS has been formally adopted by the U.S. Army, with contracts signed to equip multiple divisions.

Lockheed Martin: An AI-Assisted, Multi-Domain Architecture

Lockheed Martin is developing a layered, multi-domain architecture that integrates advanced sensors, effectors, and AI-assisted C2. The cornerstone of this architecture is the MORFIUS high-power microwave (HPM) weapon, designed to neutralize drone swarms with directed electromagnetic energy pulses that disable their electronic components. Complementing MORFIUS is the JAGM missile, a precision-guided hard-kill option for countering longer-range threats. The brain connecting these disparate elements is the Sanctum™ system, an AI-assisted C2 architecture designed for rapid innovation and integration. Sanctum leverages Microsoft Azure cloud infrastructure to create a scalable digital backbone, allowing new sensors and effectors to be added modularly.

ASELSAN DRONEDEF: A Comprehensive Multi-System Concept

With its DRONEDEF program, ASELSAN presents perhaps the most explicit conceptualization of an integrated C-UAS architecture. Rather than a single system, DRONEDEF is a holistic concept that brings together a set of purpose-built systems to create a layered defense. The architecture is built on the principle of sensor-effector integration: the İHTAR system serves as the first detection-and-tracking layer, using RF direction-finding to detect low-altitude FPV threats. Once a threat is identified, C2 can activate one of several effectors. The EJDERHA system, a High-Power Electromagnetic (HPEM) weapon, is designed to disrupt swarms and fiber-optic-controlled drones by targeting their electronics. For situations requiring a physical stop, the ŞAHİN system offers a kinetic solution using a 40mm cannon with programmable airburst ammunition (ATOM). Finally, the GÖKBERK system adds mobile high-energy laser (HEL) capability for precise, non-kinetic defeat. ASELSAN states that these systems are designed to operate in harmony, with AI playing a key role in rapid threat classification and system coordination.

5MKE TOLGA: A Fully Domestic, Layered Short-Range Air Defense System

Developed by MKE (Mechanical and Chemical Industry Corporation), TOLGA is a short-range air defense system built entirely with domestic and national capabilities to protect the layer at 3,000 meters and below within a "steel dome" air defense architecture. Designed for a broad range of missions — protecting ground forces, guarding critical facilities, escorting mobile convoys, defending bases and compounds, and protecting naval platforms — it demonstrates its effectiveness across two distinct scenarios.

In the soft-kill scenario, an enemy drone roughly 3 kilometers away is detected by the mobile radar station and then neutralized as its systems are jammed. In hard-kill scenarios, the system's two fixed 12.7mm weapons, along with vehicle-mounted rotary-barrel 12.7mm and 20mm weapon systems, precisely strike identified drones and fixed-wing UAVs using anti-drone ammunition also developed by MKE. TOLGA's ammunition consists of purpose-built rounds that fragment upon approaching the target.

TOLGA's complete system architecture includes the following components:

• 12.7mm QCB Weapon System and Turret: Neutralizes enemy drones out to a range of 300 meters.
• 12.7mm Rotary-Barrel Weapon System and Turret: Also effective to 300 meters.
• 20mm Rotary-Drum Weapon System and Turret: Extends range to 1,000 meters (1 km).
• 35mm Rotary-Drum Weapon System and Turret: Provides range up to 3,000 meters (3 km).
• Gökbörü AESA Radar, Electro-Optical System, and Signal Jammer: The Gökbörü radar detects and tracks airborne elements in all weather, day and night, using AI-assisted classification technology; the Electro-Optical System performs multi-target detection and moving-target tracking; the Signal Jammer forms the system's soft-kill layer.
• Laser Weapon System: A 20 kW fiber laser delivers three distinct effects — dazzling, blinding, and destruction.
• ENFAL Missile: Electro-optically guided, ENFAL strikes with high precision once it reaches the target; a one-way data link allows for remote access and guidance after launch.
• Acoustic System: Acoustically detects and classifies threatening aircraft, transmitting the collected data at high speed via radio frequency to rear-positioned TOLGA components.

6A Comparative Architectural Analysis of Turkish Systems

Turkey's domestic defense industry has developed a portfolio of C-UAS systems reflecting a strategic move toward integrated, multi-layered architectures. Beyond the RF-centric ILTER family, companies such as Meteksan, MKE, and ROKETSAN have produced systems that unite various sensor types and effectors on integrated platforms.

Meteksan Defense's KAPAN system exemplifies a structured sensor-effector architecture built around the Retinar FAR-AD radar. Detection begins with the Retinar radar, which provides the initial physical track; once a target is acquired, the system uses an electro-optical (EO) system for identification and precise tracking. This fusion of radar and EO data produces a track far more reliable than either sensor could achieve alone. The C2 system then activates countermeasures — the architecture includes an RF jammer for soft-kill options and a laser for directed-energy defeat. The Retinar C2Net data fusion capability points to a central processing unit that correlates sensor data and manages the countermeasure response.

ROKETSAN's ALKA NEW system offers a unique two-layer architecture that combines electromagnetic and directed-energy effectors. The first layer consists of electromagnetic jamming to disrupt a drone's control and navigation systems; the second and final layer is laser-based defeat. This dual-capability architecture allows ALKA to attempt to neutralize a drone with EW from a distance before escalating to a high-energy laser engagement, which requires a clear line of sight and sufficient dwell time. ROKETSAN has confirmed the system has successfully completed live-fire tests.

This analysis reveals a shared trend among Turkey's leading defense industry companies: a commitment to building integrated systems that go beyond single-function devices. Whether it's KAPAN's sensor-fusion approach, TOLGA's kinetic-heavy, multi-layered SHORAD philosophy, or ALKA's dual-layer effector concept, these architectures are designed to offer a more resilient, adaptable defense against an evolving and increasingly complex drone threat environment.

7A Synthesis of Architectural Philosophies and Future Trends

This review, which uses the ILTER family and its international/national counterparts as case studies, reveals a clear evolution from single-function point solutions toward sophisticated, integrated systems-of-systems. A C-UAS's architectural philosophy is now defined not by the capability of any single sensor or effector, but by the depth and intelligence of its integration. Three primary architectural paradigms emerge from this analysis: RF-centric, sensor-fusion-centric, and radar-effector-centric — with the most advanced systems blending elements of all three.

The RF-centric architecture is embodied by Boğaziçi Defense's ILTER family, built on the premise that most tactical drones are vulnerable to manipulation of their electromagnetic emissions. Its strength lies in providing a direct, effective countermeasure against the most common threats, such as FPVs and remotely piloted systems; but its Achilles' heel is its dependence on the target's emissions — which renders it ineffective against autonomous drones relying on internal navigation, and especially against fiber-optic-controlled FPVs.

The sensor-fusion-centric architecture, by contrast, takes a more holistic and layered approach, as demonstrated by ASELSAN's DRONEDEF concept. Rather than relying on a single domain, it deliberately brings together different sensors — RF, radar, EO/IR, acoustic — and effectors — EW, HPEM, laser, and kinetic weapons — under a unified, AI-assisted C2 umbrella. The core architectural value here is redundancy and resilience — if one sensor fails to detect a threat (say, a stealthy drone evading radar), another can succeed (say, by picking up weak RF emissions).

The third paradigm, the radar-effector-centric architecture, as represented by Raytheon's LIDS, prioritizes robust physical detection as the foundation of the entire system. By relying on a high-performance AESA radar like KuRFS, it guarantees detection of a physical object regardless of its emissions profile. Integration is then achieved by pairing this radar with a versatile, weapon-agnostic set of effectors.

Across all these philosophies, the most critical architectural element is the Command-and-Control (C2) system. C2 is the junction point where sensor data is fused, threats are classified, and effectors are assigned. The sophistication of C2 — its real-time data processing capability, its use of AI for classification, and its interoperability capacity — is what truly defines a modern C-UAS. The global push toward interoperability standards like NATO's STANAG 4586 is further increasing the importance of C2, signaling a shift away from isolated systems and toward networked, multi-vendor defense grids.

Sources