AI and machine learning take center stage in electronic warfare

Military electronic warfare (EW) stands on the precipice of a major revolution in capability, as enabling technologies like artificial intelligence (AI) and machine learning prepare to transform electronic combat on the airwaves.

AI and machine learning algorithms have the potential to provide capabilities crucial to EW, such as autonomous threat detection, adaptive countermeasures, and real-time decision-making, and EW systems designers are banking heavily on these technologies to bring about the next generation of EW, called cognitive EW.

AI and machine learning are transforming EW technology by enabling faster, smarter, and more adaptive systems now and in the future. These capabilities have the potential to analyze the radio frequency spectrum in real time; analyze enemy jamming and spoofing attempts automatically; and deploy countermeasures without human intervention.

Put together, AI and machine learning are expected to enable much faster responses to electronic threats. For example, drones equipped with AI targeting systems can identify and strike targets autonomously, even in heavily jammed environments.

AI also can automate fusion and interpretation of sensor data to improve situational awareness and help military commanders respond quickly to emerging threats.

What is electronic warfare?

EW describes using RF and microwave energy to control the RF spectrum, attack an enemy, or impede disrupt enemy operations, while ensuring friendly forces maintain unimpeded access to RF spectrum. It disables or disrupts enemy communications, radar, navigation, and other electronic systems.

EW has three major parts. Electronic attack involves jamming enemy radar and communications to degrade their effectiveness. Electronic protection involves protecting friendly forces from electronic attack with counter-jamming and RF emission control. Electronic support involves gathering intelligence by intercepting and analyzing enemy RF emissions for situational awareness and targeting.

EW can be conducted from air, sea, land, or space, using crewed or uncrewed systems, and involves techniques like radar jamming, deception, electronic masking, and reconnaissance.

Key capabilities include electromagnetic sensing and signal processing for detecting, identifying, and locating enemy emissions; high-performance embedded computing and open-system architectures for rapid processing and system modularity; network-enabled EW for integrating RF monitoring and cyber security; crewed and uncrewed to conduct real-time electronic combat and protection; and spectrum management and emission control to optimize and protect RF and microwave resources.

Among the most influential enabling technologies in EW operations are AI and machine learning algorithms that provide autonomous threat detection, adaptive countermeasures, and real-time decision-making to overcome the complexity and chaos of the electromagnetic environment.

The role of AI in electronic warfare

Artificial intelligence and machine learning are transforming EW capabilities by enabling fast, intelligent, and adaptive systems now and in the future. Today AI and machine learning can help analyze the RF spectrum in real time to detect enemy jamming and spoofing attempts automatically and deploy countermeasures without human intervention.

Uncrewed aircraft with AI-based targeting systems, for example, can identify and strike targets autonomously, even amid heavy enemy jamming.AI also automates fusing and interpreting sensor data for improved situational awareness and fast command decision-making.

For the future, AI-enabled autonomy is expanding into operations on and under the oceans, in space, and in missile defense for battlefield reconnaissance, targeting, and EW based on real-time data.

The next generation of EW likely will involve coordinated control of uncrewed systems with AI to enable systems to adapt to battlefield conditions dynamically. AI also is expected to improve EW by generating tailored signals to mislead or disrupt adversaries, rather than simply defending against attacks.

Integrating EW enabling technologies with emerging technologies like quantum computing and 5G networks is expected to enhance data processing capabilities drastically, and support predictive analytics and securing communication channels that are essential for EW operations.

Future capabilities in cognitive EW

Cognitive electronic warfare is expected to employ AI and machine learning to enable electronic systems to perceive, analyze, learn, and adapt autonomously in real time within the electromagnetic spectrum.

Unlike traditional EW, which relies heavily on human analysis and pre-programmed responses, cognitive EW uses AI-driven cognitive processes to recognize, classify, and respond rapidly to complex and evolving electronic signals such as RF emitters, pulsed radars, and low-probability-of-detection signals.

Cognitive EW will provide rapid situational awareness of the electromagnetic spectrum, adaptive countermeasures, signal fingerprinting, and real-time intelligence without human intervention.

It will operate at millisecond or microsecond speeds to outpace human decision-making, and enable military forces to outmaneuver and deceive enemy sensors and communications, and optimize resource usage.

Cognitive EW will work using sophisticated machine-learning algorithms deployed close to the sensors that perform rapid signal recognition, and integrate components such as sensing and data-collection units detecting signals; processing signals, and converting analog signals into digital.

AI and machine learning engines will analyze patterns, classify signals, predict threats, and learn from new encounters. Decision engines will be able to choose optimal electronic attack or protection responses, and carry out electronic jamming, spoofing, or deception. These systems will be able to adapt over time and improve their effectiveness dynamically to enhance situational awareness.

Electronic jamming systems will be aware of EW threats and adapt their jamming behavior dynamically by synthesizing the most appropriate jamming program in real time.

Cognitive EW are expected to produce systems that adapt quickly to changing enemy tactics by rapidly sensing, identifying, and countering new or unexpected electronic threats, such as neutralizing swarms of uncrewed aircraft with wide-beam jamming that adapts to the threat moment by moment.

Initial cognitive EW efforts

While the term cognitive EW is somewhat new, foundational research has been in progress for at least the past 15 years. A 2010 project of the U.S. Defense Advanced Research Projects Agency (DARPA) in Arlington, Va., called Behavioral Learning for Adaptive Electronic Warfare (BLADE) sought to push the bounds of machine learning to counter enemy threats from wireless adaptive communications such as battlefield radios, command and control networks, and RF triggers like cell phones used to detonate improvised explosive devices (IEDs).

The project pursued adaptive communications that automatically adjusts to conditions that degrade its performance, such as environmental conditions, or from intentional or inadvertent EW signals jamming.

BLADE sought to develop a networked electronic attack system that jams new wireless communications threats automatically by detecting and characterizing the new threat, learning to jam the new threat effectively and efficiently, and assessing the effectiveness of RF jamming in the field.

The goal was the ability to operate as one node or as a network of distributed BLADE nodes, with performance improving as nodes are added to the network. Companies interested in participating should use existing networking capabilities to enable information sharing among several BLADE nodes. Lockheed Martin Corp. handled the original BLADE research.

Two years later, DARPA kicked-off a project called Adaptive Radar Countermeasures (ARC) to find ways to detect and counter digitally programmable radar systems that have unknown behaviors and agile waveform characteristics.

The program involved the BAE Systems Electronic Systems segment in Nashua, N.H., which worked on developing cognitive electronic warfare technologies, adaptive algorithms, and testing for airborne electronic warfare systems. Leidos handled developing electronic warfare adaptive radar countermeasures, while Exelis (now L3Harris) worked with Leidos to implement Leidos-developed software processing to protect airborne platforms with electronic warfare (EW) systems.

Today's cognitive EW

There are a few implementations of cognitive electronic EW today, which integrate AI and machine learning to detect, analyze, and respond to threats autonomously more quickly than traditional methods.

The U.S. Air Force awarded a $6.4 million contract in 2024 to Southwest Research Institute (SwRI) in San Antonio, Texas, to develop cognitive EW algorithms to analyze the electronic environment similarly to how a human pilot interprets signals and threats, but with greater speed and accuracy. The project involves AI and machine learning to extract features from threat radar signals and identify new, previously unknown signals that are not in traditional threat libraries.

The approach includes two phases: feature extraction using AI and machine learning to identify signal characteristics, and grouping millions of radar pulses to highlight vulnerabilities and lethal signals.

Neuromorphic processing hardware emulates brain-like memory and processing to boost efficiency and speed beyond the capabilities of conventional systems to transform EW from a reactive to a proactive and adaptive capability.

Northrop Grumman also is advancing algorithms using machine learning to detect hard-to-detect threats like low-power radio frequency signals near noise floors, enhancing threat detection in GPS and navigation warfare contexts. While some cognitive EW tools have been deployed for testing and evaluation, full integration and operational use has not happened yet.

Big EW programs

Although AI, machine learning, and cognitive technologies play heavy roles in EW today, traditional EW design approaches are still major players, and are at the forefront of many large U.S. military EW programs. Among the largest and most influential EW) programs in the U.S. Department of Defense are several key initiatives led mainly by the U.S. Army and focused on rapid-response EW capabilities.

One example is the U.S. Army Multifunction Electronic Warfare – Air Large (MFEW-AL) program, which focuses on airborne electronic attack capabilities and capitalizes on commercial off-the-shelf (COTS) technology for rapid capability delivery.

The Lockheed Martin Corp. MFEW-AL airborne EW system seeks to detect, identify, locate, deny, disrupt, and degrade enemy communications and radar systems. It's a self-contained airborne EW pod for the MQ-1C Gray Eagle uncrewed aircraft, and provides battlefield commanders with electronic attack capabilities. It uses modular, open-system architecture based on the C5ISR/EW Modular Open Suite of Standards (CMOSS), and seeks to fill a gap in organic electronic attack capabilities for Army combat aviation brigades. No clear deployment date is set, however, because of a change in approach and strategy.

Lockheed Martin also is involved in the Army Terrestrial Layer System (TLS) for Brigade Combat Teams (BCTs) program, which focuses on providing extended-range SIGINT and cyber warfare at the brigade level. TLS BCT has variants tailored for different BCT types: Stryker Brigade Combat Teams (SBCT), Armored BCTs (ABCT), and Infantry BCTs (IBCT). The SBCT variant integrates onto the Stryker Medical Evacuation Double-V Hull vehicle, the ABCT variant onto an Armored Multi-Purpose Vehicle, and the IBCT variant is a man-portable system called TLS BCT Manpack.

The TLS BCT Manpack handles RF surveying, signals collection, direction finding, and electronic attack, and has been fielded on a limited basis since 2024. A broader assessment for all BCTs is planned for June 2026, with full deployment expected by 2028. The Army separates SIGINT and EW capabilities for the Stryker variant. Prime contractors are Mastodon Design LLC, a subsidiary of CACI International, which is building the TLS-BCT Manpack system. Lockheed Martin is working on EW SIGINT, and cyber capabilities, and is supporting the TLS Brigades and Echelons Above Brigade (EAB) variants.

The Army's Vehicle Mounted Multi-mission Electronic Warfare System (VMEWS) is designed to protect military vehicles from RF threats like jamming, deception, and electronic attacks aimed at disrupting vehicle systems, sensors, and communications. The system helps to identify, respond to, and counter these RF threats to ensure vehicle survivability.

The system uses rapid reprogramming through AI and advanced analytics to react quickly. VMEWS, from Pacific Defense in El Segundo, Calif., encompasses electromagnetic warfare capabilities to protect wheeled and tracked vehicles. Pacific Defense partners on VMEWS include Thales Defense & Security; BAE Systems; Palantir; MAXISIQ; Regal Technology Partners; and STC, an Arcfield company.

The Navy Surface Electronic Warfare Improvement Program (SEWIP) Block 3 seeks to surface ship defenses against anti-ship missiles through early threat detection and electronic countermeasures. It is an upgrade to the legacy AN/SLQ-32 electronic warfare system, and protects surface ships from anti-ship missiles by providing early detection, signals analysis, threat warning, and soft-kill defense. The SEWIP Block 3 prime contractor is Northrop Grumman Corp.

The Digital Radio Frequency Battlespace Emulator (DRBE) from Cerebras Systems in Sunnyvale, Calif., is sponsored by the U.S. Defense Advanced Research Projects Agency (DARPA) in Arlington, Va. It seeks to develop a large-scale, virtual RF environment to help develop, provide training, and test for advanced RF systems like radar and EW systems. It enables several RF systems to interact to replicate dense responsive real-world RF conditions.

The core of DRBE is real-time high-performance wafer-scale computing that delivers massive computational throughput with microsecond latency. DARPA is enhancing DRBE with optical interconnects and scalable wafer-scale computers to increase bandwidth to enable larger and more complex RF scenarios and expand into battlespace autonomy and digital twins.

Among the most influential EW companies in the world are Lockheed Martin Corp.; Northrop Grumman Corp.; RTX Raytheon; L3Harris Technologies; the Boeing Co.; General Dynamics Corp.; CACI International; BAE Systems; Thales Group; Saab AB; Leonardo SpA; Elbit Systems; and HENSOLDT.

Key enabling technologies

EW systems rely on several enabling technologies to enhance RF and microwave detection, interception, disruption, and protection across the electromagnetic spectrum. They involve hardware, software, and AI-driven elements.

Among the most important EW enabling technologies are high-power RF amplifiers; digital signal processing hardware and software; frequency management and frequency hopping; stealth technologies and signals hardening; cognitive radio and cognitive processing; electromagnetic deception and jamming; and distributed mesh networks.

High-power RF amplifiers and antennas boost and transmitting signals, and expand operating ranges. Signal processors and software enable real-time analysis of intercepted signals to support rapid decision-making and adaptive responses to threats. Frequency management and frequency hopping help mitigate jamming and interference by varying signal frequencies dynamically. Stealth technologies and signal hardening can involve radar-absorbent materials and cryptographic techniques that protect friendly signals from enemy detection and disruption. Cognitive radio and processing involve adaptive systems that learn and respond autonomously to enemy RF signals. improving resilience and countermeasure sophistication. Electromagnetic deception and jamming involve signal jamming and spoofing to disrupt enemy sensors and communications. Thermal signature management and radar-absorbent coatings: advanced materials science enabling stealth in harsh environments such as space. Distributed mesh networks reduce vulnerability to centralized command disruption.

The role of embedded computing

Among the most crucial enabling technologies for EW -- besides AI and machine learning -- is embedded computing. These technologies are responsible for real-time processing, machine autonomy, and the adaptability necessary for electronic attack, defense, and surveillance, and plays a central role in jamming, spoofing, signals detection, and countermeasures.

Embedded systems offer fast computation necessary to detect and respond to threats in milliseconds in missile defense, radar signal processing, and applying electronic countermeasures quickly. Certain kinds of embedded computing also are designed to be rugged and function reliably under extremes in temperature, vibration, radiation, and electromagnetic interference.

Sophisticated algorithms run on embedded systems enable system agility and stealth, and enable EW systems to process data locally at the edge to reduce latency and bandwidth loads, and can enable autonomous operation even when communications links are jammed. Embedded computing in EW systems also integrates hardware-level encryption and anti-tamper mechanisms to secure sensitive military communications and data against cyber threats and espionage.

Embedded computing refers to specialized computing systems designed to perform dedicated functions within larger systems, often operating in real-time with low power, small size, and optimized reliability. These are typically compact computers integrated into devices to carry out specific tasks without user interaction.

Yet there's another branch of embedded computing called high-performance embedded computing (HPEC) that plays an equally important role in today's and tomorrow's EW systems. HPEC extends embedded computing by delivering significantly higher computational power, comparable to data center-level or supercomputing performance, within rugged, compact, and power-constrained applications.

HPEC is designed for mission-critical applications that need real-time processing for vast amounts of data from sensors like radar, video, and SIGINT. They combine processing elements like central processing units (CPUs), general-purpose graphics processing units (GPGPUs), and field-programmable gate arrays (FPGAs) through open-systems high-throughput interconnects, to enable edge computing with advanced workloads like AI inference and sensor fusion, and maintain a balance between performance, size, weight, and power consumption.

HPEC provides real-time processing, high-speed data analysis, and the ability to perform complex electronic countermeasures like jamming, spoofing, and radar deception. They process large volumes of data from radar, infrared sensors, and signals intelligence to help detect and disrupt enemy communications and surveillance. HPEC platforms bring data center-level computing power into rugged, compact systems suitable for harsh battlefield environments.

HPEC-enabled real-time processing is essential for intercepting and responding to electromagnetic signals within milliseconds for electronic support, attack, and defense. It can enable jamming and deception via digital radio frequency memory (DRFM) jammers, which rely on embedded digital signal processors. Increasingly, HPEC integrates AI and machine learning to enhance signals recognition, threat identification, and decision-making in EW.

Combining GPGPUs, FPGAs, and high-bandwidth I/O supports the heavy computational and communication demands typical of EW systems. This approach helps provides rugged, compact computing with supercomputing performance for deployment in rugged and space-constrained applications. It also can help integrate several EW functions like surveillance, jamming, and electronic support in one system.

Embedded computing leaders

Among the most influential high-performance embedded computing companies involved in EW are Intel Corp.; Advanced Micro Devices Inc. (AMD); Curtiss-Wright Corp. Defense Solutions; NVIDIA Corp.; Abaco Systems; Aitech Systems; NXP Semiconductors; Mercury Systems; Ecrin Systems; Toyon Research Corp.; Pacific Defense; General Micro Systems Inc. (GMS); EIZO Rugged Solutions; Extreme Engineering Solutions (X-ES); Systel; LCR Embedded Systems; Elma Electronic; General Dynamics Corp.; and BAE Systems.

Intel provides CPU and FPGA processors designed for EW; NVIDIA provides rugged GPGPUs.; Abaco, Curtiss-Wright, Aitech, Mercury, GMS, EIZO, X-ES; and LCR provide rugged high-performance embedded computing modules; Elma provides rugged embedded computing chassis and enclosures for EW applications.