AI-powered radar cannot be jammed, China’s landmark test flight result suggests

China has successfully flight-tested what could be the world’s first artificial intelligence-powered radar system for a military aircraft. According to the limited data disclosed, an AI-enhanced radar system aboard an unidentified aircraft achieved near-perfect target tracking performance despite facing advanced, sophisticated jamming. In the test, when conventional radar systems failed to maintain consistent contact, losing the target in around one-quarter of engagement time, the AI boosted the detection rate to near perfection. “Radar target tracking continuity has improved from the original 70 per cent-80 per cent to over 99 per cent,” wrote project lead scientist Zhang Jie with the China Electronics Technology Group Corporation’s 14th Research Institute in a peer-reviewed paper published in the Chinese journal Informatisation Research last month. “A paradigm shift in radar design philosophy is on the horizon.” The Nanjing-based institute is China’s premier radar development hub and the cradle of its military radar industry. The advance addresses a long-standing challenge for China’s aerial forces. While the PLA Navy has made strides in deploying AI-assisted radar on surface vessels – systems that helped maintain tracking during aggressive electromagnetic suppression by the US military – the constraints of airborne platforms have slowed progress. Space, power and processing limitations in fighter jets made on-board AI integration far more difficult than in larger, ship-based systems. But this trial suggests those barriers have now been overcome. A big factor in modern warfare is what military theorists call “electromagnetic fog” – a dense, ever-shifting barrage of signals, jamming pulses, stealth platforms and decoys that makes traditional radars increasingly obsolete. In these conditions, a radar cannot be designed solely for target detection. The battlefield demands perception – a holistic awareness of the entire war environment. “Existing radar anti-jamming techniques are mainly designed based on static assumptions about the electromagnetic environment, making them often unable to cope with dynamically changing interference conditions. This shortcoming can lead to a significant degradation or even complete failure of radar performance,” Zhang wrote. “Our novel environment-aware anti-jamming technique for airborne radar aims to substantially enhance the radar’s interference suppression capability in real-time, dynamic electromagnetic environments.” Rather than merely scanning and reporting returns, the smart radar continuously analyses the full electromagnetic spectrum and identifies interference patterns. When jamming is detected, it instantly adjusts frequency, beam direction and waveform to evade suppression – much like water flowing around a stone. “By sensing electromagnetic interference in the surrounding environment, the technique adaptively selects operating frequencies and bandwidths, and integrates space-time two-dimensional adaptive processing to maximally avoid interference, thereby greatly improving the radar’s resilience against jamming,” he added. If the enemy changes tactics, the radar adapts within milliseconds. The Chinese engineers have opted for traditional machine learning algorithms over the large language models (LLMs) trending in AI development. This suggests a deliberate, safety-conscious approach to integrating AI into crewed combat aircraft. While China has deployed LLM-based AI in electronic warfare drones capable of autonomous signal analysis and jamming, the choice of a more interpretable, deterministic AI framework for manned military aircraft reflects a concern for reliability and control. All test results, including algorithm design, simulations and actual flight data, were rigorously validated, according to the study. A 2021 US congressional report warned that America was losing its long-held edge in electronic warfare. More recently, the Pakistani military claimed it shot down five Indian fighter jets – including three state-of-the-art Dassault Rafales – during border clashes in May, crediting Chinese-supplied weapons and electronic warfare support. Defence experts say such events point to an emerging trend: China may be achieving a “one-way transparency” advantage in the electromagnetic spectrum, where it can see and strike adversaries while remaining largely invisible. “If this AI radar performs as described in combat, it could lead to electronic supremacy,” said a Beijing-based aerospace expert, who requested not to be named due to the sensitivity of the issue. Beyond warfare, the technology holds transformative potential for civilian life. China is building smart cities but the electromagnetic spectrum is getting crowded. Too many devices cause interference. Autonomous vehicles and drones, for instance, rely on radar and sensors – but they can malfunction if there is strong interference nearby. The new technology helps civilian radar systems by automatically detecting and avoiding interference, making radar more reliable and safer in congested areas and reducing false alarms and downtime, according to Zhang. “This technology not only holds broad application prospects in the military domain, but also provides solutions for civilian applications in addressing issues such as spectrum interference,” he wrote.

AI-powered radar cannot be jammed, China’s landmark test flight result suggests

.css-1c6uqr6{color:inherit;font-weight:inherit;font-size:inherit;font-family:inherit;line-height:inherit;overflow-wrap:break-word;}Smart radar capable of adjusting frequency, waveform and beam direction to evade detection from enemy forces

Smart radar capable of adjusting frequency, waveform and beam direction to evade detection from enemy forces