CN-121994082-A - Deep fusion architecture based on double-spiral interweaving directional space-sky interference equipment and gallium nitride technology

Abstract

The invention discloses a deep fusion architecture based on double-helix interweaving directional space-sky interference equipment and gallium nitride technology, and belongs to the technical field of low-altitude defense. The architecture comprises four modules of detection, identification, interception and cooperative control. The detection module adopts a gallium nitride T/R component and a carbon fiber lightweight antenna, and combines a self-adaptive beam forming algorithm to realize high-mobility and high-precision detection. The identification module fuses the micro Doppler characteristics and the multi-mode data, and high-accuracy target identification is achieved through an AI algorithm. The interception module integrates three means of radio frequency, electromagnetic pulse and laser, and is driven by an AI hierarchical decision algorithm to realize low-cost and high-efficiency damage. The cooperative control module realizes multi-node resource scheduling and cooperative combat through distributed networking and semi-autonomous AI decision-making. The invention solves the problems of poor maneuverability, low recognition rate, high interception cost and weak cooperative capability of the traditional low-altitude defense system, and is suitable for protecting unmanned aerial vehicle clusters and areas of low-altitude high-speed targets.

Inventors

• HE XIANGYU

Assignees

• 何祥宇
• 广东空天抗扰技术研究院有限公司

Dates

Publication Date

20260508

Application Date

20260120

Claims (5)

1. 1. The deep fusion architecture based on the double-helix interweaving directional space-sky interference equipment and the gallium nitride technology is characterized by comprising a detection module, an identification module, an interception module and a cooperative control module; The detection module adopts a solid T/R component based on gallium nitride (GaN) technology and a carbon fiber honeycomb composite material antenna, integrates a self-adaptive beam forming algorithm based on minimum mean square error (LMS), and is used for realizing high-power and light-weight low-altitude target detection; The identification module adopts a multimode fusion algorithm based on micro Doppler feature analysis and D-S evidence theory, and combines an improved deep neural network model to accurately identify and classify low-altitude, slow-speed and hypersonic targets; The interception module comprises a radio frequency interference unit, an electromagnetic pulse (EMP) transmitting unit and a laser weapon unit which can work cooperatively, and an AI hierarchical decision algorithm based on multi-factor weighting and reinforcement learning is built in the interception module, so that low-cost and high-efficiency hierarchical interception is implemented according to a target threat level and a resource state; The cooperative control module adopts a node networking technology based on a COFDM modulation and distributed aperture interconnection protocol, and operates a semi-autonomous AI decision unit for realizing data fusion, resource scheduling and cooperative operation among multiple nodes.
2. 2. The depth fusion architecture according to claim 1, wherein in the detection module, the power density of the gallium nitride solid-state T/R component is more than or equal to 3W/mm < 2 >, a micro-channel liquid cooling heat dissipation design is adopted, the antenna is made of a carbon fiber honeycomb composite material with the density of 1.2g/cm < 3 >, and the adaptive beamforming algorithm can dynamically form a null pointing to the clutter direction, and the rejection ratio is more than or equal to 40dB.
3. 3. The depth fusion architecture of claim 1, wherein the identification module is configured to extract a target time-frequency graph by short-time fourier transform (STFT) and classify the target time-frequency graph by a convolutional neural network, and wherein the multi-modal fusion algorithm dynamically assigns weights for radar, radio frequency and photoelectric sensor data.
4. 4. The depth fusion architecture of claim 1, wherein in the interception module, the AI hierarchical decision algorithm takes a target type, a speed, a distance, a cluster scale and a current interception resource state as inputs, generates and dynamically adjusts an interception strategy through decision tree model and Q-Learning reinforcement Learning optimization, the electromagnetic pulse transmitter works in an L band, adopts a Marx generator to generate fan-shaped electromagnetic pulses with the energy density of 10 6 J/m2, and the laser weapon adopts a solid laser with the wavelength of 1.06 μm, has the output power of 10kW, and is provided with a galvanometer scanning system to realize multi-target time-sharing irradiation.
5. 5. The deep fusion architecture of claim 1, wherein the cooperative control module supports cascading networking of more than or equal to 8 nodes, data synchronization is achieved among the nodes through a time stamp calibration algorithm based on GPS second pulses, synchronization precision is less than or equal to 1 μs, an open architecture is adopted, a hardware interface accords with an STD-1553B bus protocol, and a software framework adopts an ROS system.

Description

Deep fusion architecture based on double-spiral interweaving directional space-sky interference equipment and gallium nitride technology Technical Field The invention discloses a deep fusion architecture based on double-helix interweaving directional space-sky interference equipment and a gallium nitride technology, relates to the technical field of low-altitude target defense and electronic countermeasure, and in particular relates to a lightweight and distributed collaborative defense system architecture based on the gallium nitride technology and artificial intelligence, wherein the architecture integrates detection, identification, interception and control. Background With the wide application of unmanned aerial vehicles, cruise missiles and various low-altitude, slow-speed and small targets (such as small unmanned aerial vehicles), the traditional air defense system faces serious challenges. The prior art mainly has the following defects: the maneuverability is poor, the traditional low-altitude detection radar has large volume and heavy weight (usually more than or equal to 25 kg), the deployment time is long, and the requirement of a tactical with rapid maneuver is difficult to adapt. The recognition accuracy is low, under the complex electromagnetic environment and urban terrain, the traditional radar detection blind area is large, low and slow targets such as unmanned aerial vehicles, birds and the like are difficult to distinguish effectively, and the false alarm rate are high. The method has the advantages of high interception cost and low efficiency, and the traditional interception means (such as missiles) have high single cost and are difficult to deal with large-scale unmanned aerial vehicle cluster attack. Electronic interference equipment often has single function and lacks intelligent hierarchical collaborative interception capability. The cooperative capability is weak, the existing system is mostly of a chimney-type independent architecture, information among all nodes is not communicated, the sharing of detection resources and cooperative scheduling of interception weapons cannot be realized, and the overall defense efficiency is limited. Therefore, there is an urgent need for a novel defense system integrating advanced materials, intelligent algorithms and collaborative architecture to achieve efficient, low cost interception of low-altitude, clustered, intelligent threats. Disclosure of Invention 1. Technical problem The invention aims to solve the following technical problems of the existing space defense system: the detection module has low power density, large weight, complex deployment and weak clutter resistance; The recognition module has high false alarm rate, low hypersonic target recognition rate and poor electromagnetic interference resistance; the interception module has high cost, low resource utilization rate and insufficient multi-target interception capability; the system architecture is closed, the cooperative scheduling is poor, the labor cost is high, and the life cycle is short. 2. Technical proposal The invention aims to overcome the defects of the prior art and provide a deep fusion architecture based on double-helix interweaving directional space-sky interference equipment and gallium nitride technology, which realizes the full-flow optimization of detection, identification, interception and control through material innovation, algorithm innovation and system architecture innovation, and remarkably improves the defending capability on low-altitude and near-space targets. In order to achieve the above purpose, the invention adopts the following technical scheme: A depth fusion architecture based on double-helix interweaving directional space-sky interference equipment and gallium nitride technology comprises a detection module, an identification module, an interception module and a cooperative control module. The detection module adopts a gallium nitride (GaN) solid-state T/R component, the power density (more than or equal to 3W/mm <2 >) of the detection module is more than 3 times that of the traditional silicon-based or gallium arsenide component, and the detection module can stably work for a long time in a wide temperature environment of-40 ℃ to 70 ℃ by combining a micro-channel liquid cooling heat dissipation design. The antenna is made of carbon fiber honeycomb composite material, so that the overall weight of the radar is reduced to within 5 kg, and various maneuvering deployment modes such as knapsack, vehicle-mounted and airborne modes are supported. Meanwhile, the module integrates a self-adaptive beam forming algorithm based on minimum mean square error (LMS), so that multipath clutter generated by complex terrains (such as urban buildings) can be suppressed in real time (the suppression ratio is more than or equal to 40 dB), and high-precision detection is realized. The identification module adopts an intelligent identification mechanism of multi-mode