| 摘要: |
| 针对当前智能抗干扰技术策略维度低、复杂环境应对性差的问题，提出了适应复杂环境的多维策略智能抗干扰技术。将单音信号作为探测信号，通过扫频方式减少探测阶段复杂度以实现快速实时的环境状态特征提取，之后设计了基于深度神经网络的实时智能决策引擎模型以提高决策速度和准确率。仿真结果表明所提方案能够准确地预测通信质量,最后根据目标函数在所有可通信策略中决策出最优策略，当探测信号扫频间隔选取合适时，该方案能够达到接近96%的决策准确率及较好的资源利用率，能有效进行抗干扰并取得较好的通信质量。 |
| 关键词: 无线通信 复杂信道环境 智能抗干扰 智能决策 深度神经网络（DNN） |
| DOI：10.20079/j.issn.1001-893x.211111003 |
|
| 基金项目:国家自然科学基金资助项目（U2030204）；中国工程物理研究院院长基金资助项目 |
|
| Multi-dimensional strategy oriented intelligent anti-jamming technology in complex channel environment |
| WANG Zhenyu,LIU Youjiang,XU Huiyuan,CAO Tao,YANG Dalong,LI Simin |
| (Institute of Electronic Engineering，China Academy of Engineering Physics，Mianyang 621900,China) |
| Abstract: |
| To solve the problem of low dimensionality and poor adaptability to complex environments of current intelligent anti-jamming technology strategy,a multi-dimensional strategy oriented intelligent anti-jamming technology is proposed.The single tone signal is used as the detection signal by frequency hopping to reduce the complexity in getting the feature extraction of complex environment status quickly and timely.Besides,a decision engine model based on deep neural network(DNN) is designed to improve the speed and accuracy of intelligent decision-making.The simulation results show that the proposed scheme can predict the communication quality accurately.Finally,the optimal strategy is determined among all communication strategies based on the objective function.When the detection signal sweep interval is selected appropriately,the proposed method can achieve close to 96% decision accuracy and good resource utilization,indicating that it has effective anti-jamming capability and achieves good communication quality. |
| Key words: wireless communication complex channel environment intelligent anti-jamming intelligent decision-making deep neural network(DNN) |
