ID	原文	译文
3013	最后，采用投票决策和置信度决策融合策略，提升多天线接收端协作识别精度。	Later, two decision fusion strategies ofvoting-based and confidence-based were leveraged in the multiple-antenna receiver to improve recognition accuracy.
3014	实验结果表明，所提算法能有效识别{BPSK,4PSK,8PSK,16QAM,4PAM}5 种调制方式，当信噪比大于或等于−2 dB 时，识别精度可达 100%。	Experimental results show that the proposed algorithm can effectively recognize five modulation types {BPSK,4PSK,8PSK,16QAM,4PAM}, with a 100% recognition accuracy when the signal-to-noise is equal or greater than −2 dB.
3015	为了解决僵尸网络隐蔽性强、难以识别等问题，提高僵尸网络检测精度，提出了基于生成对抗网络的僵尸网络检测方法。	In order to solve the problems of botnets' strong concealment and difficulty in identification, and improve the detection accuracy of botnets, a botnet detection method based on generative adversarial networks was proposed.
3016	首先，通过将僵尸网络流量中的数据包重组为流，分别提取时间维度的流量统计特征和空间维度的流量图像特征；	By re-organizing the data packets in the botnet traffic into streams, the traffic statistics characteristics in the time dimension andthe traffic image characteristics in the space dimension were extracted respectively.
3017	然后，基于生成对抗网络的僵尸网络流量特征生成算法，在 2 个维度生产僵尸网络特征样本；	Then with the botnet traffic feature generation algorithm based on generative adversarial network, botnet feature samples were produced in the two dimen-sions.
3018	最后，结合深度学习在僵尸网络检测场景下的应用，提出了基于 DCGAN 的僵尸网络检测模型和基于 BiLSTM-GAN 的僵尸网络检测模型。	Finally combined with the application of deep learning in botnet detection scenarios, a botnet detection model based on DCGAN and a botnet detection model based on BiLSTM-GAN were proposed.
3019	实验表明，所提模型提高了僵尸网络检测能力和泛化能力。	Experiments show that the pro-posed model improves the botnet detection ability and generalization ability.
3020	考虑点源、一对吸收机和透明机共存场景，在透明机接收分子概率模型基础上引入干扰因子，考虑空间中吸收机对该透明机接收分子的影响，建立吸收机干扰下点源−透明机信道模型，	A coexistence scenario with a point source, a pair of absorbing and transparent receiver was considered, an in-terference factor was introduced in the proposed channel model based on the receiving molecular probability in thetransparent receiver considering the influences of the absorbing receiver on the transparent one.
3021	并结合人工神经网络使用Levenberg-Marquardt 算法对信道模型参数进行学习和预测。	Furthermore, the channelmodel of point source and transparent receiver had been proposed by using Levenberg-Marquardt algorithm combined with artificial neural network to study and predict channel model parameters.
3022	仿真结果不仅验证了所提信道模型的有效性，还表明空间中任意点峰值时刻与点源到该点的距离平方成正比，与分子扩散系数成反比，峰值时刻不受空间中的吸收机影响。	The simulation results not only verify the effectiveness of the proposed channel model, but also show that the peak time of any point in the environment is directly proportional to the square of the distance from the point source to the receiver, and inversely proportional to the molecu-lar diffusion coefficient, and the peak time is not affected by the absorbing receiver in the environment.