ID	原文	译文
2663	通过莱维飞行代替了传统鲸鱼优化算法(Whale OptimizationAlgorithm，WOA)参数的随机选择，优化了鲸鱼算法中跳出局部最优的能力;	When the random selection of the parameters of the traditional whale optimization algo-rithm (WOA)is replaced with the Levy flight algorithm, the ability to jump out of the local optimum is optimized.
2664	借助改变鲸鱼算法的系数向量收敛方式明显提高了鲸鱼优化算法的泛化能力、预测精度和收敛速度。	Chan-ging the method of coefficient vector convergence results in improvements to the generalization ability, prediction precisionand convergence speed of the WOA.
2665	数据仿真结果表明，所提出的 LWOA-LSSVM 预测模型，不仅能够克服局部寻优获取全局最优解，而且具有快速的收敛速度和更高的预测精度，	Data simulation results show that the proposed LWOA-LSSVM forecasting model notonly overcomes the local optimization to obtain the global optimal solution, but also achieves faster convergence speed and higher prediction accuracy.
2666	得出预测结果的均方根误差、平均绝对误差和平均绝对百分比误差与遗传算法 BP 神经网络、遗传算法最小二乘支持向量机和传统鲸鱼算法最小二乘支持向量机相比均有着明显提高。	Prediction results of the model, concerning root mean square error, mean absolute error, and mean absolute percentage error, show noticeable improvements when compared to those of the genetic algorithm and back propaga-tion (BP)neural network, the genetic algorithm and LSSVM, and the traditional WOA and LSSVM.
2667	同时，通过调整目标命中率和训练输入样本量验证了预测模型具有更好的鲁棒性。	At the same time, through adjustments of the target hit ratio and the number of training sample entries, the prediction model is proven to bemore robust than the aforementioned algorithms.
2668	在已有的卷积码同步加扰的扰码反馈多项式重构方法中，卷积码对偶码字需要先验已知。	The existing methods need to know a priori dual word of a convolutional code to reconstruct the feedback polynomial of a synchronous scrambler placed after a convolutional encoder.
2669	为了解决该问题，本文基于 m 序列的三阶相关性，提出一种卷积码加扰的扰码反馈多项式重构新方法。	To overcome this limitation, a novel reconstruc-tion method is proposed based on the triple correlation property of m-sequences.
2670	首先对卷积码加扰序列进行分块处理，数据块的长度为卷积码编码约束长度，相邻数据块的起始位间隔一个码长;	First, the scrambled bit sequence is dividedinto multiple blocks. The length of each block equals the constraint length of the convolutional encoder, and the interval be-tween two start points of adjacent blocks is the codeword length.
2671	然后证明了加扰数据块与对偶码字相乘后，输出序列为与扰码周期相同的 m 序列，由此基于 m 序列的三阶相关峰值特性估计出对偶码字;	Then, it is proved that the generated sequence by the dot product of scrambled bit blocks with a dual word is also an m-sequence, having the same period as the synchronous scram-bler. With this result, a dual word of the convolutional encoder can be estimated based on the triple correlation property of m-sequences.
2672	最后利用三阶相关峰的位置信息重构出同步扰码反馈多项式。	Finally, the feedback polynomial is reconstructed by using two locations of triple correlation peaks.