主要研究了一种基于强化学习的无线通信智能干扰方法.目前,传统的干扰大多为单一域干扰.而伴随着越来越复杂的电磁频谱环境和抗干扰技术的不断发展,单一域内干扰难以在复杂环境中起到较好的干扰效果.同时,通信用户的侦察系统会接收干扰机能量值对其进行侦察定位,随即利用相关抗干扰技术减弱干扰.为提升复杂电磁环境中的干扰效果并降低干扰机被定位概率,结合强化学习提出了一种功率与信道联合干扰决策算法.干扰机可以通过学习和训练在动态变化环境中决策出最佳干扰功率和干扰信道.仿真结果表明,在所给条件下,所提算法可以收敛到最佳干扰策略,相较于随机干扰算法和不考虑定位因素的Q学习干扰算法,所提算法在确保干扰有效性的同时,干扰机被定位的概率分别降低了30％和60％.

Aiming at its deficiencies in digital communication, the traditional CORDIC algorithm applica-tion is optimized from the angle of computational scale and complexity. Firstly, in accordance with the symmetry of trigonometric function, the input vectors are preprocessed, the input angle is limited within [0,π/2), and the output data recovered to the corresponding angle, thus realizing full coverage of the al-gorithm. Secondly, the rotation angle is transformed into the appropriate minimum angle by judging the re-lationship of between the input vector x and y. The simulation results indicate that this algorithm could re-duce the iteration times in the same precision, while significantly improving the convergence speed.

Turbo码的编码增益高,系统工作信噪比低,低信噪比条件下载波同步是Turbo码获得优异性能的重要保障。本文对现有研究中关于低信噪比条件下Turbo载波技术进行了分类,简要介绍了几种比较典型的Turbo载波同步算。

In this paper, the influence of imperfect carrier phase synchronization on the performance of turbo-coded system is analyzed and a novel method called Logarithm-A-Posteriori Probability-Ratios (LAPPRs)-statistic-decision-aided for carrier phase synchronization and decoding is proposed. The bits satisfying the criterion presented in this paper are chosen iteratively for Data-Aided estimation and phase estimation performance influenced by a few wrongly-decided bits are also mentioned. The bit error rate is investigated and shows the performance of this method is very close to the optimally synchronized system.

目前宽带卫星通信系统广泛采用了多点波束及星上交换技术,但大都仅提供固定业务或便携式业务,如何在这种体制下实现"动中通"业务相关的研究还比较少.本文主要介绍了Spaceway3系统针对"动中通"应用的设计理念,可作为以后其他系统设计的参考.

By making use of the characteristic of the cost function MSSO (Mean of the Square Soft Outputs), a new method of carrier phase synchronization and decoding for the turbo-coded systems is presented. Firstly, a short training sequence is used to achieve the initial phase estimation and to narrow the phase searching range. Secondly, three decoding sequences with the largest MSSO values are chosen by means of parallel searching strategy. Furthermore, the three selected decoding sequences are finally processed according to the majority decision criterion. The proposed method does not have to iteratively estimate and correct the carrier phase to get the desired synchronization. The calculation is greatly reduced though the transmission efficiency will be sacrificed a little in the presence of the training sequence.