| 1. | The previous researches in the estimation theory for smn are almost focused on the case of single sensor observation
以往针对带乘性噪声系统最优估计方法的研究大多集中在单传感器观测的情形下。 |
| 2. | Furthermore , in recent years two - dimensional ( 2 - d ) systems can find more and more applications in many fields such as image processing , but all the previous studies in smn considered one - dimensional ( 1 - d ) systems
另外，以往对带乘性噪声系统的研究考虑的均是一维系统，而二维系统在图像处理等许多领域具有重要的应用价值。 |
| 3. | Signal estimation theory for discrete stochastic systems with multiplicative noises ( smn ) is very important in many applications such as oil seismic exploration , underwater remote targets detection and speech signal processing
带乘性噪声系统的最优估计理论在石油地震勘探、水下目标探测、语音处理等诸多领域都有重要的应用价值。 |
| 4. | Consequently , it is necessary to study the estimation algorithms for smn under multi - sensor observation and 2 - d smn . on the basis of the two points , further studies of estimation algorithms for smn are made in this dissertation
基于这两点，本文针对多传感器观测下的带乘性噪声系统和二维带乘性噪声系统，对其最优估计方法进行了研究探讨。 |
| 5. | The a - fe phase in parent sm - fe alloys can not be inhibited by increasing the sm content . during nitrogenation process the sm - rich phase decomposed into a - fe phase and smn phase , deteriorating the magnetic properties
取得了如下研究结果：单纯提高sm - fe合金中的sm含量不能抑制初生的- fe相，而且富sm相会在氮化过程中发生分解，生成- fe相和smn相，恶化了磁体的磁性能。 |
| 6. | For smn under multi - sensor observation , two kinds of estimation fusion algorithms are proposed in this dissertation . one is conventional estimation fusion algorithm . the other is reversed - time filtering fusion and one - pass deconvolution fusion algorithm
根据融合策略的不同，本文提出了集中式与分布式两种滤波融合算法，其中分布式融合算法又分为带反馈与不带反馈两种实现形式。 |
| 7. | The main contents of this dissertation are as follows : 1 . in this dissertation , the development and status quo of signal estimation for smn is recalled . moreover , data fusion technology for multi - sensor systems and estimation theory of 2 - d systems are introduced simply
本文主要工作如下：第一，本文回顾了带乘性噪声系统的最优估计理论的发展和现状，并简单介绍了多传感器数据融合技术及二维系统理论。 |
| 8. | The first kind of algorithm is based on s - box controlled by key , the algorithm distinguishes most of other block ciphers , its visible quality is that its s - box is not regular and fixative , but controlled by block cipher ' s key . the second kind of algorithm is based on smn ( d ) array code , it makes full use of the visible quality of smn ( d ) array code possessing big code distance ( d ) . the thesis discusses the two algorithms " reversibility , security and their qualities , the
第一种是基于“ s -盒由密钥控制”的变结构分组密码，该算法区别于其它大部分分组密码最大的特点是它的s -盒不是固定不变的而是由密钥控制产生；第二种是基于smn ( d )阵列编码的变结构分组密码，该算法充分利用了smn ( d )阵列编码具有码距大的特点；文中对这两个算法的可逆性、安全性及其性能等进行了探讨和分析，分析结果认为这两个变结构算法正确、有效且安全。 |
