A study on a distributed cooperated database structure and it's access security

(1)个人身份识别系统的特征提取方法我们研究了一种基于个人特征的个人身份识别系统，这些特征包括笔迹、声音、面部图像等，这些特征在我们的交流中经常使用。在该系统中，用户利用其个人特征进行身份验证，无需PIN码或智能卡，具有人性化的界面。从个体特征提取特征的方法是将特征参数分成几类，并根据对个体验证的重要程度对每一类进行适当的加权。当一个特征参数具有特征性和信息量时，会给出大量的权重值。我们还通过使用写入信息的测试结果研究了所提出系统的可靠性。(2)分布式协同数据库系统我们研究了一个自治协同数据库处理系统的系统架构、规范环境和用户界面；1)超分布式系统；提出了一种实现自主协调的环境。对高度信息化社会的影响也进行了研究。2)数据驱动的规范环境；研究如何提高软件的可重用性。3)多媒体接口；提出了在规范环境下建立更加人性化的界面。(3)分布式协同路由算法我们研究了分布式协同路由算法。我们使用随机学习自动机（SLA）来建立一个群体总活动的模型。

(1) Feature extraction method for personal identification systemWe have studied a personal identification system based on individual characteristics, such as handwriting, voice, facial image, which are used generally in our communication. In the proposed identification system, users utilize their individual characteristics for authentication process without PIN or smart cards, which provides human-friendly interface. As for feature extraction method from individual characteristics, feature parameters are classified into several categories and each category is weighted properly according to the importance for individual verification. In case that a feature parameters is characteristic and informative, a large amount of weighting value is given. We have also studied the reliability of the proposed system with test results using writing information.(2) Distributed cooperated database systemWe have studied system architecture, specification environment and user interface in an autonomous coordinated database processing system ;1) Hyper-Distributed System ; was proposed as an implementation of autonomous coordinated environment. Impacts on highly information society was also studied.2) Data-Driven Specification Environment ; was researched to improve software reusability.3) Multi-Media Interface ; was proposed to establish more user-friendly interface in the specification environment.(3) Distributed cooperated routing algorithmWe have studied distributed cooperated routing algorithm. We used a model of a colony's total activity using stochastic learning automaton (SLA).