Abstract:This paper is to suggest that traditional 2-valued Boolean algebra is not sufficient for representation of VLSI circuits at logic gate level, although it is the case for combinational circuits. Instead of using the Register-and-Transfer technique at RT level to represent sequential circuits, an alternative is sought and found. That is, all uncertain voltages such as oscillations and floating voltages are identified by the same value, denoted as ⊥ (called bottom). The two certain voltages are, as usual,ground and power; they are denoted by 0 and 1 respectively. An invertor, a nor-gate and a nand-gate are defined according to the physics of VLSI circuits instead of the Boolean algebra. As a result, a logic is obtained which coincides with Kleene 3-valued logic provided that Kleene's u=⊥. As it is well known, Kleene 3-valued logic is functionally incomplete. This means that not every function (or gate) can be constructed from invertors, nor-gates and nand-gates. However, by introducing a partial order  into the logic, by using general fixed-point operators instead of the least fixpoint operator to deal with feedbacks in VLSI circuits, and by applying CPO (complete partial order) domain theory to the derived 3-valued logic system, the result obtained means that this system is functionally monotonic complete. Also, the canonical normal forms for this Kleene 3-valued logic are obtained. Although the present results are mainly semantic, it is very interesting in pursuing the research further by investigating the syntactical derivability. Such research would derive more secure circuits close to reality, which is to make the work compatible with VHDL, Verilog HDL and/or EDIF. Incidentally, Mukaidono has obtained similar results, although his approach is not as coherent as the one presented in this paper.

Abstract:Ajtai and Dwork have introduced a probabilistic public-key encryption scheme which is secure under the assumption that a certain computational problem on lattices is hard on the worst-case. In this paper, the author demonstrates how Ajtai-Dwork cryptosystem can be abused. Using this kind of abuses, users can communicate secrets in a key escrowed Ajtai-Dwork cryptosystem without fearing that their secrets will be revealed later by reconstructing their escrowed private-keys. However, it is also shown that users have to trust their implementers because unscrupulous implementers of Ajtai-Dwork cryptosystem may leak their private-keys without their awareness. The author shows how one can make Ajtai-Dwork cryptosystem abuse-free.

Abstract:Liveness and safeness are important behavioral properties of nets (systems). Many powerful results have been derived for some subclasses of Place/Transition nets (systems). The aim of this contribution is to draw a general perspective of the liveness and safeness for Asymmetric Choice nets (AC nets). Firstly, this paper presents a sufficient and necessary condition for those AC nets which have liveness monotonicity and a polynomial-time algorithm to decide if a given AC system is live and safe, and it satisfies liveness monotonicity. And then the sufficient and necessary conditions of (structural) liveness and (structural) safeness for two subclasses of AC nets (Strong I AC nets, Strong II AC nets) which have liveness monotonicity are presented.

Abstract:Protocol is the lifeline of computer network. The rapid increasing of protocol complexity results in a discipline of protocol engineering. Based on an analysis of the contents and methods of protocol engineering activities and their interrelations, first discusses main formal description techniques (FDTs) and their characteristics, compares corresponding strongpoints and weaknesses, and then leads to Petri nets based FDT. Secondly, the paper points out special advantages of the Petri nets based formal techniques and the current research difficulties in Petri nets based protocol engineering, among which protocol development oriented net tools are now very important research tasks. Thirdly, the paper summarizes international research advances in terms of OSI/RM layers and expounds research trends in this area. Finally, the authors give fundamental methodologies for Petri nets based protocol engineering in protocol specification, verification and analysis, and computer-aided testing and implementation.

Abstract:In this paper, a new type Yaksha security system is presented based on ELGAMAL(NOT RSA) algorithm. The system is capable of reusing a single security infrastructure to perform various security functions——cryptography digital signatures, distributed authentication and key exchange. At the same time, how the system can be used for key escrow is also described, one of the discussions which attract public attention.

Abstract:In this paper, the authors describe the precedures of DCT (discrete cosine transform) and FFT (fast Fourier transform) which map integers to integers by using lifting scheme and the butterfly configuration of FFT. The transform is reversible, fast and suitable for the lossless image compressions.

Abstract:In this paper, from application viewpoint, a model for enhancing parallel computing performance is established. The performance of parallel computing is analyzed when different domain divisions affect convergence speed in solving the whole problem. In addition, the major factors affecting the performance of massive data parallel computing are analyzed, such as overlapping operation,the problem data size and selective algorithm. Finally some computational examples provide evidence for the usefulness of the model.

Abstract:Focusing on software productivity and software quality demanded by the development of software industry has spurred the research on software reuse technology. Therefore, the metrics about software reuse technology are also attracting more attention. The goal of Jade Bird Component Library (JBCL) system is to describe, manage, store and retrieve components in order to fulfil the requirement of component-reuse-based software development process. This article introduces the component metrics model in JBCL and discusses its implementation based on the Object-Oriented Metrics Tool of Jade Bird Program Analysis System (JBPAS) and JBCL Posteriori Metrics System, which is made up of Feedback Collection, Analysis and Processing Tools.

Abstract:In this paper, the authors first analyze the common shortcoming in the existing firewall systems. Based on these, they introduce a new concept-Agent. According to the definition of Agent, they define some Agent-based components of a firewall, and then describe the communication and collaboration of the Agents. Finally, the course of designing and deploying an Agent-based firewall is presented.

Abstract:Flow dependence is a key factor influencing loop scheduling on VLIW (very long instruction word) architectures. Current research has not exploited the lockstep property of VLIWs. By making using of this property and centering on the concept of inclusion, a complete mathematical model for flow dependence analysis on VLIW architectures is presented in this paper. It is found that loop-carried flow dependencies form a set of disjoint linear ordered sets, and that there is one and only one basis that is independent and inclusive, making unnecessary all the other dependencies. The model allows multi-cycled operations and conditional branches. It lays a mathematical foundation for research on VLIWs, and is applicable to engineering practice.

Abstract:In this paper, the authors propose a scheduling and dropping algorithm for packets in QoS (quality of service) controlling for Internet. This method is based on the QoS parameters required by users, the types of multimedia and the waiting time, etc. It schedules the packets arriving in a router and allocates the buffers to meet the quality of services. The computer simulation shows that the forwarding performance of the proposed algorithm is superior to the WFQ's (weighted fair queuing), which is the common scheduling algorithm used in routers. So, this algorithm improves the transmission of multimedia information on Internet.

Abstract:Bayesian network is a graphical model that encodes probabilistic relationships among variables of interest. It is a natural way to express the causal information, and to discover the hidden patterns among the data. Learning of Bayesian network is to find out a network model that best represents the dependent relationships of the variables in a database, that is, given sample D and prior knowledge ζ, to find a Bayesian network S that fits the maximum posterior probability p(s^h|D,ζ). In this paper, the learning process of the network is strictly derived, and a case study is presented to indicate the applications of Bayesian network in data mining.

Abstract:A field theory based adaptive resonance neural network model, FTART2, is proposed in this paper. FTART2 combines the advantages of the adaptive resonance theory and the field theory, and achieves fast learning, strong generality and high efficiency. Moreover, FTART2 can adaptively adjust its network topology so that the disadvantage of manually configuring hidden neurons of traditional feed-forward networks is avoided. Benchmark tests show that FTART2 achieves higher accuracy and faster speed than standard BP.

Abstract:Recent studies have extended the scope of data mining from relational and transactional databases to spatial databases. Spatial data mining is a promising field, where the research work on spatial data classification is still in its initial stage. In this paper, the advantages and the disadvantages of several existing methods of spatial data classification are compared first. Then an effective three-step method, which is based on the rough set theory for spatial data classification, is proposed. The validity of this algorithm to the problem of incomplete spatial information is verified by pertinent experimental results.

Abstract:After a discussion of the importance of relationships among objects as a first-class modeling concept in object-oriented technology, a new object-oriented programming language, named as RCPP (relational C++), is presented and developed in this paper, which can directly and explicitly support the relationships among objects. Object relationships can be used to dynamically manipulate object behaviors and their propagation. In this paper, the authors explain the functional properties, the language model for the specification of object relations and its implementation mechanism, especially the semantic aspects of object relations. The running of RCPP programs is implemented first by translating RCPP source codes into C++ codes, and then by compiling and linking them to form executable programs.

Abstract:Since many real-time scheduling algorithms with fault-tolerance, reported in literature, can only schedule tasks with fault-tolerant requirements, the authors present a model of hybrid real-time fault-tolerant scheduling, and proposes a hybrid scheduling algorithm for real-time tasks in this paper. The static scheduling algorithm, a part of hybrid model can schedule tasks with fault-tolerant requirements together with those without fault-tolerant requirements. An algorithm, which is used to find out the minimal number of processors needed for the real-time tasks, is also presented in this paper, so the performance of the static scheduling algorithm can be simulated and analyzed. In order to enhance the performance of the static real-time scheduling algorithm with fault-tolerance, a dynamic scheduling algorithm is studied. The performance simulation and analysis of the scheduling algorithms are presented, and experiment results show that the performance is related with the number of tasks, computation time, period and the number of processors.

Abstract:Based on CHG, the authors present a framework for analyzing object-oriented programs——OOAF, and discuss the functions, the algorithms and the design ideas of OOAF in this paper. The algorithms for identifying subobject and determining visible methods and the most-dominant-method are also given to create visible-method-class-hierarchy-graph (VM-CHG). The virtual-function-call-graph (VFCG) of the program is obtained by computing inheritance-set, override-set, and override frontier, which provides a feasible way for understanding virtual function call in object-oriented programs.

Abstract:In this paper, the authors discuss the design and implementation of Java multi-threads. A new thread scheduling algorithm, named Preemptive Round-Robin Scheduling with a Free Queue, is presented. Under this policy threads in the free queue are not assigned with a constant priority. Their priority is the same as the highest-priority threads which are in running state. The highest-priority threads and the threads in the free queue are time-sliced scheduled. This algorithm solves the scheduling problem for independent looping thread. To improve the efficiency of thread synchronization, a new design for object lock is presented. It is called the Mixed Hash Object Lock. This design is a tradeoff between the lock efficiency and its space. The experimental results have proved that the design is feasible. Compared with the traditional design, the efficiency for locking and unlocking is high and the space allocated to lock is small.

Abstract:Baum-Welch algorithm, which is often troubled with overflow, is one of the basic methods in the field of speech signal processing. People have to adjust the inner parameters constantly. So in this paper, a modified Baum-Welch algorithm is presented to avoid the overflow completely. With this algorithm manual adjustment is not needed and the computation accuracy is guaranteed. There is no significant extra cost of computation and storage. The feasibility of the new algorithm is shown in the experiment.