Abstract:Scalability is one of the most important characteristics of next generation Internet. And it makes more rigid demands for high performance routers. Since high performance routing and switching technology is important for routers to provide this characteristic, we make comprehensive a survey on three aspects of the network processor design, routing lookup and high speed switching. Related works are also summarized. The future directions of development are analyzed.

Abstract:ForCES（Forwarding and Control Element Separation）and network processor play very important roles in the router design and implementation. IPv6, the emerging standard for Internet, whose framework has been widely accepted and deployed by the Internet community. In this paper，we design an efficient framework for implementing the Ipv6 router. The approach uses the ForCES protocol to establish communication between the control plane and the data plane. To implement a high performance data forwarding system, the data plane adopts network processor to handle dual IPv4/IPv6 stacks packet forwording, tunneling. We tested the whole prototype system performance by AX4000,the result is excellent.

Abstract:The fast development of Internet challenges the traditional POP network. SPR(Single POP Router) is considered as an effective approach to resolve this problem. A novel clustering technology called aggregating switching is proposed. It can aggregate several heterogeneous routers into one large scale SPR and outperform the traditional POP network in many aspects. Some open problems about aggregating switching are discussed.

Abstract:The IPv6 routing lookup algorithms need to process huge route tables in the future owing to the huge address space of IPv6, and each lookup needs more memory accesses than IPv4 algorithms because of the 128 bits address. To solve these two difficult problems, this algorithm converts the prefix into ranges and stores them in a B-tree, then uses range fragment tree in the nodes to reduce the memory access and the storage requirement. Theoretical analysis and the experimental results indicate that the algorithm can support the high performance lookup for huge IPv6 route tables.

Abstract:Researching on the characteristic of TCP flow helps to cognize the complex structure of network and to make use of the resource of network effectively. Referring to the characteristic of TCP flow and analyzing TCP congestion control of additive increase multiplicative decrease and packet out-ordering, we present TCP friendly design of router which accelerates the processing of router.

Abstract:It is a great challenge to forward packets at line rate as the line rate grows rapidly. This paper proposes a distributed parallel forwarding engine to solve this problem based on parallel processing. Adaptive Hashing dispatch algorithm(AHDA) is proposed to dispatch the packets received, which compromises between load balance and packet ordering. Simulation results show that AHDA algorithm dispatches load in a balanced way, reorders packets at low rate, and scales well with the number of network processors.

Abstract:Along with the continuing increase of network bandwidth, the time gaps between the arrived packets are decreasing. The changed situations demand the core router to improve the ability of packet processing. The traditional packet scheduling algorithms, such as WFQ, are difficult to implement in high performance core router because of the problems of performance and extensibility. The paper proposes a new un-precise packet scheduling scheme, which improves the performance of the scheduler by modifying the classical packet scheduling algorithms to simplify the design of hardware. The scheme doesn't affect the QoS quality of the algorithms. The delay performance of this scheme is analyzed and computer simulation results show that the delay performance of SLQF is consistent with that of LFQ on the whole.

Abstract:A parameterized analytical model of network processors is presented in terms of MPSoC design methodology, called NePlat. In this model, network applications are represented by Dataflow Process Network (DPN). Heterogeneous network processor resources are described by system parameters, and then the application model is mapped into the heterogeneous resources to analyze the performances of network processors.

Abstract:This paper proposes an algorithm based on Euler Spiral to complete the missing boundary in illusory contour. By this method, an active contour model is first initialized and extended turning the predictive visual interpolation one to a pure interpolation problem; then orientation estimation and Euler Spiral are combined to complete the missing boundary.The experimental results on illusory contour figures and a remote sensing image show the efficiency of the algorithm.

Abstract:To detect the coding synchronization of multidimensional TCM in the decoding part, two synchronization methods are proposed: the method of measurement accumulation, and method of original states of surviving paths. The former is suitable for no-fully transparent codes and the latter is suitable for both of the no-fully and fully transparent ones. They both use the mid-results of the viterbi decoder and have the advantages of less calculation, simpleness and stabilization.

Abstract:Based on some limitations of the traditional DSSS in military communication application, this paper puts forward a new kind of communication technical architecture of code-hopping DSSS and code-hopping/frequency-hopping mixed spread spectrum.It will probably solve some difficult problems in traditional DS/SS and improve the anti-reconnaissance capability.of the DSSS communication system in military communication application.

Abstract:DEVS provides a modular and hierarchical system modeling and simulation framework but lacks the ability of behavior modeling. Statecharts extend finite state machine and do outstanding jobs in modeling complicated behavior of systems by means of visual and flexible state-transition diagrams. Hence, a composable modeling approach, embedding the statecharts into DEVS to complement each other, is raised. It can be used to build complicated discrete control system. DEVS, as a system modeling framework, provides simulation logical support and statecharts expand the event interaction mechanism of DEVS which can be used to modeling system behavior logic. The paper details the syntax and semantics of statecharts and DEVS, and emphasizes on the mechanism of embedding statecharts into DEVS. The advantages of building discrete control system models by embedding statecharts into DEVS are illustrated by an application example.

Abstract:This paper establishes a uniform kinematics model to describe the undulation surfaces by the long flexible fin of Gymnarchus niloticus, with various factors comprehensively considered. Simulation is carried out using the morphological measured data and some simplified hypotheses to related factors. Compared with the filmed views, the simulation results show that the kinematics model can effectively depict the movement curved faces of the long flexible fin, and has better adaptability as well as versatility. Meaningfully, the establishment of the kinematics model for the long flexible fin has made a good preparation for further finite-element-based dynamic analysis.

Abstract:Gymnarchus niloticus, propelled by undulations of a long-based dorsal fin, is a typical representative of median and/or paired fin (MPF) fishes. Elongated-body theory (EBT) is one of the most effective theoretical models which can be applied to analyze the swimming performance of MPF fishes. This paper proposes an algorithm to detect the contour curve of undulations by the long-based dorsal fin, and furthermore the contour function is indeed a key factor in EBT. In this study, some videos are recorded when the specimen (which we name MPC-I) is freely swimming. We preprocess each frame of several selected video segments with a series of operations including rotation-calibrating, origin-matching, scale-regularizing as well as noise-reducing. Finally, the contour curve is fitted with a polynomial expression which is determined by a cluster of preprocessed undulation curve samples. The experimental results show that this algorithm is effective, and moreover the detected contour function is a good preparation for further analyses on the dorsal fin propulsion thrust and efficiency.

Abstract:HLA-based simulation system, regarded as a special kind of distributed system, often adopts rollback recovery to realize fault tolerance. Checkpoint interval is an important rollback recovery parameter that will seriously influence system performance. Firstly, we analyze the differences of fault tolerance between HLA-based distributed simulation system and the general distributed system. And then according to the different degrees of the importance of the simulation process to the simulation result, we classify simulation processes into trivial parts and critical parts. Furthermore, the availability of distributed simulation system, which adopts rollback recovery mechanism, has been defined and analyzed through the utilization of Markov chain. As a result, we achieve an equation, by which the checkpoint interval in the best system availability can be figured out. The correctness of this conclusion has also been testified through a set of experimental data.

Abstract:The ordering of independent components is a hot issue in independent component analysis (ICA), and the critical step for the robustness and computing complexity of independent feature space. In time series forecasting, a novel criterion has been presented based on the mechanism of first-order differential and minimum variance error under multiple components reconstruction. To avoid the exhaustive search in combinatorial optimization, a sub-optimum approach named Adding-Testing-Acceptance (ATA) is proposed. Experimental results show that the proposed method has a better forecasting ability and more efficient in comparison with the existing ones.

Abstract:The van der Pol-Duffing equation, excited under the parameters with two time delays, is studied. The main aim is to study 1/2 subharmonic-primary parametrical resonance bifurcation of this system. Perturbation method is used to obtain the bifurcation equation with time-delays, and numerical method is utilized to analyze the effect of time-delays on the steady state response. The research indicates that 1/2 subharmonic-primary parametrical resonance bifurcation can be well controlled by time-delays feedback.

Abstract:Large eddy simulations are conducted to investigate the coherent structures of a two-dimensional supersonic mixing layer.The Governing equations of large scale eddies are solved with fifth-order WENO scheme and the effect of small scale eddies is simulated with Smagorinsky subgrid scale model.The results show that the development procedure of large structures and reveal the behavior of large eddies.Frequency spectrum analysis of the unsteady pressure data is performed to investigate the acoustic characteristics of different streamwise location in the mixing layer.Time-averaged velocity profiles obtained from the simulation are consistent with the experimental data.

Abstract:The IR radiation characteristic of exhaust plume of the liquid rocket is numerically calculated. Based on the established HITRAN database and the numerical simulation results of the exhaust plume field, the radiation transmission equation is numerically solved. The absorption and radiation of the atmosphere are considered. The calculation results indicate that the transmission conditions, height and angle, exert important influence on the distribution of spectrum intensity. The spectral irradiances achieved by solving the RTE are compared with the test data, which show that the calculation method is feasible and the solution is proper.

Abstract:A novel satellite autonomy architecture with self-organization (SAASO) for multi-satellite coordination is put forward to fulfill the autonomous control demands of multi-satellite systems. The demands and characteristics of the autonomous control are analyzed. The components, functions and relationships of agents are introduced. The mechanism of the autonomous control based on SAASO is expounded, and the form of self-organization is discussed. SAASO is designed for various space missions and various autonomous satellites, thus it has generality to a certain extent.

Abstract:To remedy the defects of modeling by using of Agent and Petri Net individually, the ways of combining Agent and Petri net nowadays are reviewed, and Agent-oriented Petri Net(AOCPN) is put forward. The running rules and analytic methods of AOCPN are presented. With keeping targets on sea under surveillance and striking them as its tasks, the evaluation model of the space equipment architecture is built and its logical characteristics and operational effectiveness are analyzed. The modeling process and simulation results show that AOCPN can describe the logical and the physical functions of the architecture completely, and be applied to analyze the performance of the architecture and evaluate its influence on combat outcomes comprehensively. The procedure of the architecture carrying out the tasks can be traced effectively.

Abstract:The dynamic model of the parachute system with the inverted ‘Y’ suspension is established, which is widely used in recovery systems. The model is comprised of the general rigid-body equations considering apparent mass and the modeling method of suspension system based on the assumption of “equilibrium points”. A simulation analysis for a spaceship's airdrop experiment is carried out and the validity of models is showed by the experiment data.

Abstract:C/SiC composites fabricated by precursor infiltration pyrolysis process are prone to oxidation above 400℃. Their oxidation behaviors have been researched on. The oxidation loss of C fiber textiles or C/SiC composites without coating increases at 400℃ to 1300℃. However,the oxidation loss of C/SiC composites with CVD-SiC coating or three-layer coating, respectively CVD-SiC layer, self-sealing layer and CVD-SiC layer, at first increases at 500℃ to 700℃, then decreases at 800～1300℃. Oxidation loss of C/SiC composites with three-layer coating are less than 6.4% at 800～1300℃ for 288h.SEM photographs indicate vitreous self-sealing layer filling in cracks.

Abstract:The solid fast ionic conductors are synthesized by sol-gel-hydrothermal method. The influences of the dispersion reagent, the mass of alkalinity and the temperature of calcinations on the surface description of nanometer-powder and electric conductivity are discussed. When PEG 12000 is used as dispersion reagent,the mass of alkalinity is 1.0%,the temperature of calcinations is 350℃, and the electric conductivity of Lithium Fast Ionic Conductor synthesized at the room temperature is 2.59×10^-3S?cm^-1.

Abstract:Spark-05 was built for conducting researches on medium/high impedance high power microwave sources. The interface between oil and vacuum was one of the keys to the system design. For reasons of compactness and simplicity, a disk type vacuum insulator was adapted. Electrostatic field along the insulator surface was calculated. By adjusting the position of the shielding rings, electric fields were well distributed, and the field around the triple junction was controlled. The insulator worked well in 1MV experiments afterwards.

Abstract:The coaxial plate-inserted mode converter is a novel high-power microwave TEM-TE₁₁mode converter. The mode conversion processes of the mode converter were reviewed, and the reflective characteristics were studied by analyzing the waveguide discontinuousness. It is revealed that, the reflection modes of the mode converter are TEM mode and TE₁₁mode, the reflection TE₁₁ mode can be canceled by choosing the right radii of the coaxial waveguide, and the reflection TEM mode can be diminished by setting matching poles and properly designing the cone at the end of inner conductor.

Abstract:Let u and v be two fixed vertices. Connect u and v by b internally disjoint paths of length a , the resulting graphs denoted by P_a,b.Kathiresan showed that P_2r,2m-1 is graceful and conjectured that P_a,b is graceful except for (a,b)=(2r+1,4s+2).Yang showed that P_2r+1,2m-1 are graceful. In this paper ,P_3，4m is proved to be graceful. So we prove the conjecture for a=3.