Journal of Northeastern University(Natural Science)

Simulation of TCP congestion control algorithms in LEO satellite networks

Gong, Chang-Qing (1); Zhao, Zhi-Gang (1); Wang, Guang-Xing (1)

2004, 25 (10): 919-922. DOI: -

Abstract ( 284 )

HTML

PDF (159KB) ( 690 )

The performance of TCP (Transmission Control Protocol) congestion control algorithms over Iridium-like LEO (Low Earth Orbit) satellite networks is analyzed, including Reno, Newreno, SACK and Vegas algorithms. The focuses are on the whole network performance of various parameters and the average numerical value. The result of simulation shows that the Vegas algorithm is superior to other algorithms in average RTT, though there are no obvious differences among the four algorithms in the average throughput. By comparison with the result of simulation of single link, using the complete network topology to simulate TCP congestion control algorithms is more positive.

References | Related Articles | Metrics

Destruction-resistant routing algorithm in LEO/MEO satellite networks

Wang, Ya-Sha (1); Li, Dong-Ni (1); Feng, Jin (1); Wang, Guang-Xing (1)

2004, 25 (10): 923-926. DOI: -

Abstract ( 238 )

HTML

PDF (779KB) ( 670 )

Proposes a destruction-resistant routing algorithm based on ATM (Asynchronous Transfer Mode) technology in LEO/MEO networks, of which the satellite network is virtualized as a fully connected network and divided into several clusters that can be regrouped according to satellites' position. The real-time topology information within a cluster is maintained by the clusterhead. When some intersatellite links fail, the clusterhead will try its best to rebuild the broken VPC (Virtual Path Connection) within the cluster, and, when fails once more, the clusterhead will diffuse the request for rebuilding VPC and the topology information on its own cluster to its neighboring clusterheads, so as to rebuild a VPC within a larger extent of known topology. The simulation result on Iridium system model shows that the diffused destruction-resistant routing algorithm can not only make the system stronger by avoiding the broken ISLs but also greatly reduce the path-finding overheads in comparison with OSPF (Open Shortest Path First) algorithm.

References | Related Articles | Metrics

H.248-based IP user media gateway design model

Li, Ying (1); Li, Hai-Xu (1); Wang, Guang-Xing (1)

2004, 25 (10): 927-930. DOI: -

Abstract ( 238 )

HTML

PDF (789KB) ( 516 )

Introducing briefly the concepts of NGN (Next Generation Network) and softswitch, and according to China's General Technical Requirements for Softswitch Devices, the design model based on H.248 protocol for IP user media gateways is presented basically. An emulator was designed to emulate the system so as to explain the function and implementations of each and every module and the voice handling flow between two user media gateways. The difficulties and key technique are discussed for system implementation. Using the design model can protect the huge investments in PSTN (Public Switched Telephone Network), resolve all existing problems of how to smoothen users' accessing NGN, implement many basic services especially the value-added services and enable PSTN to transit to NGN smoothly.

References | Related Articles | Metrics

Reconfiguration and recovery in LEO satellite communication networks

Zhao, Zhi-Gang (1); Liu, Jun (1); Yu, Hua (1); Wang, Guang-Xing (1)

2004, 25 (10): 931-933. DOI: -

Abstract ( 242 )

HTML

PDF (155KB) ( 871 )

A dynamic routing-based reconfiguring and recovering algorithm for LEO satellite systems employing ISL is proposed, by which a designated satellite with management responsibilities is available to check out as soon as possible a ranked list of failed network links and further, detect accurately the failed links through a quick and autonomous testing. Then, it will reconfigure the network topology. After the failed links have been repaired, the satellite with management responsibilities will recover all of them in network topology. The simulation in satellite networks shows that this method has the features of real-time and accuracy especially it will occupy a few bit of extra bandwidth even requires no bandwidth.

References | Related Articles | Metrics

Particle swarm optimization approach of solving communication optimization problems

Yuan, Ping (1); Wang, Guang-Xing (1); Zhang, Yang-Yang (1)

2004, 25 (10): 934-937. DOI: -

Abstract ( 258 )

HTML

PDF (109KB) ( 688 )

Briefly introduces the principle, procedure and relevant parameters of the particle swarm optimization approach. Then, a multiobjective particle swarm optimization or divided range multiobjective particle swarm optimization (DRMPSO) based on distributed computation is presented to solve the base station placement problem. The simulation results indicate the effectiveness of DRMPSO and it is convincible that DRMPSO is an efficient way to solve other problems in communications, such as channel assignment, network topological design, IP multicast, Ad hoc networks clustering, WLAN multicast.

References | Related Articles | Metrics

Mining association rule on rough entropy basis in detecting escapes from traffic accidents

Zhao, Hai (1); Chen, Yan (2); Zhang, De-Gan (1); Zhang, Xiao-Dan (1)

2004, 25 (10): 938-941. DOI: -

Abstract ( 259 )

HTML

PDF (161KB) ( 545 )

Traditional data mining algorithms have unavoidably errors arising from additional man-made uncertain factors in dealing with multi-source information. A new mining method called association rule is therefore proposed basis in view of rough set theory, with another method specially designed to assess it. Method from historical data of the crimes escaping from traffic accidents in a designed way, the association rules will provide an efficient and practical means for the police to detect the crimes escaping from traffic accidents. Some applications have exemplified the effectiveness of the method proposed.

References | Related Articles | Metrics

Detecting BGP policy conflicts through dispute arc

Wang, Hong-Jun (1); Wang, Rui-Jun (1); Wang, Da-Dong (2); Gao, Yuan (1)

2004, 25 (10): 942-945. DOI: -

Abstract ( 264 )

HTML

PDF (158KB) ( 509 )

A method to dynamically find BGP policy conflicts is proposed. Based on the theory of dispute digraph, this method will find the autonomous systems (ASs) between which the routing policies conflict with each other and corresponding routes by constructing a dispute arc. The relative preference of route is used to describe the preference relationship between two different routes to the network of the same destination. According to the formation conditions of the dispute arc, the difference of relative preference between two adjacent AS routes to the same destination indicates that the routing policies of the two ASs are conflicting with each other. In order to find route policy conflicts, a new attribute is defined in BGP to carry repeatedly the relative preference of the newly selected best route to the previous one. The simulations through SSFNET show that the method is effective and feasible.

References | Related Articles | Metrics

Packet scheduling algorithm for multistage switching Fabric's terabit routers

Fan, Li-Jun (1); Zhang, Tie-Jun (2); Quan, Cheng-Bin (1); Luan, Gui-Xing (2)

2004, 25 (10): 946-949. DOI: -

Abstract ( 232 )

HTML

PDF (168KB) ( 515 )

A novel packet scheduling algorithm named DRLB-BvN (dynamic resequencing load balanced Birkhoff von Neumann) is put forward to improve the LB-BvN algorithm proposed by Chang et al, which is easy to mis-sequence information elements. DRLB-BvN algorithm is able to regulate dynamically the threshold timestamping mechanism according to latest delay in switching information elements, thus resolving the problem of mis-sequence of information elements which the LB-BvN is easy to come about. Simulation results showed that the performance of DRLB-BvN algorithm lowers unobviously for the sake of resolving the problem in comparison with BvN algorithm.

References | Related Articles | Metrics

FEM analysis for roll deformation of a 4-high cold strip mill

Shi, Xu (1); Liu, Xiang-Hua (1); Wang, Guo-Dong (1); Li, Shan-Qing (2)

2004, 25 (10): 957-960. DOI: -

Abstract ( 270 )

HTML

PDF (195KB) ( 870 )

By way of three-dimensional elastic-plastic finite element method based on the FEM software Marc, the strip cold rolling process on a four-high strip mill is modeled to analyze the roll deformation. In the model the deformation of both roll and strip are considered as a whole, so as to resolve the coupling problem between them, and avoid the errors due to iterative processor assumption of the distribution of rolling force. A step-by-step convergent solution is thus introduced to the calculation to provide accuracy and reliability. The effects of strip width, roll's bending force, etc., on such roll deformation as bending and arc/lateral flattening, and the gap between loading rolls are obtained with different rolling parameters. The new calculation method and reference data will offer an efficient way for analysis and control of the shape.

References | Related Articles | Metrics

Development and realization of mathematical model of laminar cooling system of hot-rolled strips

Xie, Hai-Bo (1); She, Guang-Fu (2); Liu, Xiang-Hua (1); Wang, Guo-Dong (1)

2004, 25 (10): 961-964. DOI: -

Abstract ( 272 )

HTML

PDF (240KB) ( 765 )

The heat transfer process in the laminar cooling system of hot-rolled strips is analyzed to develop the air/water-cooling mathematical models so as to control cooling process. As linear regression models, they are different from the theoretical expressions in term of exponential temperature drop of which the regressive data can be collected in situ. So, they are more practical and in simpler from with higher accuracy. The setting values, including presettings and that for modification, in calculation of laminar cooling and how to realize them in programming are described in detail. High cooling capacity, wide range of cooling rate, favourable running and satisfactory control effect feature the cooling system. It will meat the requirements for smoothening rolling process and developing new products.

References | Related Articles | Metrics

High precision plate crown model for on-line application

Hu, Xian-Lei (1); Zhao, Zhong (1); Qiu, Hong-Lei (1); Liu, Xiang-Hua (1)

2004, 25 (10): 965-968. DOI: -

Abstract ( 267 )

HTML

PDF (939KB) ( 587 )

The influence function is used to analyze the influences of workpiece width, rolling force, sizes of work/back-up rolls and bending force acting on roll on the crown of loaded rolls of a conventional 4-high plate mill. With lots of data regressed in calculation, an on-line computation model is given for loaded roll crown. Then, the genetic effect of workpiece crown at entry of rolling process on that at exit is discussed, and an on-line prediction model of plate crown is thus given integrating the model of loaded roll crown with that of plate crown's genetic coefficient. The model can predict actually the crown magnitude of rolled plate with high precision because the effect of lateral plastic flow of workpiece in rolling process has been taken into account. So, it is efficient to on-line control of plate shape especially its crown.

References | Related Articles | Metrics

Study on backward slip during H-beam rolling

Xu, Xu-Dong (1); Liu, Xiang-Hua (1); Wu, Di (1); Zhang, Dong (2)

2004, 25 (10): 969-972. DOI: -

Abstract ( 251 )

HTML

PDF (300KB) ( 484 )

The deformation process of H-beam was modeled by the explicit dynamic finite element method with different rolling parameters to make sure how and why the backward slip formed during H-beam rolling. The results showed that in the rolling process of H-beam the influence of area reduction on backward slip is so apparent that the backward slip increases sharply with increasing reduction. In addition, the flange width, elongation ratio of flange to web, inner width of H-beam and initial thickness of workpiece also affect greatly the formation of backward slip during H-beam rolling.

References | Related Articles | Metrics

Adaptive learning of the deformation resistance model for tandem cold rolling process control

Wang, Jun-Shen (1); Zhao, Qi-Lin (1); Jiao, Zhi-Jie (1); Liu, Xiang-Hua (1)

2004, 25 (10): 973-976. DOI: -

Abstract ( 278 )

HTML

PDF (240KB) ( 965 )

A nonlinear formula is derived for tandem strip cold rolling process based on Bland-Ford-Hill model of rolling force and the deformation resistance model in view of both the plastic deformation of workpiece in deformation zone and elastic deformation at either the entrance or the exit of the zone. The formula is given in terms of indirect substitution, i.e., substitute the measured rolling force values for the corresponding unknown parameters in the deformation resistance model and then solve the formula to obtain the actual value of the resistance to deformation. Further, the adaptive learning coefficient of the deformation resistance model can therefore be obtained by way of exponential smoothing average. The application results showed that the accuracy and stability of the actual value of the resistance to deformation obtained in such a way are available to meet the requirement of on-line process control and improve the calculation accuracy of both the deformation resistance model and rolling force model.

References | Related Articles | Metrics

Study on growth of rutile (TiO2) single crystals

Bi, Xiao-Guo (1); Xiu, Zhi-Meng (1); Ma, Wei-Min (2); Sun, Xu-Dong (1)

2004, 25 (10): 977-979. DOI: -

Abstract ( 352 )

HTML

PDF (226KB) ( 1018 )

The processing conditions under which the rutile single crystal grows by way of flame fusion were studied comparing its growth and annealing results in different atmospheres. The results showed that the oven atmosphere is decisive to whether crystal could be formed or not, which may be controlled by the ratio of hydrogen to oxygen supplied in it. The crystal is easy to grow in a wide growth rate range to become a relatively big one if the oven atmosphere is suitable. It was revealed that no complete crystal can be obtained relying simply on adjusting (increasing or decreasing) the growth rate. The oven atmosphere in which the oxygen partial pressure is higher than oxygen's dissociation pressure at the liquid-solid interface is proved a prerequisite condition to form a complete crystal. The role of annealing is to relieve thermal stress formed in crystal growth but more important, it shall eliminate oxygen vacancies by oxidization reaction. The annealing time can be considerably shortened in an oxygen atmosphere in which a transparent light yellow rutile single crystal is obtained as its natural color.

References | Related Articles | Metrics

Effects of catalyst and substrate on the preparation of carbon nanotubes by ECR-CVD

Wang, Zhi (1); Ba, De-Chun (1); Lin, Zeng (1); Liu, Fei (2)

2004, 25 (10): 980-983. DOI: -

Abstract ( 228 )

HTML

PDF (224KB) ( 1025 )

Carbon nanotubes (CNTs) were prepared by way of electron cyclotron resonance chemical vapor deposition (ECR-CVD) using CH4 and H2 as source gases. During the preparation the Fe3O4 nanoparticle, Co nanoparticle and Fe(NO)3 sol were used as catalysts to prepare CNTs on porous silicon substrates, and the Fe3O4 nanoparticle was used singly as catalyst to grow CNTs on Si(111) and quartz substrates. Scanning electron microscopy (SEM), transmission electron microscopy (TEM) were used to evaluate the morphology and structure of the CNTs prepared. The results showed that the catalyst affects the nucleus density and growth rate and the substrate affects greatly the morphology and growth mode by controlling the properties and distribution uniformity of catalyst. Among the different CNTs, the well-aligned carbon nanotubes are those which are prepared on porous silicon substrate with Fe3O4 nanoparticle as catalyst, as revealed in the study.

References | Related Articles | Metrics

Disposal of waste plastics by coking process - The effect of organic additive (OA)

Yu, Guang-Wei (1); Cai, Jiu-Ju (1); Lei, Yong (1); Liao, Hong-Qiang (2)

2004, 25 (10): 984-987. DOI: -

Abstract ( 203 )

HTML

PDF (109KB) ( 675 )

Using a thermal balance and 200 kg pilot coke-oven, the thermal weight loss behavior and characteristics of the cokes produced from co-coking of waste plastics (WP), organic additive (OA) and the coking coal supplied by SHOUSTEEL in Beijing were studied. The results showed that the synergetic effect arising from the pyrolysis of WP, OA and the coal in their common temperature range of weight loss will affect the quality of coke. Moreover, during co-coking the mixture of 1% WP, 2% OA and 97% coking, the addition of 2% OA can eliminate the negative influence of the 1% WP on the quality of coke produced from co-coking, thus resulting in the increase of M40 and CSR and decrease of M10 and CRI obviously. Therefore, the technology proposed will have bright outlook in industrial applications.

References | Related Articles | Metrics

Factors on crystalline temperature of mold flux

Zhu, Chuan-Yun (1); Liu, Cheng-Jun (1); Wang, De-Yong (1); Jiang, Mao-Fa (1)

2004, 25 (10): 988-990. DOI: -

Abstract ( 230 )

HTML

PDF (560KB) ( 657 )

The single hot-wire thermocouple technique was used to study the influencing factors on the crystalline temperature of mold flux, and how the different compositions of flux used affect the crystalline temperature was revealed. The results showed that the crystalline temperature of mold flux rises with the increasing basicity, mass fractions of NaF and CaF2 contents, and the crystalline temperature lowers gradually with the increasing mass fractions of Al2O3, MgO, Na2O, K2O, BaO, MnO and B2O3 contents. Furthermore, the crystalline temperature decreases first then comes to its lowest value at 2% with the increasing mass fractions of Li2O content, and goes up from the lowest point later on.

References | Related Articles | Metrics

Electrolysis corrosion of large-size ferrous-nickel-based cermet inert electrode

Wang, Zhao-Wen (1); Luo, Tao (1); Gao, Bing-Liang (1); Qiu, Zhu-Xian (1)

2004, 25 (10): 991-993. DOI: -

Abstract ( 273 )

HTML

PDF (181KB) ( 784 )

A ferrous-nickel-based cermet inert anode was prepared by means of hot-pressing with a size up to D150 mm × 20 mm to apply to a 100 A electrolysis testing for 24 h at 900°C. In the whole electrolysis process the cell voltage was steady, which manifests that the anode is good at electrolytic conduction. According to the SEM results of the anode after electrolysis, it was found that the anodic corrosion due to electrolyte includes two processes. One is the reoxidation reaction between the oxygen ions produced from the anodic discharging of AlxOyFz^(2y+z-3x)- ions and the metal in anodic substrate, which causes the metallic oxides to be dissolved in electrolytic molten salt. The other is the anodic deposition of AlF3 into gaps, which is produced from the anodic reaction. A conclusion is thus drown that the coefficient of the anodic corrosion layer is different from that of anodic substrate and such a difference will result in the delamination even peeling off of the surface of anode in the cooling process. The corrosion rate is evaluated as 18 mm/a.

References | Related Articles | Metrics

Thermodynamics of irreversible process in homogeneous single chemical reaction

Zhai, Yu-Chun (1); Wang, Jin-Xia (1)

2004, 25 (10): 994-997. DOI: -

Abstract ( 345 )

HTML

PDF (249KB) ( 1164 )

The prevalent linear thermodynamic expressions of irreversible process are just available to those physicochemical changes which are close to equilibrium. On the other hand, the studies on the physicochemical changes far from equilibrium are mainly focusing on dissipative structures. A nonlinear dynamic equation is therefore set up for the chemical reactions far from equilibrium by way of extending the linear expressions to the homogeneous single chemical reaction systems far from equilibrium with no dissipative structure involved. The equation has been applied to water-gas reaction with a quintic expression obtained in high conformity to measured data with a maximum error of 8%. The result reveals that in such a way the thermodynamics of irreversible process can be applied to homogeneous single chemical reaction systems far from equilibrium.

References | Related Articles | Metrics

Synthesis and research of spinel LiMn2O4 by sol-gel method

Xu, Cha-Qing (1); Tian, Yan-Wen (1); Zhai, Yu-Chun (1); Wang, Zi-Xia (1)

2004, 25 (10): 998-1001. DOI: -

Abstract ( 244 )

HTML

PDF (219KB) ( 1001 )

The spinel LiMn2O4 for lithium-ion battery was prepared by sol-gel method using Mn(Ac)2·4H2O, (AC)Li·2H2O and harmless low-molecule citric acid as raw materials. The mechanism of synthesizing LiMn2O4 with dry gel was found by TGA-DTA analysis at different heating rates. In the process of dry gel burning, large quantity of the LiMn2O4 were formed because of the small grain size of the products Mn3O4 and Li2O2 resulting from burning, and the residual Mn3O4 and Li2O2 reacted completely at about 300°C, thus lowering greatly the synthesizing temperature. XRD pattern of the LiMn2O4 showed that the product samples are all of a pure phase with spinel-type structure, sharp peak and perfect crystalline. Investigating the effect of pH values on the grain size and electrochemical property of samples, the SEM analysis showed that the electrochemical capacity of the LiMn2O4 samples increases with the increasing pH value, of which the grain size distribution is uniform at a sub-micron level if pH value is 6.0. The charge/discharge testing result at 10% current of initial discharge capacity and 3.30-4.35 V showed that the initial discharge capacity of the synthesized samples is 121.0 mAh/g. It implies a good electrochemical property.

References | Related Articles | Metrics

Dynamic distribution control of braking force on turning a vehicle

Yu, Yan (1); Xie, Li-Yang (1); He, Hui (2); Zhang, Li-Jun (2)

2004, 25 (10): 1002-1005. DOI: -

Abstract ( 281 )

HTML

PDF (175KB) ( 708 )

To distribute the braking force to both the left and right wheels of a vehicle is one of the most important factors for braking stability when the vehicle is braking on turning. A new method to control dynamically the braking force distribution on turning a vehicle is proposed on the principle of forward-feed control of steering angle, with the intention of meeting the demand of braking stability on turning a vehicle under any road condition. A 3-wheel vehicle model of 2 degrees of freedom is thus established for a simulation of braking force distribution, of which the results show that this method greatly reduces the difference between the actual and ideal lateral wobble frequencies in the process from the very beginning of braking operation to its end. The steerability and effectiveness of the method are therefore verified.

References | Related Articles | Metrics

Damage analysis of side joints of HSRC frame

Xu, Ya-Feng (1); Jia, Lian-Guang (2); Wang, Lian-Guang (1); Liang, Li (1)

2004, 25 (10): 1006-1009. DOI: -

Abstract ( 264 )

HTML

PDF (775KB) ( 423 )

Based on tests, the influencing factors on the damage as shown in the topic were considered to ascertain the damage variable and damage model which is adaptable to the side joints of HSRC column/HRC beam under low cycle repeated load. The different phases of damage growth and their features were analyzed under the cyclic earthquake load, as well as the relation of damage to displacement. The result showed that the damage level of joints with increasing axial compression ratio, and the smaller the rate of stirrup used in core area is, the higher the damage level of joints rise but the ductility decreases slightly will be with ductility decreased slightly. On the contrary, the damage level of joints lowers but the ductility increases.

References | Related Articles | Metrics

Modeling prediction of composite foundation settlement

Wang, Feng-Chi (1); Zhu, Fu-Sheng (1); Kang, Yu-Mei (1); Chen, Bai-Ling (1)

2004, 25 (10): 1010-1012. DOI: -

Abstract ( 259 )

HTML

PDF (670KB) ( 504 )

Based on an analysis of the developing mechanism of composite foundation settlement in combination with the characteristics of Verhulst gray model, a Verhulst gray model for the composite foundation settlement was specially developed to reflect the characteristics of settlement-time curve, such as its monotone increasing, boundedness and S-shape. The precision check proved that by such a Verhulst gray model less observing data are required to set up a prediction model with reliable precision through a mathematical reasoning process, which will reveal the long settlement process of composite foundation.

References | Related Articles | Metrics

Analysis of environmental isotopes for groundwater in pediment tableland region of China's western northeast plain - Exemplified with the middle upper reach of Taoer River basin

Jia, Wei-Guang (1); Di, Zhi-Qiang (2); Jin, Hong-Tao (2); Wang, En-De (1)

2004, 25 (10): 1013-1016. DOI: -

Abstract ( 243 )

HTML

PDF (235KB) ( 506 )

Based on the testing results of isotopes for collected samples of surface water and groundwater, the relationships of δD to δ¹⁸O and δ¹⁸O to H were plotted so as to analyze the relations among the recharge, runoff and discharge of groundwater in the pediment tableland region. The following equations were thus given, i.e., for the regional rainwater relation line δD=7.1781δ¹⁸O-8.1151 and for the elevation effect relation line δ¹⁸O-0.0049H-8.209. Meanwhile, the Greater Xing'an Mountains are reckoned the recharging areas of groundwater sourcing from meteoric water. Inside the areas the circulation of groundwater is rather slow and the recharging source is mainly the meteoric water accumulated since 1999 and formed within 3.26-10.6 years. The annual average temperature during the formation of groundwater is a little bit lower than that at present.

References | Related Articles | Metrics

Criteria for strictly generalized and diagonally dominant matrices and nonsingular M-matrices

Shao, Xin-Hui (1); Shen, Hai-Long (1); Gao, Yi-Ming (2); Li, Chang-Jun (1)

2004, 25 (10): 1017-1019. DOI: -

Abstract ( 182 )

HTML

PDF (155KB) ( 1494 )

Based on the other earlier works as shown in reference and the characteristics of the elements of an irreducibly and weakly diagonally dominant matrix, the row elements of the complex matrix A are divided into three parts, then the module of the elements of each and every part are summed up to obtain the three values αi, βi, and γi. Comparing the magnitudes of the values of hik and Hjk, which are both constructed by αi, βi, and γi, the sufficient conditions are given to judge whether an irreducibly and weakly diagonally dominant matrix is a generalized strictly diagonally dominant matrix or not. Then, the results can be applied to the judgment of a nonsingular M-matrix so as to extend the main conclusions as indicated by other earlier works.

References | Related Articles | Metrics

Archive