Showing papers in "Journal of Nanjing University of Posts and Telecommunications in 2011"

Survey on the Internet of Things

Abstract: This paper presents a survey on the Internet of Things(IoT).The origination,development and application of IoT are first discussed,then the definition,architecture,transport network layer,sensing node and the terminal,pervasive sensor network,smart processing techniques,service platform techniques,security,and standardization of IoT are discussed in detail.Finally the key technologies and primary research topics are proposed.

Joint Optimization of Sensing Period and Sensing Interval in Cognitive Radio Networks

Abstract: Sensing/monitoring of spectrum-availability has been identified as a key requirement for dynamic spectrum allocation in cognitive radio networks(CRNs).Sensing duration and sensing interval are two significant parameters associated with spectrum sensing,which directly affect the detection probability of available spectrum and the collision probability of the primary and secondary users.In this paper,a two-dimensional single-objective optimization problem of sensing period and sensing interval is proposed under the constraint that the primary users are sufficiently protected.Based on the energy detection sensing algorithm,the mathematical model of joint optimization is given.We present the approach of solving the sensing period and sensing interval optimization problem which makes the maximum utilization of spectrum for the secondary system.Simulation results show that the spectrum utilization increases at first and decreases then as the sensing interval and the sensing period increase.So there exists a set of optimal values of sensing interval and sensing period to maximize the spectrum utilization of the cognitive networks.

A Camera Self-calibration Method Based on Genetic Algorithm and LM Algorithm

Abstract: A camera self-calibration method is proposed by combining genetic algorithm with LM algorithm.Firstly,we use genetic algorithm to get several groups of initial value of camera parameter,then we use the LM algorithm to optimize each group of parameter,and select the group of parameter with minimum projection error.If the minimum projection error meets the requirement,this group of parameter is the final result of camera self-calibration.Otherwise,we use genetic algorithm to cross and mutate these groups of parameters,and reuse LM algorithm to optimize each group of parameter until the projection error meets the requirement.Finally,we estimate the disparity with the camera parameters from camera self-calibration to judge the correctness of these parameters.Experimental results show that the proposed camera self-calibration method based on LM algorithm and genetic algorithm is correct and effective.

Underdetermined Blind Source Separation Applied to the Ultra-Wideband Communication Signals

Abstract: A novel two-stage method of underdetermined blind source separation named ladder diagram-minimum angle is proposed for the ultra-wideband communications signalsIn the first stage,where the mixing matrix is estimated,cluster calculation is performed on the uniform sampling of data points by using ladder diagram,thus a engineering model for estimating the mixing matrix has been established,which is used to directly estimate the number of source signals and get high precision estimates of the mixing matrixIn the second stage,the source signal is estimated by using the minimum angle separation criterion and the better separation performance is thereby obtained

Network Traffic Prediction Based on Wavelet Transformation and FARIIMA

Abstract: A new network traffic prediction scheme is proposed based on wavelet transformation and FARIMA ModelFollowing the decomposition of original traffic with wavelet,each branch is reconstructed with mallat algorithmThen prediction of reconstructed branch is conducted using FARIMA Model and finally the traffic is integratedSimulations with real network traffic have verified the precision of proposed prediction algorithmCompared with the method that first predicts the wavelet coefficients with FARIMA model and then reconstructs the traffic,the proposed method reduces the prediction error

P2P Traffic Identification Model Based on Ensemble Learning

Abstract: P2P traffic identification model is constructed based on the ensemble learning algorithm,which integrats DTNB,ONER and BP neural network algorithmUsing network flow characteristics and the integrated classification algorithm for rule generation in machine learning,network traffic flow is divided into two types,P2P and non-P2P trafficThe model consists of three steps,iegaining network flow characteristics,P2P traffic feature selection and the establishment of flow classification modelThe rationality of the model and the effectiveness of the proposed method are evaluated using method of combining T-fold cross-validation and test setsThe experiment results have shown that the average precision of traffic classification reaches 9727% and the model has relatively high P2P flow identification accuracy

Application and Analysis of the Polarization Characteristic of Radio Wave Propagation Based on the Method of Ray Tracing

Abstract: The polarization model is the significant polarization characteristics in the radio prapogation.In this paper,the prapogation characteristics of indoor wireless radiowave has been studied based on the method of Shooting-and-Bounding Ray/Image(SBR/Image).Furthermore,the Field strength including the polarization characteristics of two-times reflection has been put forward.Finally,The path loss and field strength has been predicted in typical indoor office environment.A good agreement is achieved between the simulated results and the published results.

The concept and communication process of network buzzwords

Abstract: With the popularity of the Internet in China,Chinese network buzzwords update in an accelerating pace and gradually penetrate from the Internet to the traditional media and social lives,even go into public language.This paper,starting from the concept of network buzzwords and their differences from network common words,points out event correlation is the cause of the spread of network buzzwords.The psychological factors to affect the popularity of network buzzwords include both individual and group ones.And network buzzwords feature in social,entertainment and non-standard characteristics.

Quantum McEliece Public-Key Cryptosystem Based on Quantum QC-LDPC Codes

Abstract: Quantum public-key cryptosystem includes unconditionally secure one and computationally secure one.In this paper,we presents a quantum public-key cryptosystem of computationally secure by using classical public key algorithm.Quantum LDPC codes is a linear codes over GF(4),and there exists a quick decoding algorithm,such as BP algorithm.We propose a quantum McEliece public-key cryptosystem by using quantum QC-LDPC codes.We give the encryption and decryption of this cryptosystem.By numerical simulation,we analyze the performance of this cryptosystem over quantum McEliece channel.Compared with classical McEliece algorithm,quantum one expands the keyspace greatly,and improves the security of cryptosystem.It is shown that the work factors have been improved from 274(classical McEliece algorithm) to 2570(quantum McEliece algorithm) when transmission rate is 0.60,and quantum one can effectively resist quantum Grover algorithm attack.

Research on Supply Chain Resources Optimal Allocation Support System Based on Linear Programming Inverse Optimal Models

Abstract: To help the supply chain response efficiently to the market and operate in coordination with low cost,the Linear Programming Inverse optimal(ILP) models are set up for supply chain resources optimal allocation(SCROA)And then a support system SCROA_SS with the ILP models' building and solving module as core is analyzed and designed based on Knowledge Representation,it is realized through J2EE framework and XML related technologyThe applied case shows that SCROA_SS has unique advantages:According to the market,it can provide decision support of supply chain resources optimal allocation quickly,moreover,improve the comprehensive competitiveness of the entire supply chain

Study on Replica Placement Algorithm in Tree-based Data Grid Environments

Abstract: Data replication is an effective method to reduce data access cost,decrease network bandwidth requirement,increase ability of fault tolerance and so on.But the number of replica is limited by some factors,such as storage capacity of nodes.Replica placement has become a problem worthy of study in replica management of data grid.For a tree-based data grid with network bandwidth-constrained,considering load balance of each server with replica,this paper proposed a k-replica placement algorithm based on dynamic programming and binary search,and the time complexity is less than O(n2logn).Experiments show that this replica placement algorithm can get better load balance and demonstrates a good performance in data access.

A Novel Replica Selection Strategy Based on Combination Algorithm in Data Grid

Abstract: After researching replica selection strategies based on genetic algorithm and ant algorithm and considering their advantages and disadvantages,we propse a new replica selection strategy based on combining algorithms.The fitness function of genetic algorithm is used to determine pheromone distribution in ant algorithm,and then ant algorithm is used to select the best replica,so the new algorithm solves the problem of low efficiency of genetic algorithm and lack of pheromone in the beginning of ant algorithm.Simulation results are displayed graphically.Through analysis and comparison of simulation results,the evaluation demonstrates that the new algorithm can be effective in reducing the operational running time and improve the speed of replica selection in the whole grid environment and in a single site,so it improves overall performance.

Echoed images and cultural metaphor in The Kite Runner

Abstract: By weaving in The Kite Runner with various culturally metaphorical images as kite,pomegranate tree,slingshot and harelip,Khaled Hosseini brings poetics to his narration.These recurring images are arranged to echo in the novel,so the structure of its extensive narration has been packed and the narrative tension has been strengthened.Meanwhile,this narrative technique also enriches the spiritual dimension of the characters and enhances the theme of the novel.

A Combination Algorithm Based on DES

Abstract: With the growth of computing speed,DES algorithm can't resist brute force attacks and no longer meets today's network security requirements of data encryption because of short key length.One way to improve its anti-attack performance is to increase the length of the key and the texts.The design and implementation of an algorithm is introduced,which improves DES Algorithm by adding the key's length to 112 bits and adding the plaintexts'/ciphertexts' length to 128 bits at the same time.So the algorithm can resist the attacks from brute force more effectively.

An empirical study of producer service agglomeration in Yangtze River Delta region

Abstract: The agglomeration degree of producer service is a significant index to its development.This paper utilizes the data of Yangtze River Delta region as samples and takes the Spatial Gini Coefficient and Location Entropy as the analysis methods.The results indicate that the producer service in this region at present exhibits high degree of agglomeration,which is higher than the national average level.Secondly,the relatively more agglomerated producer industries lie in transport,storage,post,finance,leasing and business services.Finally,Shanghai,Nanjing and Hangzhou show greater agglomeration of producer services.

An empirical study on the relations between E-service quality and online shopping satisfaction of college students

Abstract: Based on a review of relevant literature at home and abroad,this study constructed a conceptual model for the online shopping customer satisfaction research.It chose college students as the subject and taobao mall as the survey website,and obtained research data with questionnaires.The study used Partial Least Squares modeling method of the structural equation model to verify multiple assumption relations put forward in the structure model.The results show: E-service quality has significantly positive effect on customer trust,perceptive value,customer satisfaction and customer loyalty;Customer trust and perceptive value are the transition links in the process of E-service quality influencing customer satisfaction and customer loyalty;Customer satisfaction is the most direct factor on customer loyalty.

Multi-dimension Optimization and Criterions of Collaborative Spectrum Sensing

Abstract: Taking into account the constrained cognitive user transmission power,the random distribution of relative positions between two cognitive users and the licensed user in the three-dimensional space,the context effects of geographical information,the RF environment and the environmental background,a model is proposed capable of evaluating the performance of collaborative spectrum sensing under the wireless environment.The feasibility of the model is well explained and supported by extensive simulations.The study results indicate that the collaborative spectrum sensing performance is affected by multiple factors including path loss exponent,fading factor,the relative position changes between users and user transmission power.Optimization can be achieved under certain conditions,maybe with great improvement in some cases.

On Stability Conditions of Congestion Control in High-speed Long-latency Networks

Abstract: The stability conditions of congestion control in high-speed long-latency networks in FAST TCP are studied in this paperA continuous time fluid flow model of the link buffer is considered to create a linear control system related to FAST TCPLinearing the fluid flow model and window control model,with Laplace transform version of them,a negative feedback system with open loop transfer function is presentedThen,in the case of the link is shared by many FAST flows,a sufficient condition on asymptotical stability of FAST TCP congestion window related to the parameteris obtainedPacket level NS2 simulations are used to verify the theoretical results

Progress in Organic Field-Effect Transistors and Its Applications

Abstract: Organic field-effect transistors(OFETs)are the transistors using organic semiconductive materials as the active layersIn comparison with conventional inorganic transistors,OFETs have attracted extensive attention due to their potential applications in large area and flexible opto-electronic devices,such as organic light emitting diodes(OLEDs),organic photo detectors(OPDs),organic solar cells,pressure sensors,organic memory elements,flexible flat panel displays,electronic papers,and so onThe device structure and operating principle of OFET are briefly introduced in the beginning of the paperThe development of organic materials,fabrication technologies and potential applications of OFETs in the past few years are reviewed comprehensivelyFinally,the development of OFETs in the future is prospected

Multi-Constrained QoS Routing Algorithm Based on Quantum Genetic Algorithm

Abstract: This paper presents a approach to the QoS routing problem on quantum genetic algorithm.The network resource consumption function and the distribution of network loads are considered as the objective function,besides network width and delay.Quantum bit(qubit) has been used for encoding the problem.The quantum chromosomes are updated by the quantum gate operation in order to get the path that will take less resource,balance the network loads,use the network resource reasonably and reduce network congestion.

A Improvement for Chaotic Image Cryptosystem Based on Skew Tent Map

Abstract: A new chaotic cryptosystem based on skew tent map was proposed by making use of a mixed chaotic sequences,which are produced by combining the dynamic encryption parameter and a simple single-dimensional chaotic map.The simulation result shows the proposed cryptosystem inherited the advantages of the original one and it shows that the proposed algorithm greatly enhances the security of image encryption without image distortion comparing with the traditional skew tent map chaotic encryption algorithm.

An Improved Dynamic Spectrum Sharing Scheme Based on Game Theory

Abstract: Aimed at the problem of spectrum sharing among a primary user and multiple secondary users based on Game Theory,an improved utility function of the second user is proposed.Not only the different spectrum requirements as a result of the different service among second users,but also the influence of the shared spectrum on the spectrum demand have been taken into consideration in the spectrum demand parameter of the function.The more spectrum the second user shared,the demand for spectrum will be relatively less severe.The demand degree in the utility function is a function of spectrum sharing.It is a variable which can more accurately describes the cognitive radio system features in meeting the needs of users of the spectrum,and reflect better the spectrum allocation of the equity.This paper also analyzed the impact of the spectrum demand function on the spectrum sharing algorithm based on game theory.The theoretical analysis and simulation demonstrate the validity of the proposed algorithm and the existence of Nash equilibrium.

Performance Analysis of 802.11b Based on Improved Markov Model

Abstract: The theoretical maximum data rate of wireless LAN based on IEEE 802.11b is up to 11Mbit/s.However,in practical applications,we find that the actual user data rate is much lower than that.By use of two-dimension Markovian analysis on the framing efficiency and the DCF efficiency,we provide an extensive throughput performance evaluation of the basic access and the RTS/CTS access mechanisms of IEEE 802.11b.The accuracy of the proposed new model is validated by comparing it with simulation results,with respect to saturation throughput.Based on this,this paper gives a compartmentalization of the RTS/CTS threshold,with which the system throughput and average packet delay could be improved.

A Game-theoretic Approach to Spectrum Sharing in Cognitive Radio Networks

Abstract: A game-theory based static spectrum sharing model is proposed for cognitive radio,which supports multiple secondary users competitively sharing one primary user system's idle spectrumThe primary user system's benefit function based on Self-discipline mechanism is establishedReasonable price is determined considering both primary user's spectrum demands and no wasting of the remaining free spectrumDynamic spectrum sharing algorithm is introduced,in which the secondary users can continuously adjust their strategies to reach Nash equilibrium and maximize their benefits

Wireless Heterogeneous Network Convergence Scheme Based on Mode Awareness

Abstract: Based on overall IP core network,this paper describes an architecture of heterogeneous network convergence,which integrates different networks,such as cellular network,WLAN,UWB,Bluetooth and Internet.The hardware and software schemes of network-related mode-aware module,Multi-Mode Wireless Access Server(MMWAS),routing and QoS guarantee are presented.Finally,based on the architecture,it introduces a communication platform including network devices,multi-mode terminals and MMWAS.Experimental testing shows that SMS,voice and file communications can be realized between terminals with different wireless communication modes.

An empirical research on stock market and monetary policy in China:the analysis based on VAR model

Abstract: The empirical analysis has been done on the relationship between the Chinese stock market index and the monetary policy using the vector auto-regressive(VAR) analysis method.Research has found that stock price index has a significant impact on interest rates and money supply M0 and M1,however,the effect of monetary policy on stock index is very limited.China still doesn't have the ability to adjust stock price index by using monetary policy at present,but we can cultivate the ability of influencing the stock price with monetary policy,and pay close attention to the stock price fluctuation.

Signal Filtering EEG Data in Matlab Simulation Design Analysis

Abstract: The pretreatment of collected EEG signals is quite important in processing EEG signals.By adopting the knowledge of signal analysis and processing,we simulate the impact of different wavelet basis functions on the decomposition and reconstruction of EEG signals based on matlab platform.Then we decompose and reconstruct EEG in a 5-layer multi-scale way by using db5 wavelet basis function,and simulate the vibration of the signals.By comparing the simulation results of denoising under different thresholds with that under lowpass filters,we analyze and study the results of specific frequency bands.The research shows that in the process of denoising EEG signals,wavelet analysis can extract EEG micro signals effectively,and thus has important value on the practical use of EEG in broader fields.

Research of Region Based Stereo Matching Algorithms

Abstract: In the paper,region based stereo matching algorithms are developed for extracting depth information from two color stereo image pair.It shows that global error energy minimization by smoothing functions method is more reliable but more time consuming,and that line growing method is more convenient for the sequential computing architectures because of promising higher speed.Filtering unreliable disparity estimation can increase reliability of disparity map.Finally,a comparison of the algorithms is presented.

A Novel Construction of Quantum Stabilizer Codes

Abstract: Quantum error correcting code is an efficient tool to correct and prevent quantum error.It is the basic for quantum computation and quantum communications.Taking advantage of the cyclic difference set,we present a novel approach to construct quantum stabilizer codes in this paper.It's not difficult to construct the well-known quantum code by our approach.By analyzing the property of the quantum code constructed by this method,such as ,as well as ,we find all the constructed codes are the quantum Single-bound codes,and we can use this method to encode different length information.Besides,the code rate is high.

A Recursive Algorithm for Modular Exponentiation Algorithm

Abstract: Modular exponentiation algorithm is widely used in public key encryption technology.This paper analyzes the existing modular power algorithm and proposes an implementation method for modular exponentiation based on a recursive algorithm.