- EI核心期刊
The design and implementation of defect testing system (DTS)-a testing system based on defect model was presented. It first classifies, analyzes and summarizes the main software defect models and gives a description method to describe the proposed models based on state machine. A unified testing framework was proposed to test the proposed models. In the unified testing framework, interval computation was proposed to reduce the false positive and provide necessary information for defect testing. The experiment shows that, compared with other tools, DTS can test the code more efficiently but produce less false.
Image retrieval in discrete cosine transform (DCT) compressed domain is proposed by introducing the spatial information and the energy of DCT blocks into the feature extraction. According to the distribution of DCT coefficients, the spatial distribution map is constructed and spatial distribution feature is given. Meanwhile, the effect of the energy of DCT blocks on retrieval precision is considered, an energy distribution feature is presented as another feature. The energy distribution of DCT blocks and the spatial distribution of DCT coefficients are taken into account for avoiding the retrieval mistaken and lose. The experiment results show that the new method has a good performance both in retrieval efficiency and effectiveness.
An information-theoretical interpretation of necessary conditions for secure network coding is presented. Including Shannon’s secrecy system as a special example, a lower bound of secret key information rate of secure network coding for a special multicast network is proposed, which holds for nonlinear network coding linear network coding as well.
How to protect the security of secret keys is an important problem of security. A proactive and publicly verifiable server-assisted secret sharing scheme is proposed. The secret in this scheme is publicly verifiably shared between a user and a group of servers, and the secret reconstruction is activated by the user. Another important trait of the scheme is that it has two-side initial proactive property, that is, the shares of the user and servers will be renewed in each time period and the renewal operation is initiated by the cooperation of the user and servers. This will make the scheme very fair and make it difficult for an adversary to compromise the secret.
The relationship existed in the nonlinear coefficient, the geometry parameters and wavelength is derived when considering photonic crystal fibers of total internal reflection type with triangle cladding by calculating the effective refractive index. The curves showing the relation are given. The relations of the nonlinear coefficient are discussed. Theoretical support for photonic crystal fibers in nonlinear research and application is provided.
To solve the selfishness problems caused in energy saving in Mobile Ad hoc NETwork (MANET), an incentive model is studied by introducing the principal-agent theory into the relationship between the source nodes and the participating nodes. The income and outcome of both sides can reach the Nash equilibrium by using the simple game theory. This model changes the cooperation plan designing into the optimal incentive contract designing, with which the source nodes will theoretically take the lowest cost for making the participating nodes work hard. Therefore the selfishness can restrained to a certain degree.
A multiple-trellis-coded-modulation (MTCM) symbol sequence design method for Rician channels as well as its corresponding decoding algorithm, state-symbol sequence decoding, are proposed. The complexity of maximum likelihood detection algorithm, such as general Viterbi algorithm for MTCM, is very high. Based on the design characters of MTCM, an applicable decoding algorithm is presented: first, the state sequence of MTCM is determined, and then the symbol sequence between two successive determined states is decoded. With this algorithm, it is not necessary for the decoding output to wait for until the whole symbol sequence is received, that will lead to an acceptable short decoding delay. For MTCM symbol sequences suffering from wireless fading, the decoding performance can approximate that of the maximum likelihood decoding algorithm by changing one parameter. Theoretic analysis and simulation can prove the proposed.
An active queue management (AQM) algorithm based on queuing delay is proposed. The algorithm aims to achieve higher throughput, lower queuing delay, short queue size, lower loss rate and better fairness; it has characteristics of simple implementation relative to other AQM algorithms. The key idea is to monitor congestion by queuing delay instead of packet dropping in Drop Tail gateways, and set the estimated value of round trip time (RTT) as the single threshold of congestion detection. If the queuing delay of a packet exceeds the threshold, the packet will be marked according to the mechanism of explicit congestion notification (ECN) to notify transmission control protocol (TCP) source to take measure to respond congestion. Simulation on ns-2 shows that the algorithm is with desired performance.
Initial gap increasing (IGI) is an effective algorithm for available bandwidth measurement. However, its accuracy will be deteriorated when there exists a few bursts in network traffic. In such cases, the intervals of probing packets are largely affected by the factors of queue size and surges of network load, which are overlooked by IGI. Accordingly, an algorithm is proposed. It approximates the available bandwidth with reference to an iteratively estimated queue size. Furthermore, it is also contained the error caused by the surges of network load. Thus, the approach will result in an improved and consistent performance. In addition, as to the bandwidth of the bottleneck link, instead of being measured alone as is done in IGI, it is gauged as the invariant mode of packets intervals during the measurement process. Therefore, the proposed algorithm will improve the efficiency of IGI.
A finite element full-scale entity model for 6-UPS parallel kinematic machine (PKM) and its accessories are established based on substructure method .It is further confirmed through a modality study and a finite element modality analysis. The displacement harmonic response curves of the nodes on the virtual x, y and z axes of PKM are obtained through a harmonic response analysis on PKM performed with a finite element method. Based on these resultant curves, a law of harmony response on the PKM is investigated. The resultant parameters clearly demonstrate a harmony response characteristic of PKM and its response scope magnitude, which can provide an important theoretical basis for the optimal structure design of PKM.
To ensure the validity of software project schedule estimation and improve the success rate of software projects, combined with some actual project cases, a quantitative and systemic method of software project schedule validation was studied. A schedule verifying guide was established by modeling. The regression result was found not obvious by validation. It indicates that this guide could only be used to verify, but not used to arrange the schedule. The usage of the schedule verifying guide was studied combined with some actual project cases. In addition to the validity validation of single project, the whole schedule estimation method validity of the organization was studied by modeling based on the collection and statistics of historical data. The actual data indicates that the model can improve and ensure the validity of the schedule estimation result and the estimation method, and can obtain continuous improvements.
A new random access protocol assisted by fixed relay stations under the transparent mode in multi-hop relay networks over the time division multiple access was presented. In the protocol, collision problems are resolved when many users attempted to send access requests simultaneously. The key idea is to obtain benefits of multi-antenna systems by cooperative retransmissions among relay stations and users when considering protocol layers. The receivers will recover the collision packets successfully, while channel resources will not be wasted without extra scheduling overhead and without additional antenna hardware. Simulations show that the network performance would be improved compared to those of previous literatures.
A low complexity bit and power allocation algorithm is proposed to minimize the total transmit power under transmission rate and bit error ratio constraints in multiple-input multiple-output systems with inaccurate channel state information. With monotony, the non-convex optimization problem is simplified to a convex one. Then the optimal solution is derived in the nonnegative real field. And the final bit and power allocation is obtained after adjusting the real solution into discrete bit numbers. The simulations show that the proposed algorithm has the same power efficiency with the optimal robust greedy algorithm while its computation complexity is much lower.
Taking the dynamic topology characteristic of mobile sensor networks into account, a probability distribution-based time synchronization protocol (PDTS) is proposed. In PDTS, synchronized nodes send out synchronization messages with self-adaptive probability according to their current states to synchronize other nodes. By this means, the sufficiency of synchronization messages is guaranteed and the redundancy in the networks is reduced remarkably. Additionally, there is no need to establish network hierarchy and merely one communication is required to synchronize a pair of nodes in PDTS. Due to the before-mentioned advantages, PDTS can be applied to mobile sensor networks effectively.
The mechanism forecasts the traffic for many cycles through the prediction function of interpolating polynomial produced by the reports of many key cycles, and grants many cycles at a time. Not only can it save bandwidth, but also reduce the data packet delay. A knowledge support system（KSS）is set up. The key cycle number and predictive cycle number is optimized by the differential coefficient (DC）optimizing method, the adaptive factors of the non-lineal prediction function are optimized by the neural network. Simulation verifies its merits very well.
Focusing on the case that existing orthogonal space-time block codes (O-STBC) have not achieve full rate, a type of novel transmit matrix for quasi-orthogonal space-time block code (Q-O-STBC) is proposed based on Tarokh`s design criterion of space-time code. A full diversity space-time constellation-rotating (A-STCR) block code is obtained through appropriate modulation constellation rotating. It is introduced the encoding structure and decoding principle of A-STCR code. The validity of this coding is proved. The performance merit of A-STCR block codes is confirmed by Matlab simulations and compared with other space-time block codes.
A location assisted energy optimal routing (LAEOR) algorithm was proposed for deep space backbone network based on the theoretical analysis for the energy routing over deep space backbone network. LAEOR not only takes into consideration of the equipment constraints and energy restriction, but also makes use of the node location information. It introduces a threshold value to design and optimize routing strategy. Simulations demonstrate that this algorithm can decrease and balance energy consumption, and reduce both transmit time and the packet dropped rate.
To calculate the life cycle cost of mobile networks. A concept of mobile network element life cycle cost is put forward. Thereafter, a cost evaluation method of network module is employed to determine components of network element life cycle cost with a new cost calculation model of mobile network element. Then life cycle cost of mobile network of certain area is achieved. Applying these models to certain real wireless network environment, the life cycle costs of most types of equipment at a base station is worked out and analyzed. And many conclusions with practical significance are concluded.
A result merging method based on text / rank analysis and Group Decision Making activity is proposed. Utilizing text-based information obtained from search results, a definition of query-match grade is presented, and an approach on text normalization for meta search is described. The relevant scores of relevant documents are normalized by incorporating text analysis measure with existing rank-based method. When estimating scores of non-relevant documents, different assumptions are discussed, and an improved shadow document method is proposed as well. So a merging method based on Group Decision Making activity is adopted to sort the search results. Four different search engines are tested in the practical web environment. The experimental results show that this method is more effective than other three merging methods: Round-robin, CombSum and CombMNZ.
A cooperative theory based convergence mechanism for heterogeneous wireless networks is presented, and a theoretical optimization model for the cooperative relay node selection is proposed, in which the utility function is involved. The proposed model can determine the optimal relay nodes from the global viewpoint under the conditions of equivalent transmission rates between the first and second hops. In order to simply the computation complex, a sub-optimal relay selection algorithm is presented. Simulation shows that the proposed cooperative relay node selection algorithm can improve the overall performances dramatically.
A subband method with general-parameter-filter banks (GPFB) for wideband array adaptive beamforming is proposed. The received wideband signal is divided into multiple subband signals by the analytical filter bank. Each of the subband signals is then fed into its corresponding subband array processor, of which the outputs are combined by the synthesis filter and recovered into wideband signal. As subbands are running at down-sampling rate and in parallel manner, the loads of computational processing can be effectively alleviated, and the efficiency of array improved. The simulation results show that the convergence speed of weights and deep suppression of the interference are enhanced with the subband method except for a little worse mean square error (MSE) of the output signal. In addition, compared to the subband method using quadrature-mirror-filter (QMF) banks with the same number of channels, this method possesses a superior ability of interference suppression, and it is also capable of reducing the MSE significantly.
The existed password management methods are not secured enough but inconvenient to change each account password. Based on the key technologies of trusted computing, such as key management, security storage and authorized access control, a new password management scheme is proposed. This scheme helps users manage multiple accounts by turning a single memorized password into a different password for each account. The implementation of the scheme is discussed and compared its strength and usability to those of related approaches. Unlike previous approaches, our scheme is both highly resistant to brute force attacks and convenient to execute a password change for each password. It also can prevent phishing.
A new OFDM frame synchronization algorithm is proposed, two identical sequences are included in the synchronized symbol. The diversity of multipath channel is considered in the new object function so that performance of the proposed algorithm will be not influenced by frequency selectivity of channel. Analytical form of missing detection and false detection probability of the proposed algorithm are given. Numerical simulations validate the analytical results: when the receiver’s SNR is 20dB and missing detection probability is 10-10, the false detection probability tends to zero in our algorithm.
A new divergent block-dependent watermarking method is proposed to authenticate the integrity of the image and localize the image tampers. The divergent block-dependent watermarking method is prior to conventional block-dependent watermarking method with improved security and localization accuracy. the original image is firstly divided into non-overlapping blocks, and then the block watermark is independently generated. the characteristics of the image block is mapped into the initial value of a chaotic map, and the block watermark is generated by iterating the chaotic map from the initial value. In the meantime, a block related to several blocks is made by sorting chaotic sequences to form the divergent block-dependent structure. Then the block watermark is inserted into the least significant bits of the related blocks to obtain the watermarked image. The experiments show that the proposed divergent block-dependent watermarking method can resist to the counterfeiting attack, and localize the tampers correctly and accurately.
Spectrum sharing mechanism among distributed cognitive radio nodes is studied. Based on interference temperature model, the spectrum sharing problem is changed into a nonlinear power optimization problem. A distributed power allocation algorithm is proposed to maximize the network utility under multiple interference temperature limits. The presented algorithm solves the nonlinear optimization using primal-dual method with limited information exchange among cognitive radio nodes. Analysis and simulation show that the network utility is enhanced dramatically using the proposed algorithm.
In order to differentiate real-time traffic from best-effort traffic in multi-hop Ad hoc networks, a new differentiated medium access control (MAC) scheme, differentiated label-swiching distributed coordination function (DL-DCF) is proposed. Forwarding information is obtained by label in MAC layer and is piggybacked in Request-acknowledgement (RACK) frame. Clear to send (CTS) handshake of the downstream node was triggered by RACK that made the data frame transmit consecutively across multi-hop. The probability of performing consecutive transmission is calculated according to traffic type and occupied ratio of the higher layer data buffer in MAC layer that can realize service differentiation and avoid data buffer overload. Simulation shows that compared with 802.11 DCF , under high loads, DL-DCF can remarkably decrease the end-to-end delay and normalized control overhead, increase the throughput, and realize service differentiation.
A new coded cooperation methodology based on LDPC codes is presented. The diversity gain and bit error performance is deduced theoretically, and the performance of the new methodology is simulated for some typical wireless communication scenarios.
Based on MMSE-SIC (Minimum Mean Square Error-Successive Interference Cancel) receiver, a fairness adjustable grouping pairing scheduling algorithm is proposed in VMIMO (Virtual Multiple Input Multiple Output) systems. This algorithm makes use of the proportional fair (PF) scheduling to select the primary user, and applies grouping as well as adjustable metric criterion designed to select the paring user. The comparison simulation results show that the scheme exploits the multiuser diversity gain, meantime, improves the service quality of the users at the cell edge. Furthermore, the tradeoff between the system throughput and user fairness can be achieved by varying an adjusting coefficient in the algorithm.
A radio resource allocation and scheduling algorithm for multiuser MIMO/OFDMA
systems is proposed to efficiently support the diverse quality of service (QoS)
requirements of heterogeneous services. Firstly, the priority is given by taking
into account QoS requirements, channel conditions, throughput, and queue state
information. Then, the user selection on the same subcarrier based on the channe
l correlation, power allocation, and bit loading are implemented. The algorithm
aims to maximize the system throughput by fully exploiting multiuser diversity
gain in space, time, and frequency domain while guaranteeing the delay requirem
ent for realtime services and minimum rate requirement for nonrealtime ser
vices. The simulation results show that the proposed algorithm improves the syst
em performance in terms of the system throughput, packet dropping probability, a
nd the ratio of unsatisfied minimum data rate of nonrealtime services.