- EI核心期刊
Joint optimization assisted by the time and power allocation of spectrum sensing and data transmission is a new paradigm for cooperative wireless network. By taking full account of different cognitive user`s transmit power over channels and their mis-detection/false alarm probability of licensed users, a new joint optimization scheme with amplify-and-forward mode based on system energyefficiency maximization is proposed for sensing-transmission time allocation and power control. It is proven that the global optimal solution of this scheme can be tractable by sequential optimization method. Simulation shows that the energy efficiency of the relay-assisted transmission using our proposed scheme outperforms the non-relay transmission, and the system performance will be improved with sensing-transmission ratio decrement and signal-to-noise ratio of the licensed user increment.
In practical communication systems, the prefect channel state information (CSI) is hardly obtained by transmitters owing to imperfect estimate or feedback of the channel coefficients. In order to mitigate the adverse impact of imperfect CSI on bit error rate， based on the constraint of zero-forcing channel error, a robust minimum interference leakage algorithm is proposed for multi-cell multi-user multi-input multi-output system. The formulation of system sum rate is derived under new objective constraint with channel error. Simulations show that if the imperfect CSI is known, the performance on bit error rate and average sum rate of the user in cell edge can be improved by the proposed robust algorithm, especially at the high signal noise rate values.
In linguistic scene descriptions, how to automatically build the description of spatial relations between objects is an important task of linguistic scene description. A method of fuzzy directional relations that rely on the computation of quadtree histogram is introduced. And a system is constructed able to choose the suitable directional relations terms, the distance relations terms, a humanlike linguistic description of the spatial relations is generated as then. Excellent results from several synthetic and real image examples show that the applicability of this approach works well.
In order to improve customer satisfaction an optimal replica selection algorithm is proposed for content delivery network based on local request characteristic. In the case of limited storage capacity of surrogates, according to request characteristic of local surrogates, the local request characteristic is used as heuristic information to select replica. Simulations show that the proposed method can obtain higher request of coverage compared with several other algorithms.
In order to realize the channel estimation for multiple input multiple output (MIMO) orthogonal frequency division multiplexing (OFDM) systems in fast timevarying environments, an algorithm of channel estimation based on the polynomial basis expansion model (BEM) and the parallel interference cancellation algorithm with iterative decision feedback is proposed. With help of the parallel interference cancellation algorithm, the accurate channel estimation and the removed intercarrier interference are achieved based on the feature of which the BEM can transfer the channel estimation to a small amount of parameter estimation. Simulations shows that the signal to noise ratio gain reaches approximately 5 dB when the bit error rate is about 10-3, compared to the traditional estimation algorithms of least square and linear minimum mean square error. Furthermore, the algorithm doesnt need to know the channel characteristic of time domain. The phase shift orthogonal pilot sequence is adopted, and the computational complexity of channel estimation is reduced thereafter.
Clustering is an important energy-saving mechanism in sensor networks. Most existing clustering algorithms ignore the reverse-data-forwarding phenomenon suffering from limitations in energy efficiency. A directional clustering algorithm (DCA) is proposed for data aggregation in sensor networks. DCA ensures that node’s data is always transmitted towards to sink but not away from it during the whole forwarding route, so as to avoid energy waste caused by the reverse-data-forwarding phenomenon. Simulations show that DCA possesses higher energy efficiency and can prolong the lifetime of sensor networks.
To solve high computational complexity problems in selection algorithm of precoding codebook for time division duplex(TDD) long term evolution (LTE), two traditional selection algorithms, i.e. maximizing mutual information and minimizing error rate are studied. And based on singular value decomposition of channel matrix, a new precoding matrix index selection algorithm is presented. The algorithm reduces the computational complexity enormously by using the idea of channel average and maximizes effective power of transmitted signals. Simulations show that the computation of the algorithm presented is 26%~33% of those above these traditional algorithms but performance loss is very small.
To improve the spectrum efficiency and reliability of relay system, two transmission protocols combined network coding and hybrid automatic repeat request (HARQ) are proposed, the HARQ protocol is based on determination network coding(DNC-HARQ) ,and the HARQ protocol is based on opportunity network coding(ONC-HARQ). The performance of outage probability of system and the spectrum efficiency of these two protocols are analyzed. Furthermore, an optimize problem is elicited to maximize the spectrum efficiency of system. As the optimize problem is nonconvex, an iterative algorithm is designed to solve it. Simulations confirm that the two protocols presented is with higher reliability than the protocol with network coding, more spectrum efficiency of systems is obtained than the protocol with HARQ. Simulation shows the iterative algorithm designed can effectively maximize the spectrum efficiency of system.
To solve the difficulty of quantitative estimate in system structural analysis, a new system structural analysis algorithm is presented. Based on fuzzy relation matrix of integrant-to-attribute and attribute-to-attribute, the contribution degree and relation degree are acquired. Moreover, the extent of effect and relation is analyzed. Structural analysis on communication network system shows that the proposed algorithm can provide different contribution of integrant-to-system and effective relationship between communication network system and sub-networks and running indexes, realizing the goal of analyzing qualitative problem by quantitative method.
A method for codebook design is proposed to support beamforming mechanism in a 60 GHz millimeter-wave wireless communication environment. The codebook is designed with only phase shifting but not any amplitude adjustment to meet the low power consumption requirement. N complex numbers of the same phase interval and amplitude of 1 are designed to generate a codebook matrix. Simulation shows that, compared with the codebook in IEEE802.15.3c, a similar mainlobe patterns can be obtained in the proposed high-order codebook, while the sidelobe level appears lower, and the antenna gain at the direction of main response axis is higher. Meanwhile the interference decreases obviously with an improved total channel capacity in the practice environment, and the codebook is also robust to the beam shifting deriving from phase errors.
With the aid of secure scalar product protocol, a privacy-preserving vector equivalency determination protocol is proposed. Based on privacypreserving vector equivalency determination protocol, a privacy-preserving dynamic point distance determination protocol based on semihonest and an n-dimension privacy -preserving dynamic point distance determination protocol are presented. The correctness and security are also analyzed. At the end, a general solution to a privacypreserving dynamic point distance determination is discussed.
To solve the problems of weak computing and storage capacity of nodes while providing aeronautical communication services, a multiple metrics aware ant routing algorithm for high altitude platform station (HAPS) networks is presented. The proposed routing algorithm works to find the shortest path from source to destination in HAPS networks, and satisfies the requirements of services. Many simulations have been performed. It is shown that the proposed algorithm is capable of reducing call blocking probability and better link utilization achieving traffic load balance.
A method of similarity computation for executable objects homology detection based on structural signature was proposed. At first step, considering the characteristic of clone code, a signature of basic code block was designed. On the basis of the hierarchical structure of filefunctionbasic code block, similarity measurement of basic code block was built based on its basic properties and structural information. At second step, to evaluate the homology between original and object files, the similarity was calculated through constructing function weight by means of function structural information. Aiming at the most common clone patterns, some experiments were conducted between the proposed method, the method without considering weight and some mainstream similarity detection tools. Comparative results demonstrate that the proposed method can measure the similarity of two executable objects more accurately than other methods.
A two-stage game model between duopoly platforms is established based on twosided markets theory and game theory, the scale equilibrium, the price equilibrium and the profit equilibrium of duopoly platforms are analyzed. The coopetition strategies is discussed between the first movers platform and the late entrants platform, It is concluded that the more prosperous of the video market turns, the more advantages the late entrants platform gets. Therefore, the late entrants platform should adhere to imitative innovation strategy and scalization development using lower pricing and targeting low-end customers, while the first movers platform should insist on innovative and differentiation strategies through targeting high-end customers and providing high quality service.
Dynamic model for a front-wheel drive bicycle robot free from regulator weight was presented and balanced controller for rectilinear motion was investigated. By analyzing the turning radius of the robot, kinetic energy were derived with the driving angle velocities of front-wheel and front bar, and then the dynamic model was established based on Lagrange formulation. The balance controller for rectilinear motion was constructed by linearizing front-wheel and front-bar input angle and considering the under-actuated frame rolling angle as the internal dynamics. Simulation result shows the effectiveness of the proposed controller. Physical experiment result further verifies that the proposed controller can realize rectilinear motion with reasonable driving torque.
Dynamic threshold attenuation characteristics and firing periodicity of spiking cortical model (SCM) are analyzed. It is concluded that the negative time matrix of this model conforms to Weber-Fechner-law. It processes lighter areas coarsely and darker areas accurately. And then a new image enhancement algorithm based on SCM is presented. The hue saturation intensity (HSI) color space that satisfies with human visual system is chosen. Hue component is kept unchanged; but saturation component is changed by power stretch while luminance component is processed by SCM. Experiment shows that this algorithm is feasible, and the enhancement effect is obvious.
Based on cyclic autocorrelation, an algorithm of blind recognition of orthogonal frequency division multiplexing (OFDM) signals in the multi-path fading channels is proposed. Derivation of the cyclic autocorrelation characteristics for OFDM signals with a cyclic prefix in the multipath fading channels is done, it verifies that the cyclic autocorrelation of OFDM signals will appear peaks at a certain time delay, while single-carrier signals is without peaks. So, OFDM signals can be recognized from single-carrier signals in multipath fading channels. The new recognition algorithm eliminates the preprocessing tasks, such as symbol timing estimation, carrier and waveform recovery, and signal to noise ratio estimation as well. Also, the proposed algorithm is with low computational complexity and high recognition rate. Simulation verifies that the proposed recognition algorithm is correct and efficient.
The nature of scanning algorithm for remote electrical tilting antenna control system based on antenna interface standard group (AISG) protocol is investigated and determined, it is made up for the inadequate AISG protocol. The algorithm is optimized when considering the influence in which the antenna device number is exerted on the scanning time during the process of increasing mask bits. Meanwhile, a tree collision splitting algorithm based on the B1 maskbit increasing is proposed. In accordance with the characteristics of the scanning algorithm for AISG protocol, this algorithm reasonably arranges the numbers and steps of increasing mask bits, and solves contradictions between idle cycle quantities and collision cycle quantities, so that the effective identification of antenna line devices is achieved, and the scanning time is reduced as well.
Solving the problem of algorithm in modern communication channel, a new method of quasiNewton least mean square adaptive equalization is proposed. By replacing inverse matrix calculation of LMS/Newton correlation function by symmetric positive-definite Hesse matrix, the proposed algorithm reduces the influence of traditional autocorrelation matrix estimate error for the algorithm convergence. Equalization simulation shows that higher convergence speed and lower bit error rate are obtained.
A new algorithm called heterogeneous nodes based identifier space partition is put forward to get balanced distributed Hash table space partition in heterogeneous environment and balanced storage load of each peer. To alleviate the route and reply load unbalance due to skewed objects queries, another algorithm called virtual ring based loading transfer is utilized to transfer route and reply load from heavy loading nodes to light loading ones. Experiments running on OverlayWeaver show that the above two algorithms can be synthetically applied to achieve good load balance among heterogeneous peers in peertopeer (P2P) systems.
A Dirichlet process mixture model which can make use of information of known topics efficiently is proposed for topic tracking. Prior knowledge of known topics is combined in Gibbs sampling for model inference, and similarities between new story and known topics can be gained. Experiments show that the model, without a large scale of indomain data, can improve the performance of topic tracking significantly even with a few ontopic stories.
A timing synchronization scheme using constant amplitude zero auto correlation sequence as training sequence is proposed for multiple input multiple output (MIMO) orthogonal frequency division multiplexing (OFDM) systems in multipath fading channels. A modified cyclic delay sequence is used in each antenna. The cross correlation between local sequence and received signal and the symmetrical correlation of received signal are calculated respectively. The two correlations are then combined together to detect the time offset of each transmitter element. Furthermore, multipath backward searching technique is employed at the receivers to cope with the situation where the first path is not the strongest. The threshold is variable at different signal to noise ratios in order to guarantee acceptable synchronization performance. Performance of this proposed scheme is verified and compared by simulations in both additive white gaussian noise and multipath fading channels.
To deals with the popular mechanism, groupbuying auction, a secure protocol without auctioneer in semihonest model was presented. Taking advantage of homomorphic lemma of distributed ElGamal scheme, the bidders can compute the result by themselves without disclosing own messages. The correctness and security are also analyzed. Then the protocol is extended and a basic problem in secure multiparty computation, secure zone distribution problem is proposed. At last, the application is discussed.
In order to solve the problem of resource location in content delivery networks, a new algorithm for selecting optimal service nodes is proposed. The algorithm provides a good solution for both service performance and overhead. First, by taking advantage of the network coordinates approach, the computation and communication overhead of different service nodes are computed depending on network and server status change. Then, based on the hierarchical clustering algorithm, all the service nodes are reorganized in the coordinate space. Finally, the user requests are adaptively redirected to the service nodes with minimum overhead without sacrificing the service performance. Experiments verified that the precision of resource location is improved and the overhead is reduced with acceptable time performance by the new algorithm.
A compressive-domain filtering method based on oblique projector is proposed. The oblique projector used for filtering is constructed without exactly knowing the sparse positions of the target signals. It can eliminate the unwanted signal’s contribution to the measurements in compressive domain without first resorting to a full-scale reconstruction; meanwhile, it can preserve the information of target signal for reconstruction. Simulations show that this method can filter the unwanted signal effectively in compressive domain. Furthermore, and it is better than that of the filtering method based on orthogonal projector.
The factors influencing the achievable secrecy rate of analog network coding over twoway relay wiretap channel are analyzed. A model of wiretap channel with an eavesdropper is established, and the achievable secrecy rate is derived. By numerical simulation and analysis of special scenes, it is concluded that: when eavesdropper locates in some area, the achievable secrecy rate may be 0; by increasing the distance and angle between eavesdropper and the node, or by optimizing the power allocation, the achievable secrecy rate can be enhanced.
For a two-user layered Alamouti-coded multiple-input multiple-output （MIMO）system, the complexity of the maximum-likelihood decoding is high while full diversity is achieved. Using low-complexity partial interference cancellation (PIC) group decoding and its combination with successive interference cancellation (PIC-SIC) group decoding, one space-time coding scheme is proposed. Full diversity gain is proved to be achieved with the PIC-SIC group decoding by the analytical derivation of the symbol error rate. Simulation confirms the derivation and its better biterror rate performance over layered with the PIC and PIC-SIC group decoding.
In order to infer more accurate network topologies, a 3-tuple packets measurement method is proposed. It could measure the time to live (TTL) and queue delay of the shared links between destination nodes. A topology inference algorithm is designed based on queue delay and TTL. The measurement and topology inference algorithm are analyzed. Simulated by network simulator 2, the results indicates that the topologies inferred based on 3-tuple packets are more accurate than the topologies inferred by other algorithms.
A new program for BitTorrent network detector is designed. The detector adopts a similar crawler technique to measure and explore the actual networks, solving the problems that the logs of tracker servers are hard to acquire in related works. The measurement data discovers an interesting phenomenon that even in a small network area, there are more than one peer downloading the file. It also finds out the time distribution character of the number of peers, the distribution of peers’ geographical locations and the rank about different types of BitTorrent client. All of the mentioned above provide data support and basis for peer-to-peer networks.