- EI核心期刊
The performance of a low complexity time domain pilot scheme for orthogonal frequency division multiplexing-multiple input multiple output systems is analyzed. The scheme is turned out to be of relatively high estimate accuracy with low calculation complexity, and the estimate performance is unaffected with numbers of transmit antenna. There exists no interference between pilots and data symbols. Compared with least square estimator in frequency domain, the present scheme can obtain gains with specific parameters and simple linear correlation under the same mean square estimation level. Also compared with linear minimum mean square error (LMMSE) estimator, the scheme approximates the optimization of LMMSE's favorable performance in high signal to noise ratio region with knowledge of channel taps length only and no more priori information in the estimate process.
A moving caption detection method based on relevance vector machine (RVM) and the context of moving caption is proposed. Harris corner detector is used to determine caption region of video keyframes, and then the sparse optical flow field is obtained from Horn-Schunck(HS) optical flow algorithm, meanwhile, the motion and static text features is extracted respectively as well. A spatial-temporal context relationship among multiple text frames is described by features cascading. Finally, the relevance vector is learned and a two-class classifier is constructed. Experiments show that the performance of the proposed method is better than the existing four approaches, and supports vector machine-based algorithm.
To maximize the system throughput, a subcarrier and power allocation scheme for multi-service multicast system is proposed. The new scheme works in three steps. Firstly, to ensure the fairness among services, subcarrier and power allocation among services is performed according to the target data rate of each service; Secondly, to maximize the system throughput while ensure the fairness among users with the same service, a subcarrier cooperation scheme and a simplified scheme are derived; Finally, the water-filling method is adopted to get further capacity improvement. Simulation results show that the proposed scheme can significantly improve the system throughput, while ensure the fairness among multicast services and among users with the same service.
An intelligence control algorithm for friction compensation of low-speed servo system is proposed based on self-recurrent wavelet neural networks. There’s of no necessary to predict the system dynamic model parameters,and the high-precision compensation of nonlinear friction is realized by using few neurons and iterations through only position feedback. Lyapunov stability analysis shows the bounded convergence of tracking error and network weights. Also the servo experiments from a robot joint show that the servo positioning accuracy can be greatly improved by introducing the proposed compensation algorithm.
To meet the increasing requirement on the transmission characteristics of optical communication systems, based on the structure of parity-check matrix of Richardson-Urbanke algorithm, a lower-complexity constructing method is proposed for the quasi-cyclic low-density parity-check (QC-LDPC) codes in the optical communication systems, and a QC-LDPC(4288,4020) code with 93.7% higher code-rate is also constructed. Simulation shows that the constructed QC-LDPC(4288,4020) code has net coding gain of about 1.7 dB over the classic RS(255,239) code which has been widely used for optical communication systems and 0.3 dB more than that of SCG-LDPC(3969,3720) code with the same 93.7% higher code-rate, and the NCG of the QC-LDPC(4288,4020) code is about 1.3 dB away from the Shannon limit at bit error rate BER of 10-7; furthermore, the error floor of QC-LDPC(4288,4020) code is lower than that of the randomly constructed LDPC(4288,4020) with the column weight of 3.
In the distributed storage system, the storage node needs to be repaired when its data is damaged. So far, the network coding has been applied in the distributed storage system for greatly reducing the total amount of data transmitted from the undamaged storage nodes called repair bandwidth to repair the data. For network-coding-based distributed storage system, the data repair problem is studied that minimizes the repair bandwidth under the constraint of given rebuilding time. The optimization problem addressing the data repair problem is formulated, and its optimal solution together with the optimal data recovery approach is presented. Simulations show that the proposed approach minimizes repair bandwidth within the given rebuilding time threshold and the repair bandwidth quickly decreases with the increase in the rebuilding time threshold.
According to structural characteristics and availability requirements of cloud storage, a semi-random linear network coding (SRLNC) algorithm is proposed for data redundancy. The data block is split into pieces, encoded by coding matrix composed of unit row vectors and random row vectors on finite fields GF(2s), stored in data nodes dispersedly. In decoding process, the data block can be restored with the strategy of unit row vectors priority. The probability of unique solution is analyzed for SRLNC decoding equations, a new operations per character index is defined to measure algorithm efficiency. Experiments prove that the operating time of the decoding process in the proposed algorithm exponentially reduces with the increase of the node availability. When the availability of each node≥0.8 and redundancy≤3, the encoding operation of the proposed algorithm is faster up to 33%, and decoding operation of the proposed is 5 times faster than that of random linear network coding algorithm, respectively. It is shown that the proposed algorithm is applicable to "write once read many" cloud storage system.
In order to improve the verification efficiency of coordinate rotation digital computer (CORDIC), a new verification model that derives from orthogonal statics theory is proposed. Using this model, we set up orthogonal tables with the test factors, and choose the best optimized test method to verify according to orthogonal tables results. Moreover, the proposed model can randomly generate test vectors and automatically check the precision errors of CORDIC calculation results. Simulation results show that the orthogonal model can greatly cut down verification period and improve working efficiency.
In order to detect virus variants and unknown viruses effectively, inspired by the biological immune system, a computer virus detection model based on artificial immune systems (AIS) is proposed. The dynamic clonal selection algorithm is improved to solve the problem that the self-space is static during the training. The proposed model has enhanced the adaptability of virus detection systems to the continuously changing virus environment. Experiment shows that the proposed model has good adaptabilities, it can effectively detect viruses and has a low false positive rate.
In order to analyze configurations of metamorphic mechanisms, a description method of displacement subgroup for structure transformations of metamorphic mechanisms is proposed. Methods of representation and calculation of displacement subgroup for adjacent and non-adjacent link combinations are presented. And methods of representation and calculation of displacement subgroup matrix for metamorphic transformations are also presented. A spatial 8-SSSS parallel metamorphic mechanism is proposed as well. Its structure transformation is represented and calculated by methods of displacement subgroup and its matrix. The effectiveness and correctness of the proposed method are further proved. The method can transform issues of metamorphic transformations into the representation and calculation of displacement subgroup. Furthermore, the applicability of the description method for structure transformations of metamorphic mechanisms is extended from topological structure to geometric structure.
For the multi-relay two-hop multiple input multiple output system, a kind of relay transmission scheme and interference alignment algorithm is put forward. Firstly, the transmitting beamforming is adopted by the scheme between sources and relays to utilize the spectrum efficiently. Then the relays forward the signals using interference alignment technology. Finally, the interference is eliminated at the ends with a low computation interference eliminate algorithm. Simulations show that the transmission scheme can achieve better system capacity and energy efficiency.
For wireless sensor network of energy hole phenomenon, based on the energy efficient and balance proposed a new uneven cluster algorithm for dynamic routing. Its core is divided networks through the analysis of single-hop and multi-hop energy consumption in same interval network model. Through the calculation of the network energy consumption, obtaining the number of the optimal cluster to building an uneven clustering network to resolve the energy hole through. In the data transmission stage, cluster heads consider energy consumption and residual energy, according to the probability to select the next hop nodes. Simulation results show that the algorithm can effectively prolonging the network life.
To study the use of intelligent ant-colony optimization (ACO) to improve agglomerative hierarchical clustering (AHC) and to attain high-quality cluster results of hierarchy, a hybrid clustering based on ACO and AHC (HCAA) is proposed. The modified AHC and a new objective function are used to generate the dendrogram of clusters and the internal index is utilized to evaluate the solution. The mechanism of pheromone feedback and pheromone volatilization supported by the ACO is employed to control the search of the ant colony in the solution space. The method will accelerate the search, avoiding the results of local optima because of using meta-heuristic optimization. Experiments on several datasets of university of California, Irvine verify the feasibility of this method.
A combination of application layer content identification and deep packet inspection is proposed to realize efficient fine-grained allocation of resources. A better adaption of optical network resources with high-layer application needs is achieved. With network traffic model analysis and packets content-aware, this technology prioritizes different traffic loads. Based on queuing theory, a new optical network traffic model is built and analyzed. Experiments testify the effective optical network resource allocation.
Aimed at the problem of slow convergence speed and low efficiency for moving target contour extraction,a kind of new algorithm based on difference multiplication and multi-level set with Chan-Vese model are put forward,dealing with the problem of sensitive to the initial position and noise of C-V model. The method uses four adjacent images frames for differential multiply to inhibit and filter most of the background edge,and then extracts moving object contour through multi-level set with C-V model. Experiment results show that this method not only overcomes the difficulty of multi-objects recognition,but also accurately extracts moving target contour more quickly.
Machine type communications (MTC), defined as machine to machine communication over cellular mobile network is an integral part of future ubiquitous network and has broad application prospects and market potential. According to the typical application scenarios and features of MTC described in 3GPP TS 22. 368, four service models for MTC are proposed, namely non-mobility model, extended non-mobility model, low mobility model and full mobile model. The traffic flow and signaling overhead estimating method is developed to assess the bearing capacity requirement of operators’ networks under massive concurrent access from MTC devices, which is of prime important for designing efficient network planning and optimization strategies to enhance network performance and user experience.
An new routing algorithm has been proposed for the problem about important power communication service concentrated in a few paths,which based on service risk equalization degree. First,two risk degree models of communication nodes and channel segments have been built,and an equilibrium risk degree model was proposed based on those tow models. Second,the improved Dijkstra algorithm was used to search for K-shortest paths as candidates, and maximum and minimum model was employed for routing decision. Finally, the performance of risk balanced K-shortest path (RBKSP) is compared with two existing algorithms. Simulation verifies the effectiveness of the RBKSP algorithm.
Despite of significant progress on speech recognition, current techniques cannot satisfy the demands of real applications in robot controls, the main reason is that various noises in environments of robot control substantially degrade the performance of speech recognition. A feature extraction method is proposed based on sparse coding. This method makes use of the de-noising merit of sparse coding and extracts features after removing noise in Mel-frequency domain. Such a strategy integrates spare coding into speech feature extraction and can reduce the effect of noise. Experiments in speech recognition tasks show that the feature proposed possesses strong robustness against various noises and improves the performance of speech recognition in noisy environments.
A modified mobile location algorithm based on received signal strength indication (RSSI) is presented. The geometric location of the anchor node is taken into account with the revised positioning algorithm. And then the common line of the three reference nodes is excluded to get the best three reference nodes. RSSI values gives that the same reference node collected takes the median. According to the median of the RSSI values, RSSI values are amended. Finally, the unknown node is located by using the weighted centric algorithm. The algorithm didn’t need additional hardware cost, but easy to implement. It improves the positioning accuracy and suits for the positioning of nodes on wireless sensor networks when their operation abilities and power supplies are restricted.
Binary shuffled frog leaping algorithm (BSFLA) in population diversity is proposed for a cognitive radio allocation model, which is a NP-hard problem. The integer coding manner and the searching method is applied in the shuffled frog leaping algorithm. BSFLA utilizes binary coding manner, explores application of the cultural orientation method to accomplish the searching for the optimal solution. Based on schemata theory, the population diversity is defined using hamming distance in order to judge the premature phenomenon. When the premature phenomenon appears, the population reconstruction is done under the elitist retention conditions to avoid falling into the local optimal point. To demonstrate the effectiveness of the proposed algorithm, An experiment for contrast on 100 various network topologies is done. For three fitness functions, the success rates that the proposed algorithm is superior to particle swarm optimization, genetic algorithm and quantum genetic algorithm are equal to 100%, 75% and 100% respectively.
In the coordinated multi-point systems with limited feedback, a multi-cell coordinated scheduling algorithm is proposed for inter-cell interference coordination and cell-edge user’s throughput enhancement. The concept of delta-channel quality indicator (delta CQI) is used to indicate the interference level of users in one cell when the users in adjacent cells of cooperation region using different precoding vectors. The base stations exchange the information of delta-CQI which is fed back from the users in each cell and calculate the weighted interference factor, then the user scheduling are performed by the multiple cells in the cooperative cell set along with the priority factor given by the proportional fairness scheduling algorithm used in each cell. The proposed algorithm considers the user interference among different cells to improve the data rate of users in cell edge, which gives a good tradeoff between system fairness and spectrum efficiency. Meanwhile, the overhead of the control information exchange among coordinated cells is reduced by the delta-CQI feedback and exchange.
A valid approach to analyze the performance of multi-service cellular systems is proposed. Appropriate models of user’s hybrid traffic and its behavior are established with a reasonable description of its characteristics. Accordingly, a multi-dimension continuous time Markov chains based method is used to derive the relationships between the expected system blocking probability, the traffic outage probability as well as the resource utilization ratio and the users’ attributes. Simulation results verify that the proposed approach, compared with the existing ones, evaluates system performance more accurately.
To study the displacement mechanism of the spatial five-link RCRCR, dual-number matrices are adopt to set up mathematic model and close-form equation. Firstly, 3×3 unitary matrix and Euler formula are introduced to make some dual-number matrices diagonalizable, then, to find relationships, the primary part and dual part are decomposed to derive a 4th degree input-output polynomial equation with a single unknown by symbolic computation. At last, other middle variables are solved. Numerical example confirms that the numbers of analytical solutions for five-link RCRCR mechanism are 4.
To integrate quality, cost and time of project development in multi-outsourcing resources selection, a quality/cost/time trade off model used for optimal scheduling of product planning was presented based on set pair analysis. Modified aho-corasick algorithm based on compressed matrix was used to solve the optimization model, and the optimal product quality/cost/time control concept alternative was obtained, and reasonable outsourcing resources was selected for each component. Finally, three needle lockstitch sewing’s quality/cost/time control process case was evaluated, based on which each component’s outsourcing resources was identified. The result shows the validity of the proposed method.