- EI核心期刊
It is reported the discovery of indoor spatial spectral holes in television white space based on the real measurement in the urban area of Beijing China. The report shows more than 18% urban indoor area of Beijing has spatial frequency holes, and proves that the observed spatial spectral holes can be used by cognitive radio users forming a low power short distance small cell without causing interference to nearby television receivers. The research on spatial spectral hole provides extensive spectrum access opportunities for employing emerging dynamic spectrum access technology to meet the massive demands for future wireless services.
In correlation between five common exchangeable image file (EXIF) parameters and image statistical noise for digital photos is built. The authenticity of the image is judged by this correlation. The statistical noise and five EXIF features from the training image are extracted, following the correlation established. Make use of the correlation to estimate EXIF information of testing images which will be compared with the real values. The comparison will determine the authenticity of testing images. Experiments on both tampered images and real images indicate the effectiveness of the proposed method.
An improved unscented Kalman filter (UKF) algorithm is called dual adaptive UKF. It could not only stabilize model noise by the method of resizing noise, but also inhibit observation interference by monitoring the trace of the adaptive matrix based on innovation characteristics and amending the adaptive matrix in time. Simulation results show that the dual adaptive UKF algorithm is very sensitive to model noise and observation interference. And it has strong robustness which could stabilize the model noise quickly.
A construction algorithm with optimized shift value of circulant permutation matrix for quasi-cyclic low-density parity check (QC-LDPC) codes was presented. Through analyzing the cycle relationships between the basis matrix and the check matrix, the number of short cycles was reduced and the girth of check matrix in QC-LDPC code was maximized. The simulation results show that the girth of QC-LDPC codes constructed by proposed algorithm could at least extend to 6 or 8, while the number of short cycles was decreased significantly. In the end, the bit error rate performance was effectively improved.
A novel linear precoding scheme is introduced for CoMP multiuser multiple-input multiple-output downlink Systems based on joint-processing mode. By carrying out a channel extension approach and applying a matrix decomposition method with low complexity, this precoding algorithm can effectively suppress noise and multiuser interference simultaneously and can be applied to the system configuration with arbitrary number of antennas for each user. Additionally, a suboptimal power allocation algorithm is introduced under the total power constraint of all the base stations. Complexity analysis and performance simulation demonstrate that the proposed precoding algorithm can not only degrade computational complexity of the central control station in general, but also can achieve enhanced system performance.
An optimal decision threshold for energy detection based broadband is presented. Upon the analysis and research on whether the certain performance spectrum detection algorithm can achieve the target requirements, it is concluded that the threshold settings depended on the impact of false alarm probability and the probability of missing detection. Through mathematical derivation, the closed-formsolution of the minimum-cost decision threshold is deduced, and then the selection criterion of the optimal detection threshold is proposed in which the detection threshold selection is upon the length of observation time in domain. Simulation results verify the correctness of the article.
A new algorithm based on adaptive Metropolis algorithm for recognizing phase-amplitude-modulated signals in multipath fading channel was proposed. The method uses the full history of the chain to tune covariance of the proposal distribution adaptively. This adaptation strategy can approach an approximation of the target distribution, which can avoid the problem of selecting proposal distribution in traditional Metropolis algorithm and improve the sampling efficiency effectively. Simulation show that the proposed method has high accuracy which verifies the validity of the algorithm.
In wireless sensor networks, the localization of nodes is vital and promising to a wide scope of applications. Based on characteristics of DV-hop, an improved algorithm for the typical range-free localization was proposed. The main principle of the improved algorithm is to introduce hop coefficient to revision hops, which could address the problem that DV-hop localization of wireless sensor networks produces error of hops stimulation. Then hop coefficient could bring down the error of per hop average distances and be sure precisely unknown nodes and reference node. Simulation demonstrates that the results of improved algorithm are more accurate, and the errors down to about 10%.
The enhanced distributed channel access mechanism is adopted by IEEE 802.11p medium access control (MAC) layer,which is used for communication in vehicular ad hoc networks. An analysis model based on two-dimensional Markov chain, colligated the saturated and unsaturated condition, was proposed to study the quality of service (QoS) requirements of different applications, taking different contention windows, internal collision and frozen mechanism into account. On this basis, the performance metrics, such as normalized throughput and drop rate, were studied. The results show that the value of normalized throughput decreases with the increasing number of vehicles, but increases with the packet generation increasing. The drop rate increases with the increasing number of vehicles as well. However, the higher priority categories have larger value of normalized throughput, which verifies that the IEEE 802.11p can provide QoS for different messages.
To propose the method of evaluating and improving the real-time performance of Ethernet for plant automation (EPA) industrial ethernet, the real-time performance of EPA nonperiodic data transmission was studied. By analyzing EPA communication scheduling mechanism and its time delay generating regularity, nonperiodic messages were categorized as some modes according to their entering-queue time. The scheduling characteristics, communication delay of each mode and their interacting relations were studied, and the models of real-time performance of EPA system were established. Analyses and an instance verification, 3 key factors limiting the real-time performance of EPA nonperiodic messages transmission were found: ① priority inversion during the messages transmission; ② communication overhead of deterministic scheduling mechanism; ③ the processing method of nonperiodic messages that enter queue after the device gives up its competition right of transferring nonperiodic messages.
In order to realize a low complexity tamed spread spectrum system based on long codes, a novel tamed spread spectrum system based on quasi-regular low density parity check (QR-LDPC) code was proposed. This system employs a new LDPC code called quasi-regular LDPC which is constructed using degree distribution optimization method. A synchronization algorithm exploits check constraint was proposed, and then the complexity of synchronization for long code tamed spread spectrum was reduced. Simulations show that the proposed tamed spread spectrum is an effective and feasible scheme adopt long codes, both spreading gain and coding gain could be achieved. The bit error rate performance of proposed system is also better than traditional direct sequence spread spectrum system since the spread factor is equal.
In distributed storage systems, an efficient multi-node regenerating program with healthy nodes collaboration was proposed to make a trand-off between the repair bandwidth and reliability for multi-node regeneration. Specific multi-node repair process and the construction of regenerating code were introduced, and the max-flow-min-cut of information flow graph was used to illustrate its lower bound. It is shown that the program ensures the repair bandwidth maintain to minimum during the repair process, and makes the repair time synchronization and repair process easier. The most important is less that of desired transmission channels when ensuring reliability of regeneration.
For purpose of energy saving in wireless access networks, a regional autonomous energy saving mechanism, consisted of energy saving phase division based on traffic load, was proposed. The method of stepwise choosing closed base stations was adopted when energy saving period begins as well as choosing closed base stations to resume before end of saving period. Improved energy consumption model and evaluation method are also proposed to estimate the efficiency of proposed energy saving mechanism. Simulation using situation similar to existing network was given to verify the proposed energy saving mechanism and reveal that effective energy saving of 17.5% could be achieved, while guaranteeing regional coverage and service quality. It is shown much better in terms of block rate control and user received signal strength than that of the other energy saving mechanisms.
Both circuit and packet traffics are supported in present communication networks. Forwarding those two traffics over the same switching fabric favors the sharing of network devices and the integration of networks. A combined circuit/packet switching fabric and dispatching scheme was proposed based on Clos-network switches. Dedicated paths were reserved for circuit traffic by dispatching scheme in the switching fabric. Meanwhile the remaining bandwidth is used to provide packet traffic with best-effort forwarding. Simulation shows that quality of service requirement of circuit traffic is satisfied, and a high throughput is achieved for packet traffic.
A sum-space interference alignment (IA) scheme for downlink coordinated multi-point joint transmission system was proposed. The interference from the collaborative base stations to each user equipment was aligned onto a space of specific dimension, which was accomplished by joint precoding for multiple coordinated cells, and then the interference space was compressed. The interference can be eliminated by interference suppression matrix which was designed at the each user equipment side. In the performance analysis, the outer bound of the system total achievable degree of freedom (DoF) was derived, and the relationship between the achievable DoF and the dimensions of the aligned interference space was given. Numerical analysis shows that the sum-space IA scheme can obtain more DoF and data rate compared with traditional orthogonal schemes and traditional IA schemes.
A K-means cluster evaluation technique using bi-dimensional entropy components was proposed. The attack dataset on the basis of network entropy was preprocessed, a two-dimensional plane was mapped. The output of preprocess as the input of clustering was utilized. And a relation between the attack dataset and the effect category on the basis of K-means algorithm was established, thus an explicit division of attack effect set was achieved. Efficient evaluation was given. Experiment shows that the method can process attack dataset with high efficiency, as well as provide a visualized evaluation result by form of evaluation cluster diagram.
In order to improve the cell edge users' performance, a quality of service-aware inter-cell interference coordination method using a graph-based approach was proposed. Considering that the users may require a different quantity of physical resource blocks, the problem of assigning the radio resource to the users was modeled as the graph colouring. To solve the problem, an improve sequential colouring method for introducing resource allocation type was applied. Simulation shows that the proposed scheme could effectively increase the cell edge users' performance and the cell average throughout.
A multi-agent enforcement learning method based on user clustering as well as a variable learning rate was proposed for solving the problem of channel allocation and power control within multi cognitive radio users. Firstly, a hierarchy processing method was used to separate channel selection and power control. The channel allocation was implemented by fast optimal search combined with user-number balance. Secondly, stochastic game framework was adopted to model the multiuser power control issue. In subsequent multi-agent enforcement learning, K-means user clustering method was employed to reduce the user number in game and single user's environment complexity, and a variable learning rate scheme for Q learning and policy learning was proposed to promote the convergence of multiuser learning. Simulation shows that the method can make multiuser's power status and global reward converging effectively, moreover the whole performance can reach sub-optimal.
The stabilities of finite-state machine (FSM) seriously affects the normal operation of satellite borne system, and is important for control logic. Single event upset caused by space radiation environment often leads to FSM instabilities. Current fault-tolerant methods can correct single error. However, the correction of double errors for those high reliability systems is also necessary. A new design of field program gate array(FPGA) based on (16,8) quasi-cyclic code was presented. With advantages of high reliability, uncomplicated hardware and short delay. The design can correct single or double errors, and detect triple errors as well.
An improved back propagation neural network algorithm based on dichotomy was proposed for assessment of electric power communication network risk. The dichotomy was used to adjust the learning rate in the training process. It helps to change the learning rate automatically to the direction of optimization. Simulation shows that the improved algorithm's performance is optimized, such as convergence rate、error accuracy and training time.
Applicable for digital cameras equipped with single-sensor covered with a color filter array (CFA), a new joint color demosaicking and zooming algorithm was proposed, the interpolation directions from the raw CFA image was estimated based on the local gradients. The estimated directions were exploited consistently in the demosaicking and zooming. Experiments show that the proposed algorithm is of good performance compared with the several recently published algorithms.
A new space-time coded transmission scheme is proposed for X channel with two antennas at each user. Each user employs rate-2 space-time block code (R2-STBC). The multi-user interference is eliminated by proper design of transmission method of R2-STBC and pre-coding at the transmitter. The feedback amount is reduced and the transmission efficiency is higher as well, compared with the existing scheme for the same scene. Simulation results demonstrate that the reliability of the proposed scheme significantly outperforms the existing scheme.
An admission selection algorithm utilizing delayed rejection model for heterogeneous wireless networks (AS-DRM) was proposed. The concept of buffer was introduced, elastic services were allowed to wait in the buffer until they could be serviced successively. Five-dimensional Markov queuing model was built by queuing theory for resource occupied in heterogeneous networks. The system performance was analyzed and evaluated. Simulation shows that this algorithm evidently decreases the blocking rate for elastic services, and the system performance is improved.
An uplink resource allocation strategy for cognitive single-carrier frequency division multiplexing access (SC-FDMA) system was proposed. After defining the new marginal utility function, the available resource blocks and transmission power were jointly allocated. Simulation shows that although its achieved throughput is slightly lower than other algorithms, on one hand, the proposed algorithm can achieve the continuous resource block allocation, protect the minimal transmission rate for each user very well, and more, avoid the interference to primary network and guarantee the communication of primary network.
Compared to traditional random mobility models, a social-based mobility model was proposed, aiming to generate synthetic traces to capture the statistical properties detected from real traces. the driving force of complicated social behaviors from an evolutionary point of view was explored. A genetic algorithm-based mobility model(GAMM) was presented. Using Gain/Cost Ratio as the metric of trace's fitness, complicated movement patterns were emerged from generations of evolutions. Explorer's model and transportation model were presented to show the expandability of GAMM. The influence of social-based mobility model on MANETs network protocols are also investigated by simulation.