For issue of Virtual Machine (VM) migration in cloud computing environment when it comes to meeting the demands of load balancing, auto scaling, green energy-saving, etc. This paper design a scheduling algorithm that is cloud computing infrastructures oriented and workload prediction based. By organically integrating the active control technology based on workload prediction and the passive control technology based on status information of actual system, as well as with the exponential smoothing prediction model to predict the workload condition in future time, a VM consolidation algorithm is put forward which takes the maximum future workload as first in the VM selection stage and compares the resource demand queues in the VM placement stage. The simulation results show that the algorithm uses the prediction-based resource integration to reduce the number of servers and virtual machine migrations as well as service level agreement violations, effectively increasing the overall resource utilization of data center as the core of the cloud infrastructure.

To solve the issues of malicious modification and privacy disclosing of aggregation value during the data aggregation phase of Wireless Sensor Networks (WSNs), a novel private data aggregation with data integrity checking algorithm is proposed, which is called Integrity-ChecKing Private Data Aggregation (ICKPDA) for WSNs. This algorithm first hides the sample data through combining it with secret seeds; then it enhances the privacy efficiency by data slicing and assembling technique. Finally, it converts the data to two-tuples to provide integrity checking by the relevance of binary data. The simulation results show that ICKPDA can effectively preserve data privacy, check the data integrity, and get accurate data aggregation results while having less communication and computation overhead.

Recommender algorithms based on Singular Value Decomposition (SVD) performs better in prediction accuracy and stability, while the speed of error descent slows down gradually and the iteration number is huge when applying Stochastic Gradient Descent (SGD) to SVD , which greatly limits its usage in actual projects. To address this problem, the difference matrix of the rating matrix is used to represent the local structures and to optimize SVD as a new target function. Experiments on MovieLens and Netflix dataset show that the improved SVD algorithm can obtain better performances with fewer iterations compared to classic SVD. When compared with other state-of-the-art recommender algorithms, the proposed algorithm is only second to SVD++, however within far less training time than SVD++.

This paper investigates the propagation pattern of malicious data packets in Mobile Wireless Sensor Networks (MWSN) through employing the stochastic cellular automata. The corresponding network model, channel allocation mechanism, signal interference model and malicious packet propagation model are designed based on information communication characteristics of MWSN. The new cellular space, neighborhood and states as well as state transition rules are defined. Then a stochastic cellular automaton model for modeling the propagation process of malicious packets in MWSN is proposed to study the spatial-temporal behavior characteristics of malicious data packet propagation in the uncertain environment. The results show that the proposed model can efficiently predict the propagation trend and the spatial distribution state of malicious data packets in MWSN, and overcome the disadvantages that no one of the Markov Random Field, the differential equation and the difference equation can describe the spatial distribution state of malicious data packet propagation in MWSN.

Considering the problem of 3D void area in Wireless Sensor Network (WSN) routing, 3D routing algorithm based on Cell Space Routing (3D-CSR) is presented. Analyzed the possible situations on the area which is unable to route by the greedy algorithm, it is the proposed algorithm that can determine the boundary of void area in 3D cell space model, and accomplish the routing process with Cell Routing Mechanism (CRM). Meanwhile, adaptive election mechanism is used in every single cell to keep the cell leader node active for routing and others sleep, so that the cost of network energy is balanced. Simulation results show the correctness and effectiveness of 3D cell space model and 3D-CSR. Compared with 3D Grid Position-based Routing (3D-GPR) and 3D Coordinate Face (3D-CFace), 3D-CSR has higher message delivery rate and node survival rate.

As the next Internet research hotspot, Content Centric Networks (CCN) provide content dissemination through distributed content cache machines and change the traditional host-centric network architecture. Cache replacement policy is one of the core issues in CCN which will directly restrict the data dissemination performance of the network. Considering the content popularity distribution characteristic, this paper proposes a cache probability replacement policy based on content popularity and derives the closed-form expressions which calculate the average request miss probability of different levels in a L-level tandem network scenario. This policy is designed to select a data chunk replacement position in cache according to its popularity and try to balance the proportion of different popularity contents stored in the network nodes. Numerical results show that this policy is suitable for the network application with higher requests’ aggregation. In this scenario, it can obviously promote the dissemination performance of low contents compared with the traditional Least Recently Used (LRU) policy.

Future wireless network must consider the problem that utilizing the same frequency band with other wireless network because of lacking of frequency spectrum. Recent research results find that interference alignment can make the all users in network utilize the same frequency efficiently. In order to utilize the same frequency band with other interference alignment network and have no influence on it, an iterative algorithm for interference alignment between two networks is proposed. This algorithm based on eliminating the interference in new network can make all users communicate well. Numerical results show that new users will communicate as no interference and have no influence on other users when transmitters and receivers have enough antennas. If the transmitters or the receivers lack of antennas, the interference alignment network will degrade performance. Lack of antennas at transmitter will have more harm.

In the smart grid, risk of power communication has become the problem of power system needs to solve, the service route is an important factor of communication risk. However, existing routing algorithm does not take into account the special nature of power communication services, it could cause unbalanced distribution of service link, and further lead to increased risk of the overall network. Therefore, a risk balance based service route allocation mechanism is proposed to reduce the risk of communication circuit and the overall risk of network. First, service important degree, link availability and other factors are integrated to establish a business-oriented risk balance model. Secondly, a business-oriented route risk balance mechanism is proposed based on this model. Simulation based on the province electric power communication network topology data, compared with the Availability-Aware Routing (AAR) and Load Balancing (LB) algorithms, the proposed mechanism can improve the service link risk balance degree more than 30%, and overall network risk is reduced by at least 6%, which has a very strong practical significance.

Energy-efficient resource allocation issue in uplink OFDMA systems is investigated. Firstly, the uplink OFDMA system is modeled as a M/G/1 queuing system with the consideration of the packet retransmission and the serve rate of the system is determined by both the resource allocation result and the current channel condition. With the constraint of average delay of traffic, an optimization model for energy-efficient resource allocation is formulated based on the traffic statistical information. Since the optimization problem is NP-hard, a heuristic algorithm is proposed which allocates the resource by two steps: the number of suchannels determines, and allocates the subchannels to each user. Simulation results show that, the proposed algorithm can improve the system energy efficiency effectively while guaranteeing the average delay requirement of the traffic.

Common-frequency full-duplex systems send and receive signals simultaneously in the same frequency. Therefore, self-interference will exist in the local receiver and interference cancellation is necessary. Digital interference cancellation is applied after Analog-to-Digital Converter (ADC) sampling. ADC bit number, Interference-to-Signal Ratio (ISR), and quantizing judgment scheme affect both interference cancellation and Symbol Error Ratio (SER). In this paper, the essential relationship among ADC bit number, ISR, and SER of Quadrature Amplitude Modulation (QAM) is analyzed; A closed-form expression of SER is developed; And simulation results are presented to support the SER expression. The simulation results show that: when ISR decreases or ADC bit number increases, SER of this kind of system decreases in large scale, and fluctuates in small scale; some fluctuations of SER are dramatic, which can approach to the SER of the system without quantization error.

Traditional physical layer security methods, such as artificial noise, commonly require that the sender has more antennas than eavesdroppers, which is not practical in the multicasting system. To solve this problem, a multicarrier-based physical layer security scheme for the multicast systems is proposed, which takes the advantage of both space and frequency domain to form a relatively virtual multi-antennas system. Through subcarrier allocation the existence of null space in every subcarrier channel matrix is guaranteed, so that the artificial noise can be added to ensure a secure communication. Based on the Kuhn-Tucher condition, an optimal power allocation method is also proposed to maximize the security rates under the restriction of total power. Simulation results indicate that this scheme is unlimited by the amount of the sender antennas, and it improves the security rates by  than the average power allocation method.

According to the characteristics of two-way amplify-and-forward relay channel, this paper extracts secret key by exploiting the reciprocity of wireless multipath channel in time division system and the commutative law of convolution operation. So, the secret key can be generated with the channel estimation to overcome the drawback of the key pre-distribution mechanism. The recovery algorithm based on compressive sensing theory is used for channel estimation because of the sparsity of the wireless multipath channel. This paper proposed two new schemes based on the relative-amplitude and the relative-phase for secret key generation and compared them to the scheme based on the relative-delay. Simulation results show that all three schemes can achieve high key generation consistency probability, and present eavesdropping by the third party to ensure the security of the physical layer communication.

A new timing synchronous algorithm for Filtered MultiTone (FMT) system is proposed based on investigation for FMT character. The method achieved good performance with shorter train sequence for getting synchronous symbols by linking the correlation of the FMT symbols to the strong autocorrelation of PN sequence. Otherwise, a simple effective estimation for carrier frequency offset is introduced which is a integer multiple of sub-carrier spacing. The simulation results show that these two algorithms perform well in the additive Gaussian noise channel, and obtain a good estimate performance in the Doppler shift of the Rayleigh channel with diversity techniques.

A combination code estimation method based on the fuzzy unitary matrix is proposed for the blind descrambling and dispreading of the TD-SCDMA burst signal. The method take the short spreading code and the scrambling code of the fixed length together as the district’s spreading code of TD-SCDMA system, also named the combination code. Firstly, according to the relevance of midamble code’s height to acquire the midamble code and to estimate the residual of carrier frequency. And then according to the midamble code's relative position to intercept the data domain signal with the length of the scrambling code. Finally, the blind estimation of the combination code can be achieved by using the second-order moments of the correlation function, the singular value decomposition method and the unitary matrix method of the eliminating fuzzy relations. Simulation results show that the algorithm has a good effect, and it is unnecessary to estimate the combination code’s cycle before the estimation of combination code.

Strong sequences in cryptography should not only have a high linear complexity but the linear complexity should also be stable. Such sequences are called excellent sequence in this paper. Cat swarm optimization is an intelligent global optimization search algorithm. Based on given reasonable conditions, it can automatically generate excellent results hoped. By designing rational and effective fitness function, and appropriate parameters, cat swarm optimization is selected to find excellent sequence, and binary excellent sequence is got where the period N is 32, 64, 128, 256, 512, 1024 etc. and the error k is less than or equal to N/4. After combining a large number of experimental data, the conjecture is given that k-error linear complexity of excellent binary sequence with period N meets the law that  LC_k(S)≤N-2k+1.

The function projective synchronization and parameter identification between the fractional-order chaotic system and hyper-chaotic system with uncertain parameters are researched, in which the fractional-order Chen chaotic system and the new fractional-order hyper-chaotic system are as examples. First, based on the fractional theory of stability and nonlinear dynamic theory, the nonlinear controller and parameter identification rules are designed. And by the nonlinear controller the function projective synchronization between the fractional-order 3D chaotic system and 4D hyper-chaotic system with uncertain parameters are realized. At the same time, the uncertain parameters are identified. Second, based on the fractional-order theory of stability the synchronization are proved strictly in mathematics. Finally, by the Predictor-Corrector scheme numerical simulation, the validity of the presented method is verified.

As it is resistant to quantum attacks, lattice theory is widely applied to various cryptosystem. Currently, lattice-based Password Authenticated Key Exchange (PAKE) protocols are all designed for two-party circumstance, so it can not meet the application requirements for large-scale communications system. In this paper, a three party PAKE protocol is proposed based on lattice from the two party PAKE framework of Gorce-Katz, and its security is proved in the standard model. At the same time, through the realization of explicit mutual authentication between user and server, the protocol can be used to resist undetectable online dictionary attacks. The new protocol is considered to be the first three party PAKE protocol based on lattices. Compared with general construction, the new protocol reduces the number of communication round, and can avoid the weakness that cryptosystem based on the integer factorization problem or discrete logarithm problem is vulnerable to quantum attacks.

To reduce the computational complexity and memory access bandwidth of motion estimation in H.264 video code, an algorithm to decide the search range for motion estimation is proposed. Firstly, the Search Range (SR) is adaptively selected based on the accuracy of Predicted Motion Vector (PMV), which is measured by using the Sum of Absolute Difference (SAD) of neighboring blocks. Then, the motion activity of coding region is also considered for search direction selection in the search area. The extensive experimental results show that the proposed algorithm is able to reduce 91% and 18% of the computation time on average compared to full search and UMHexagon algorithm, respectively, with almost the same bit rate and PSNR.

This paper proposes a novel algorithm for determination of decoding order for the Gaussian Interference Channel (GIC) based on max-min fairness. The algorithm is developed using rate-splitting technique to improve the sum-rate of GIC. Based on the proposed algorithm, an expression of the number of rate-splitting is derived to make the sum-rate approach the maximum value in two-user GIC. Numerical results show that, in the two-user symmetric GIC, the proposed decoding order algorithm improves significantly the performance over the greedy algorithm on the two-user symmetrical GIC within a wide range of channel coefficients, and the sum-rate of the proposed scheme can approach the inner bound of Sason as increasing the interference factor on two-user symmetric GIC.

Most of previous blind parameter estimation methods of Frequency Hopping (FH) signals with a single channel do not adapt to overlapped signals. Moreover, the single- and multiple-channel-based methods use batch processing techniques almost, so they are not able to estimate FH signals in real time. To get real-time tracking reliably, a novel method for single- and multiple-channel-based frequency estimating and hop timing detecting for FH signals is proposed to track the frequency of multiple frequency-hopping signals based on Sparse Bayesian Learning (SBL). Firstly, overlapped FH signals sparse representation model is established. Then, the Sparse Bayesian Learning is used to estimate hopping frequencies and detect the frequency hops once they happen. Numerical examples are carried out to demonstrate the effectiveness of the proposed method.

In view of the lack of theoretical studies on vector tracking algorithm, the linear model and the tracking error are comprehensively studied. To improve the dynamic performance of vector tracking loop, the terms of the partial differential of the pseudorange-rate to position, it is usually ignored in the existing papers, are revised in the modified linear measurement matrix. A new loop feedback is put forward to simplify the operational process. The frequency jitter of vector tracking loop caused by thermal noise is deduced, showing the tracking threshold is lower than scalar tracking loop. And the tracking threshold of vector tracking loop becomes lower with increasing in the number of observation satellites in the same situation. The dynamic stress threshold is deduced as well, it is proved that vector tracking loop can be utilized to track weak signals in high dynamics. The Monte Carlo simulation validats the effectiveness of the theoretic analysis.

To reduce the high Peak-to-Average Power Ratio (PAPR) of modulated signals based on Prolate Spheroidal Wave Function (PSWF) pulse, a novel PAPR reduction method for PSWF pulse based on Givens rotation is proposed. The method represents firstly orthogonal PSWF pulses with weighted eigenvectors, and then uses Givens matrix G(i,j,Θ) to rotate all the two-dimensional planes of coordinate (i,j) with any angle. The rotation matrix which most greatly reduces the PAPR of pulse set is searched and the original PSWF pulse set is transformed with the matrix. To reduce the realizaition complexity, a flipping iteration scheme based on grouping is proposed. It is theoretically proved that the transform can maintenance orthogonality of pulse set and high energy concentration in the given frequency domain. Simulation results show that the method can effectively reduce the PAPR of pulse set and preserve the transmission efficiency, the Power Spectral Density (PSD) and Bit Error Ratio (BER) performance of modulated signals.

The Time Adaptive Support Vector Machine (TA-SVM) algorithm exhibits good performance for nonstaionary datasets. However, insufficient information from adjacent subclassifier may lower the reliability of the obtained classification model and weakens its usefulness. A novel classifier named Evolving Support Vector Machines (ESVM) is proposed in this study by defining the relationship decaying function of the subclassifier serial. The evolving relationship between all the subclassifiers are considered in ESVM, thus a more smoothing subclassifier serial can be obtained by constraining the weighted variance between all subclassifiers, conforming with the characteristic of drifting concept hidden in the data. The effectiveness of the proposed ESVM is also experimentally verified.

Low-dose Computed Tomography (CT) is widely used in modern medical practice for its advantage on reducing the radiation dose to patients. However, excessive quantum noise is present in low dose X-ray imaging along with the decrease of the radiation dose; thus, there are obvious streak-like artifacts in reconstructed images. For this problem, an adaptive restoration algorithm based on local fuzzy entropy is proposed in this paper. This new algorithm modifies the statistical information based anisotropic filter, distinguishing edges and smooth areas by a local fuzzy entropy. The new diffusion model can effectively control the diffusion degree, thus improve greatly the diffusion rate to achieve the purpose of rapid recovery of the projection data. Simulation results show that higher signal-to-noise ratio reconstructed images can be obtained by the new adaptive diffusion algorithm. In addition, compared with conventional algorithm, the proposed algorithm shortens processing time in projection domain and thereby reduces the hazards of radiation to patients.

A kind of object recognition algorithm based on Affine Histogram of Oriented Gradient (AHOG+SVM) is proposed to solve the poor effect of object recognition algorithm based on HOG (HOG+SVM). In order to have scale invariance, this paper builds multi-scale pyramid gradient images, and then computes HOG feature on them. In order to increase the rotational invariance and shear invariance, this method firstly expands 2D HOG grid to 3D HOG grid, then maps 3D grid to 2D affine grid according to the relationship between the world coordinate and image coordinate. Finally, inverse transformation of HOG feature in affine grid is carried out to remove the influence of affine transformation. The experimental results show that, the proposed method has the ability to solve the low recognition rate because of scale changes, rotation changes and shear changes (3D perspective changes) of object, and its performance is better than HOG+SVM.

Due to the reduction in computational complexity, Fast Factorized Back-Projection (FFBP) has an obvious advantage over conventional BP in improving the processing efficiency of SAR. Based on the FFBP algorithm presented before, a novel FFBP algorithm which can reduce computational complexity further is proposed in this paper. The main characteristics of the novel FFBP are as follows: (1) with the imaging plane set on the  coordinate system instead of polar coordinate system, there is no need to calculate the arcsines of the back-projected angle; (2) meanwhile, two-dimensional polynomial fitting method is introduced to obtain the values of back-projected slant range and angle without calculating point by point. Finally, the processing results of simulated X-band SAR data show the proposed algorithm is accurate and effective in reconstructing well focused images.

By applying the prior knowledge of targets to be detected, enhanced imaging for the targets can be realized and beneficial improvements in image quality and detection performance can be expected. Body Of Revolution (BOR) has the unique feature of aspect scattering invariance, which can be used for the classification of BOR targets. In this paper, Aspect Scattering Entropy (ASE) is introduced to measure the aspect invariance of target scattering. Then the enhanced imaging algorithm based on ASE is proposed to emphasize BOR targets. Afterwards, the effect of fluctuant location and attitude on enhanced imaging is analyzed, and compensation method is integrated with aspect scattering feature extraction. The results of experimental data demonstrate the effectiveness and practicability of the proposed algorithm.

There are two problems for the Adaptive Non-Negative Eigenvalue Decomposition (ANNED). One is that the search solution to the Non-Negative Eigenvalue Decomposition (NNED) takes much time to obtain the volume scattering power by repeatedly calculating eigenvalues. The other is that the remainder covariance matrix of the ANNED may have negative eigenvalues which indicates that the decomposition result is invalid. For these two problems, an improved ANNED is proposed. Firstly, a fast solution to the NNED is derived by calculating the principal minor zeros of the remainder covariance matrix. Since it does not need to repeatedly calculate eigenvalues, the fast solution is faster than the search solution for calculating the volume scattering power. Secondly, the fast solution is also used to adjust the scattering powers of the ANNED to deal with the problem of the negative eigenvalue. The real-POLSAR-data experiment shows that the improved ANNED can markedly enhance the double-bounce scattering power and weaken the volume scattering power for urban areas, and enhance the accuracy of classification.

In Airborne/Stationary Bistatic Synthetic Aperture Radar (A/S-BiSAR), the assumption of translational invariance is no longer available. After the Range Cell Migration Correction (RCMC), the range compressed signal in the same range gate exhibits azimuth-variant Doppler rates which make the phase error difficult to be estimated from the echoes by using the available Phase Gradient Autofocus (PGA) algorithms. To deal with this problem, a modified PGA algorithm is proposed. Comparing with the traditional PGA, the Residual Quadratic Phase (RQP) caused by the azimuth-variant Doppler rates is additionally estimated and compensated. As the influence of the azimuth-variant Doppler rates is greatly reduced, a phase gradient estimator is subsequently applied to accurate phase error retrieval. Furthermore, the range-variance of the phase error is also concerned in the phase error estimation and correction, thus the proposed algorithm is suitable for A/S-BiSAR with wide swath in range. Raw data experiments are presented to verify the advantages of the proposed algorithm.

As an important technique in both civil and military affairs, airborne SAR absolute positioning can absolutely locate the ground targets without the support of ground reference point. In this paper, a new absolute positioning approach based on the sum-different SAR is presented to determine the locations of ground targets without any reference points. In this approach, three-dimensional coordinate information of ground targets is obtained by monopulse processing between difference and sum channels. Moreover, the influence of parameters mismatch on the accuracy of geolocation is analyzed. The new presented method is compared with SAR geolocation techniques based on the R-D model. Simulation results are provided to show effectiveness of the approach.

This paper focuses on the methods to design the transmit beam pattern and waveforms separately, and utilize the waveform diversity to suppress adaptively the interferences and clutters for co-located MIMO radar. Firstly, an improved fast beamforming algorithm is proposed to optimize the above two designs simultaneously. In order to guarantee the full rank of the correlation matrix of the transmitted waveforms, the eigenvalue loading method is applied. Then, an array signal processing framework for MIMO radar which employs the partially waveforms is adopted to recover the diversity of the waveforms by using the matching filter banks. Simulation results show that this kind of MIMO radar will take full advantages of the extra degrees of freedom brought by waveforms diversity to suppress the clutter and interference efficiently. 

For the compensation of the system distortion in the wideband radar working with STRETCH processing, a new distortion compensation method based on time-variant property removal is proposed. The STRETCH processing distortion and its time-variant mechanism is expounded in detail, and a time-variant property removal algorithm is obtained. Joint time-variant property removal and distortion compensation method is established finally. The concrete implementation steps of this compensation method are also presented. Experimental results based on the real data, which prove the effectiveness and efficiency of the presented strategy, demonstrate that the proposed compensation method can compress echo pulse with low computation complexity, simple implementation and sidelobe suppression above 45 dB.

Deployable truss reflector antenna will be applied broadly to Synthetic Aperture Radar (SAR) on satellite in the future, while the systematically theoretical analysis of the truss reflector antenna’s pattern is still not fully developed now. Based on the special reflector shape of deployable truss reflector antenna which is composed of a number of space triangle elements, a method that combines Physical Optics (PO) method and substitution is proposed to calculate the pattern of truss antenna illuminated by arbitrary numbers and locations of feeds in this paper. To solve the pattern, only coordinates of vertexes on the reflector are used. The method is applied to simulate the pattern of a 2.8 m×6 m-aperture deployable truss reflector antenna which is illuminated by rectangular horns line array, and the simulation results of the main pattern parameters coincide with the practical measured values and show a better performance than geometrical optics method’s simulation results. The method shows direct ties between the construction of truss antenna’s reflector and its pattern, and it realizes the electromechanical integration design of deployable truss reflector antenna.

Considering the issue of the analysis and calibration for complete mutual coupling of conformal arrays with nonideal ports fed, the concepts of the “structure mode mutual coupling” and the “antenna mode mutual coupling” are proposed. These two kinds of mutual coupling induce the complete coupling effects. Based on the full wave numerical analysis and microwave net theory, the analysis and effective calibration techniques for conformal arrays’ complete mutual coupling are proposed. Furthermore, the 12-element microstrip cylinder array with nonideal ports fed is simulated, which confirms the validity of the proposed method.

The characteristics of the Electromagnetic Band-Gap (EBG) structure composed of regular hexagonal patches is analyzed in the frequency domain and time domain. Respective influences of the side length of patches, space between patches and the radius of vias on the band-gap and transmission characteristics are investigated from the theory of the equivalent circuit. Mathematical expressions which estimate accurately the upper and lower cutoff frequency and bandwidth of the band gap for different side lengths of patches are obtained and verified. Studies show that space between patches changes the bandwidth, but has no influence on the characteristics of the left part of the band gap. When the radius of vias is reduced, the band-gap shifts left and becomes narrow. Finally, The time-domain characteristic of the signal transmission line using EBG structure as a return path is also analyzed. Experiments show that the smaller the period of structure, the worse the signal quality.

According to the well-performed ionospheric heating experiments at Arecibo in the low latitudes as well as at TromsΦ in the high latitudes, the large-scale modification effects are simulated. The findings are extensively exploited to verify the validation of the proposed model by comparison to the experimental results. Further, a thorough research is carried out on the influences of the background electron density gradient as well as the ratio heating frequency f_HF to the critical frequency  f₀F2 of ionospheric F2 region on heating effects. Conclusions are drawn as follows: A smaller electron density gradient of background ionospheric F region leads to a better ionospheric heating effect; During over-dense heating, the heating effects are enhanced if the ratio of f_HF to f₀F2 increases, which is limited by the resultant elevation of the reflection height. However, there might be a better ratio range with small values of the ratio of  f_HF  to f₀F2. Finally, how to select heating parameters efficiently under adverse conditions is analyzed so as to obtain relatively effective results.

Matrix Factorization (MF) based Collaborative Filtering recommenders proves to be highly accurate and scalable, nonetheless, most models of which have the defect of serial training. It is generally accepted that the models can be more scalable if they are able to training in parallel. In order to attain the goals, a new parallel model based on the Regularized Matrix Factorization (RMF) recommenders, namely Parallel RMF (P-RMF), is proposed. P-RMF works by employing the alternative stochastic gradient decent instead of stochastic gradient decent to train the features, whereby the dependence between user features and item features is removed and then the training process can be parallelized and improved. The experiment results show that this new model is more effective to increase both the scalability and computing speed for solving collaborative filtering problems, compared with the existing similar models.

Spectrum informations in the scope of the mainlobe are the main research object of spectrum analysis methods, in which noise component always decreases the accuracy of spectrum analysis. Firstly, the concept of the Mainlobe Spectrum Distortion (MSD), according to the mathematical definition of SNR, is proposed to describe the confidence level of spectrum analysis results quantitatively. Then, to show the obvious relationship between MSD and the error of spectrum analysis, some phase measurements using energy gravity method are carried out as examples, whose results show that, MSD can trace the error curve of phase measurements more closely than that of SNR. Finally, a MSD self-estimation algorithm is provided in details, which can estimate MSD quickly with high accuracy and good anti-noise performance, without measureing the frequency and initial phase.

The impossible differential property of LBlock block cipher is analyzed. Based on the property of the structure of round function and the technology of key-byte guessing, two impossible differential attacks on 21-round and 22-round reduced LBlock are presented. It is shown that the attack on 21-round requires about 2⁶² chosen plaintexts and 2⁶² 21-round encryptions, and on 22-round requires about 2^62.5 chosen plaintexts and 2^63.5 22-round encryptions. The presented results are the best impossible differential attack on reduced-round LBlock so far.

To solve the problem of time-variant communication links in Wireless Body Area Networks (WBANs), a novel energy efficient and reliable Adaptive Routing (AR) protocol is proposed. Based on the analysis of data propagation model, a computing method for the quality of outer links, which are trained by Hidden Markov Model (HMM) as the observation value in order to find the optimal routing path, is also brought forward. The simulation results indicate that compared to the existed prediction-based routing protocol such as Prediction-based Secure and Reliable routing framework (PSR), etc., the performances such as the network packets loss rate, the network time delay and the nodes’ lifetime are improved significantly in the proposed.