The Proceedings of the Third International Conference on Communications, Signal Processing, and Systems

Different from the statistical or bound method to build the interference model of cognitive wireless networks, in this paper, we propose an exact mathematical interference model based on the primary channel and the secondary interference channel under Rayleigh and Nakagami fading respectively. Under the rigorous mathematical derivation, the proposed model can cover many parameters such as spatial distribution, the spectrum sensing schemes, the transmission and channel propagation characteristics of nodes, etc. In addition, the analysis result can be extent to a number of applications including spatial density settlement of ST nodes, ST power control, spectrum sensing schemes analysis, irregular geographical shape evaluation of cognitive radio etc. The simulation results have verified our analytical model.

For solving the problem of positioning primary user (PU) in cognitive radio networks (CRNs), we propose a new algorithm which is based on the mobility of secondary users (SUs). The algorithms has the following advantages. First, it needs minimal prior information of PU, so it is more suitable for CRNs; Second, the algorithm utilizes the relative span weighted factor as weight of each SU to overcome the drawback that Centroid Localization (CL) algorithm is too dependent on the connectivity of network; What is more, the algorithm is based on the mobility of SUs, therefore, it can improve the SUs density which fits the localization scenery with little SUs. Simulation results show that the proposed scheme has lower complexity and better robustness compared with traditional algorithms, meanwhile, the accuracy of the algorithm is relatively higher.

Localization of primary user (PU) is the key problem for location-aware spectrum allocation in cognitive radio networks (CRNs). In order to improve spectrum efficiency and reduce the interference from secondary users (SUs) to PU, localization of PU must be rapid and accurate enough. Aiming at the drawback of existing localization algorithm for PU localization, a PU localization algorithm based on node selection is proposed. The algorithm is suitable for PU localization with unknown transmit power and selects nodes based on geometric precision. The simulation results show that the proposed scheme has a good performance in terms of localization accuracy. Meanwhile, the complexity and communication traffic are reduced because of selecting part of the secondary users.

The existing ZigBee security system shows less attention to forward security especially when a node leaves the network. In this paper, we present a key management mechanism in ZigBee hybrid network. This mechanism includes a rekeying method for hierarchical network key and a prompt revoking process of application link key, which can efficiently prevent the damage caused by the key leakage of leaving node. We build a network model to verify the key management mechanism. A modified routing protocol based on LEACH is adopted in the model to gain a cluster topology. Then we use MATLAB to evaluate the performance of this mechanism. Results show that the reduction of consumption is considerable especially in large scale network.

For Cognitive Radio (CR) systems operating within the range of low power incumbent wireless systems, effective and efficient radio resource management (RRM) technique is vital for the spectrum efficiency improvement. By using robust and efficient channel estimation and synchronization techniques, CR based RRM algorithms can be integrated into CR enabling techniques such as Ultra Wideband (UWB) to dynamically allocate the radio resource across the operating frequency band for the optimal spectrum usage. For OFDM based CR-UWB system, we proposed a hybrid RRM (HRRM) algorithm aiming to optimize spectrum efficiency. The HRRM algorithm involves the joint optimization of power and time resource allocation, in which the spectrum sensing window size is dynamically assigned in order to optimize the use of the optimal power distribution algorithm. Our numerical simulation indicates that HRRM algorithm outperforms traditional RRM in terms of spectrum efficiency enhancement and the gain contributed by the HRRM algorithm outperforms the complexity generated.

Security protocol issues is an essential factor in network communication. The research of authenticated key exchange protocol is a hotspot in information security field at present, and the related research theories are quite mature. However with respect to the emerging wireless body area network (WBAN), there is few appropriate security protocol to guarantee the security of this network. This paper proposes an authenticated key exchange protocol for wireless body area network, which support the selective authentication between nodes pertain to the net, simultaneously two pairs of session key being generated efficiently and succinctly in the process of each certification, afterwards the security proof by BAN logic of such protocol is given out. The analysis indicates that the proposed protocol meets the expectative objectives.

Expected frame cancellation is a cross-layer interference cancellation technique proposed to increase the degree of concurrence in WLANs. Using the strong semantic correlation of the control frames exchanged in a typical 802.11 WLAN, a receiving node sometimes is able to know every digits of an incoming interfering frame. Thus it is possible for the receiver to reconstruct the signal of the interfering frame and then linearly cancel it from the receiving streams. In this paper we present our simulation study on the performance influencing factors to this technology, including SINR, frequency and timing differences. It is shown that the signal reconstruction and cancellation method based on the information provided by the preamble can greatly improve the BER performance, especially when the expected interfering frame and the frame to be received are synchronized.

Due to energy limitation of sensor nodes, the conventional security algorithms with high computation complexity are not suitable for wireless sensor networks (WSNs). We propose a compressive sensing-based encryption for WSNs, which provides both signal compression and encryption guarantees, without introducing additional computational cost of a separate encryption protocol. In this paper, we also discuss the information-theoretical and computational secrecy of compressive sensing algorithm. For proposed WSN, if only a fraction of randomizer bits is stored by an eavesdropper, then the probability that he/she cannot obtain any information about the plaintext approaches zero. Simulation results show a trade-off can be made between the sparsity of a random measurement matrix and the number of sensor nodes used to reconstruct the original signal at the fusion center.

Traditionally, jamming to the wireless in-home system is a fatal threat for Smart Grid communications, which impedes the two-way data transmission between electric devices and the smart meter, and thus deteriorates the reliability of the in-home communication of Smart Grid. To enhance Home Area Network (HAN) security for Smart Grid application, in this paper, orthogonal frequency division multiplexing (OFDM)-based power line (PL) system is incorporated into HAN, and a hybrid model of PL and wireless communications are proposed with transmit diversity. For the combinational model with transmit diversity, Alamouti code is employed at the transmitter part. Simulation results validate the feasibility of the presented combinational solution, and furthermore show that the hybrid model could tolerate jamming to the wireless system effectively.

In cognitive radio network, a Grouping spectrum allocation algorithm allocates lots of spectrum to cognitive users, but not to care for the users’ requirement, which leads to unreasonable spectral allocation. For this problem, this paper proposes an improved Grouping algorithm based on users’ requirement. According to the graph theory, the topology of network is divided into many groups by two different grouping ways (frequency Grouping and user Grouping) respectively. In the distributing process, each group reports the distributed information after completing one distribution. Whenever a user’s demand is satisfied, every group is informed to not allocate the spectrum for the user any more, and delete it from the topology in every group at the same time. The improved method can get high spectrum utilization with little sacrifice of time that cost on the spectrum allocation. In addition, the fairness is improved by deleting the node whose demand had already been met. Because of doing this, the other nodes being interference with the node on the same channel can participate into spectrum allocation. The simulation result proves that the utilization of the improved algorithm is much higher than the traditional parallel algorithm, and users’ satisfaction increases a lot.

ZigBee provides a simple and reliable solution for the low cost networks. However, the current routing algorithms cannot fully satisfy the energy consumption issue. In this paper, we propose a minimum physical distance (MPD) broadcasting algorithm the transmission of the monitoring and command packets which are from or to the ZC. The physical depth is introduced to indicate the least hops to the ZC, and the transmission paths are decided based on the neighbour table information. The simulation results show that the MPD could improve the performance of the monitoring and controlling packets transmission by providing shorter paths lower delay.

In the ZigBee networks, the Z-AODV routing algorithm could use the global shortest path for data transmission by flooding the routing quest when necessary. However, this mechanism may lead to a heavy routing overhead. In this paper, we propose a directional broadcasting algorithm in routing discovery (DBRD) to reduce the routing overhead. The network is divided into several continuous “clusters” based on their relative position information. The devices which are not in the clusters covering the shortest path may not take part in the routing request rebroadcasting. The simulation results show that the DBRD could improve the performances of routing discovery; the routing overhead was effectively reduced.

Feature extraction is a crucial step in synthetic aperture radar (SAR) automatic target recognition (ATR). This paper proposes a new feature extraction algorithm named optimal multiple kernel local discriminant embedding (OMKLDE). Based on kernel local discriminant embedding (KLDE), OMKLDE introduces multiple kernel functions and constructs an optimal model. Under this model, we use optimal method to obtain the expected mapping, which enhances interclass separability and maintains intraclass compactness, and more important, solves the parameter selecting problem in kernel trick. Experimental results based on MSTAR database show that the proposed method improves the stability and accuracy of recognition effectively.

To improve the ability of dealing with inaccurate of model and statistic characteristics of noise, as well as the abrupt change of state of cubature Kalman filter (CKF), a new nonlinear filter, called Cubature Kalman Filter based on strong tracking (CKF-ST), is proposed in this paper. Inspired by the idea of strong tracking, a time-variant factor is introduced into the recursive process of cubature Kalman filter such that the filter gain can be updated along with the measured values, thus endowing CKF-ST powerful ability to deal with abrupt changes of state. Meanwhile, such merits of CKF as high accuracy and being easy to implement can be entirely preserved in CKF-ST. Simulation results on one classical examples demonstrate that CKF-ST is overall superior to CKF and other filters involved, especially when target motion changes suddenly.

This paper focuses on the issue of nonlinear data filtering in radar tracking. By through the analysis on nonlinear filters, we find that the accuracy of the extended Kalman filtered (EKF) data image was not ideal for radar tracking data filtering. In our study the unscented Kalman filter (UKF) was introduced to achieve better performance. The evidences show that, while comparing with using EKF, using UKF for radar tracking can get more accurate results because the mean and variance of the nonlinear function can be estimated more accurately with unscented transformation, and the computation complexity is reduced significantly by avoiding to calculate the Jacobian matrix.

Ultra wide band (UWB), can be used to detect human target hidden by walls. Information theoretic algorithms like entropy, relative entropy and mutual information are proven methods that can be applied to data collected by various sensors for detecting target. In this paper, we propose to use entropy and relative entropy to detect target. Breathing motion in human will cause periodic changes in the received signal at a distance where target is located. Relative entropy can detect the change in histogram quickly. After applying weight using relative entropy, we can apply entropy based detection. We conducted study in three different kinds of walls. When target is behind gypsum wall, accurate detection can be achieved by applying this method. We can identify the human behind the brick wall as well. However, human was undetected using this algorithm while hidden by wooden doors.

Target detection is the front-end stage in any automatic recognition system for synthetic aperture radar (SAR) imagery (SAR-ATR). The efficiency of the detection directly impacts the succeeding stages in the SAR-ATR processing chain. This paper proposes a target detection method for SAR images based on visual attention mechanism. In the paper, a new texture feature extracting method using Local Walsh Transform (LWT) is employed and a target-saliency map is computed based on fusing the primary visual feature maps. Experiments are tested on two kinds of images with simple or complex background. The experimental results show that the detection time cost by the proposed algorithm is less than traditional visual attention methods and the detection results are visually more accurate.

Joint sparse representation (JSR) is mostly used in face recognition area. While in this paper, JSR is adopted in the area of SAR automatic target recognition (ATR). In our method, the whole training dictionary is divided into several sub-dictionaries, according to the label of training samples. And classification is based on the minimum representation error criterion. Independent and identically distributed (IID) Gaussian random projection is used to extract feature of SAR images. Experiments are carried out on moving and stationary target acquisition and recognition (MSTAR) public database. Experiments results show that recognition rates can be improved by our method, by combining more useful information and reducing interference information of target.

As the importance of analyzing system resource constrains on SAR automatically target recognition (ATR) has been noticed, more researchers start contributing to this field. In this paper, we focus on discussing the relationship between the performance of target detection and the resolution of SAR image which is regarded as one of the most important parameters of SAR imaging system. It is assumed that the resolution does not need to be good enough in target detection, but the modest reduction of resolution will provide a better trade-off between accuracy and speed. Firstly, an empirical operation is used to modify the resolution of raw images. Actually, the images with different resolutions should be derived by SAR imaging independently, but that’s not the point in this paper. Then the comparison using two methods is described, the optimal detector based on Bayesian expression and the sub-optimal detector, which is known as CFAR detection. The simulated data is used to evaluate the performance of detection, and the correctness of our suspicion is demonstrated in the experiment.

This paper studies genuine and dummy ballistic warheads identification using multiple-input and multiple-output (MIMO) radar. State space model (SSM) is used to describe the kinetic characteristics of the warheads. Identification of the genuine and dummy warheads is accomplished by solving a binary hypothesis testing problem. Sequential detection method is employed. Since sophisticated dummy warhead are made very similar to the genuine one, it is very difficult for conventional sequential detectors to tell them from the genuine ones. We consider to employ locally optimal unknown direction (LOUD) test, which has been proved to have the advantage of distinguishing small differences. Sequential LOUD (SLOUD) test is proposed. It is shown that the performance of the SLOUD test based identification method is superior than the sequential detection method based on the conventional mismatched likelihood ratio (LR) test or the generalized LR (GLR) test.

Scanning radar systems have extensive and significant applications, but theirs angular resolution usually is low because of many constraints. In this paper, a Landweber iteration algorithm of improving the angular resolution is proposed. Firstly, a signal model in the azimuth of the scanning radar system is illustrated. Then the Landweber algorithm is derived in theory. At last, Simulations and real radar data experiments show that the algorithm can effectively improve the angular resolution of the scanning radar system.

For the data transmission in hydrophone linear array, This paper focused on digital baseband transmission system and proposed the cascade channel transmission model, introduced the application of Turbo codes and Turbo channel coding and decoding system briefly, studied and verified the feasibility of improving the channel performance that add Turbo codes through computer-aided simulation.

Terrain following radar (TFR) is a particular airborne equipment that can assist the pilot in his task of staying close to the ground and yet avoiding obstacles. Angular resolution is crucial for TFR, but it is limited by the beam width of antenna. To overcome this problem, angular superresolution technique is adopted in TFR. First, received signal in vertical plane is modeled as a mathematical convolution of the antenna pattern and the targets’ scattering. Then principles of the angular superresolution algorithm and range migration correction method are presented. Simulation result validates that the method can effectively improve the angular accuracy of TFR in vertical plane.

This paper presents a new change detection method based on coherence characteristics between channels in Polarimetric Synthetic Aperture Radar (PolSAR) images to change detection. It aims at solving the problem that information sources of change detection measure are limited to image intensity information usually in PolSAR change detection. In this method, by using channel coherence information extracted from polarimetric covariance matrix, and relying on the entropy character, we obtain the similarity factor of improved information sources to change detection. Finally, set a threshold to distinguish the changed targets. Simulations and experiments are carried out to assess and evaluate the performance of the proposed method. A comparison between the proposed and the other well-known change detection methods is shown, which indicates that the proposed method performs well in change detection.

The paper focuses on the angle measuring technology and analyses the angular accuracy. The total differential method is used to derive the angle measurement precision in the MIMO radar and the phased array radar. A strict formula of the angle measurement error is proposed, while the angular accuracy of the two types of radars are compared under the same SNR condition. The simulation results show that the formula of the angular accuracy is accordant with the simulation and the angular accuracy of MIMO radar is better than the phased array radar.

Different from the traditional static data anomaly detection, the subjects of big data stream anomaly detection has been attracting extensive attention. Through wall human being detection with UWB radar has become popular recently due to its many merits. And it is a typical big data stream mining problem when detected human being in real time. In this paper, we proposed a statistical algorithm based on spectral method for big data stream anomaly detection. The through brick wall human detection experiment was designed and the results showed that the proposed method could detect the human being with high confidence level.

Video-SAR can produce sequential images at high frame rate, providing a continuous surveillance over a region of interest (ROI). To accelerate the speed of image processing, the whole circular path of Video-SAR is divided into several sub-apertures. However, due to the severe coupling of range and cross-range direction of the circular path, standard autofocus techniques as Phase Gradient Autofocus (PGA) fail since the direction of image blur changes for each sub-aperture image (sub-image). We have developed an algorithm for sub-aperture processing of Video-SAR that estimates the phase errors from strong scatters in the unfocused sub-image based on echo regeneration, regardless of the sub-aperture position. After phase compensation, more focused sub-images can be obtained before they are compiled into a SAR video. Thus the efficiency and utility of Video-SAR sub-aperture processing is retained, while image quality is not compromised due to platform motion errors.

A bandwidth allocation method based on smartphone users’ personality traits and channel condition is studied in a unified mathematical framework in this paper. Based on the relationship between automatically extracted behavioral characteristics derived from rich smartphone data and Big-Five personality traits, the service provider could estimate each user’s probability of each personality trait using diagnostic inference, and then based on predictive inference to calculate each user’s usage of bandwidth using Bayesian Network. This could help the service provider to better allocate the smartphone bandwidth. For our proposed smart bandwidth allocation scheme, both the outage capacity and the outage probability are studied in fading channel. The service provider could adjust the bandwidth allocated further on account of the real channel condition, which makes our proposed algorithm more robust.

The channel inconsistency will generally degrade the performance of a multichannel system. So it’s necessary to equalize the mismatch of each channel. In the navigation anti-jamming receiver based on array antennas, the channel inconsistency will lead to a distortion of the beam and degrade the anti-jamming capability. This paper proposes a frequency-domain equalization method to compensate the channel inconsistency and improve the anti-jamming capability of the receiver. The simulations show that the method is highly effective and can improve the system performance.

In order to improve the stability of the beam pattern, a newly adaptive beamforming method (SPAPES, Space-Polarization Amplitude and Phase EStimation) for vector sensor array is presented in this paper. The signal model is introduced first, and the principle of the proposed method has been discussed, and then the optimal weight vector is derived. The desired signal will be cancelled out when calculating the covariance matrix, thus the optimal weight vector is irrelevant to the desired signal. Simulation results indicate that SPAPES can achieve a good filtering performance under the circumstances of high desired signal power, low sampling snapshots and coherent interference signal.

Due to the problems of high cost and complication in indoor positioning, a new kind of positioning method by using a mobile phone based on magnetic, angular, rate, and gravity (MARG) sensors is more favored in recent years. This method estimates the orientation of pedestrian by quaternion. The quaternion-based extended Kalman filter (EKF) used for data fusion and orientation correction reduces the error of linear acceleration and avoids magnetic-field interference. We conduct pedestrian gait detection and step length estimation by using an accelerometer and verify the positioning performance of this method in an mobile phone. Testing results indicate that the positioning accuracy can reach 30 ‰ and 20 ‰ in the complex magnetic-field and non-magnetic interference environments respectively.

Pulse compression technology is very useful in modern radar system, with the reason that it can solve the contradiction between the radar range and range resolution. Now, nonlinear frequency modulation (NLFM) is researched widely, because the pulse compression of the NLFM signal can lower the range sidelobe better than −40 dB without window weighting. But in some applications, the reducing level also cannot meet the requirements. This paper put forward a continuous nonlinear FM waveform designed bring down the level of the range sidelobe, which can suppress the range sidelobe to −86 dB. Therefore, it can be used in many radar systems, especially in some kind of weather radar.

This paper focuses on the power allocation and relay location for decode-and-forward (DF) relaying in the presence of co-channel interference. Firstly, we derive the approximation outage probability in the high signal-to-interference ratio (SIR) regime. Then, three optimization problems are formulated to minimize the obtained approximation outage probability, namely optimal relay location with fixed power allocation, optimal power allocation with fixed relay location, and joint optimization of relay location and power allocation. The simulation results validate our analysis and show that joint optimization obtains the best outage performance.

In this paper, we propose a robustness enhanced beamformer which does not involve any additional constraint on weight vector beside the distortionless response constraint. The proposed algorithm enhances its robustness against steering vector error by incorporating a term which aims to minimize the cross-correlation between the real and the imaginary parts of the desired signal in the objective function. Extensions of the proposed algorithm to

l

1

-norm minimization and incorporation of robust constraint are also addressed. Computer simulations verify validity and advantage of the proposed algorithm.

For the fading channel effect on the performance of cooperative spectrum sensing in cognitive wireless network, an improved clustering cooperative spectrum sensing algorithm based on double-threshold energy detection is presented in this paper. Within each cluster, the cognitive user that the channel quality is best to the data fusion center (DFC) is chosen as the cluster head (CH), other cognitive users in the cluster use the double-threshold energy detection on the local position. The detective information is sent to the CH, the CH makes the decision of the cluster, and then the decision information of each cluster is sent to the DFC by each CH. The DFC use the “or” rule to fuse each clusters’ results, and make a final decision whether the authorized user (AU) exists. Simulation results show that the proposed algorithm can further improve the detection probability of cognitive wireless network than traditional cooperative spectrum sensing based on double threshold detection.

When a primary user uses frequency hopping communication, cognitive radio users using typically spectrum detection method is the time-domain autocorrelation. But the primary users signal is interfered by a fixed-frequency interference (FFI), the method is invalid. For the problem, this paper improves the traditional time-domain autocorrelation method by using the power spectrum cancellation, and the improved method can effectively avoid the fixed-frequency spectrum interference to increase the spectrum detection performance. Simulation results show that signal-to-noise ratio (SNR) is below −10 dB, and the false alarm probability is 0.05, the detection probability of improved method is greater than that of the traditional time-domain autocorrelation method. In low SNR, the improved method has a good detection performance, and also can prevent the collision between frequency of FFI and frequency of the primary user as well.

A false sensing information attack can cause the detection performance down for cognitive radio, for this problem, this paper presents an improved weighted sequential probability ratio test (WSPRT) algorithm by based on an accuracy combining effectively using the user data of small weight. The method enhances weight by recording the accordant times between the previous spectrum sensing report and the final spectrum sensing decision, at the same time, the malicious users are distributed smaller weights in order to use the sending data of malicious users. Simulation results show that the improved algorithm can be effectively resist spectrum sensing data falsification (SSDF) attacks by comparing with traditional WSPRT when there are more malicious users.

In orthogonal frequency division multiplexing (OFDM) system, basis expansion model (BEM) has been commonly used because of the significant variation of doubly-selective channels. In such a model, the channel taps are estimated by linear combinations of prescribed basis function. We propose a method of making better use of discrete prolate spheroidal sequence, also called Slepian sequence with an optimized parameter in a mathematical relationship with Doppler shift, and then using inverse reconstruction method to recover the BEM coefficients of the channel taps. Simulation results have confirmed the validity of this improvement and also proved an enhanced use of Slepian sequence than other basis functions in high mobility environments.

Although the mobile commerce has wide prospect, it has to face with huge challenges because of potential security risks, privacy problems and efficiency due to its limitations of processing capability, and storage space and power supply in mobile terminal. The secure and efficient open PKI (Public Key Infrastructure) identity authentication scheme is one of the possible solutions in secure identity authentication of the mobile commerce. In this paper, we propose a novel secure and efficient mutual open PKI identity authentication scheme, named SEMOPIAS, to solve the problems of validity, one-time use of identification and unauthorized application system in open PKI identity authentication scheme. Analysis and evaluation are performed to demonstrate that the SEMOPIAS achieves stronger security and higher efficiency by simplifying the message format and optimizing the authentication processes. Especially, the advantages of the SEMOPIAS will be achieved more significantly when there are multiple clients or merchants participated in the scheme.

Multi-Input Multi-Output (MIMO) are known as increasing the link throughput by providing a multiplexing gain which scales with the number of antenna cellular systems. Based on noncooperative cellular systems with multiple antennas, uplink capacity is analyzed in this paper. The concise expression of the approximated capacity is derived in the reverse link by utilizing Bernoulli’s law of large number. Furthermore, the impact of several factors on the noncooperative cellular systems in practical application is discussed. Finally, computer Simulation is presented to verify the theoretical analysis.

Aimed at these problems that the transmission rate changes greatly, the slow start threshold (ssthresh) sets statically and the abrupt transition causes the multiple packets losses from a window of data and retransmission timeouts on the stage of slow-start of TCP congestion, this paper proposes an improved slow-start algorithm called TCPBP, which utilizes a passive end-to-end bandwidth estimation technique to estimate in real time the available bandwidth, then realizes the dynamical updating of ssthresh according to the network status. Furthermore, we adopt a phased approach to adjust the size of the congestion window (cwnd) during slow-start phase. Simulation results show that TCPBP can effectively avoid the phenomenon of multiple packets losses with respect to the traditional TCP and achieve the smooth access and the transition to congestion avoidance stage, which improve the performance of network, increase the bandwidth utilization and throughput, and lower the packet loss rate.

In this paper, we propose a new spectrum sensing method which could detect the time-variant fading channel gain and primary user state jointly. This joint estimation algorithm is based on the maximum a posteriori probability criteria and the particle filtering technology. Experimental simulations verify the superior performance of our presented joint detection scheme over traditional detection methods such as matched filtering detection and energy detection under time-variant flat fading channel.

In this paper, spectrum sensing is investigated when the decision statistic is computed using energy detection (ED) with Welch’s method in the frequency domain. First, we assume an estimated noise variance is used to calculate the threshold, instead of the priori exact noise. We present an analytical model to evaluate the performance of the conventional Welch’s ED. The characteristics of this model are also analyzed to show the effect for the performance of spectrum sensing. Then an optimal threshold is proposed to achieve high detection probability and low false alarm probability at low SNR levels. The analytical results and simulations demonstrate the effectiveness of the proposed optimal threshold.

In searching for alternative solutions of secret key generation in wireless networks, many physical-layer-based methods have been proposed. These methods exploit the inherent randomness and reciprocity of the multipath fading channel to generate secret keys. Multiple-antenna system provide higher key bit generation rate than single-antenna one because of more channel randomness. However, traditional MIMO key generation schemes produce prohibitively high bit mismatch. To address the tradeoff between key bit generation rate and key agreement probability, we propose a compressed sensing (CS) method for key generation in single carrier MIMO sparse multipath channel. Theory analysis shows that the CS-based key generation scheme provides high entropy key bits from fewer probes. Performance simulation reveals the proposed scheme achieves a high key agreement probability at a high key generation rate.

In traditional CDMA networks, transmit power control (TPC) is main technology to mitigate interference of cell-edge users (CEU), and in 4G-LTE networks with OFDM, fractional frequency reuse (FFR) for CEU is main technology for its flexibility of subcarrier pre-coding. In this paper, we propose a FFR scheme based on cell-edge interference alignment (IA) in DCF infrastructure networks first, then present a performance analysis and compare it with TPC by using probability theory with same node density based on BM model, simulation are given at last. Results show that TPC and soft FFR are almost same in terms of average throughput, former has a little more outage or error probability; TPC with upload and download in different channel has the best throughput performance.

Teletraffic demand in important activity regions is high, thus congestion shappen frequently. So how to meet the wireless communication demands of these places become a major concern of operator. In this paper, we have explored this problem by careful analysis of business statistics, business model building and total demand prediction. Relevant radio network construction scale and need were then introduced. Finally, we designed the corresponding three-dimension network scheme.

In this paper, the basic theory of amplitude equalizer was mentioned, with the main specification and type. Model of resistor-aided ultra wide band (UWB) microstrip equalizer was presented and the equivalent circuit was analyzed. According to the analysis result, the length of a short circuit resonance stub was calculated roughly. Microwave simulating software High Frequency Structure Simulator (HFSS) was used to develop the model and perform simulation optimizing design. At last the equalizing module was produced, and the 2–18 GHz microstrip equalizer was finished. The measure results showed good performance, what’s more, the power equalizer was small in size and was already in use.

With high operating frequencies and high emission power, 60 GHz millimeter communications may suffer seriously from realistic hardware impairments. Among this, nonlinear power amplifier (PA) will significantly degrade its transmission performance. In order to decrease the bits error rate (BER) of 60 GHz millimeter communications, in this investigation a promising polar coding scheme is proposed. Based on the derivation of the general formulas of polar code, a new construction scheme for polar code is proposed. This designed polar code is further applied to 60 GHz systems with nonlinear PA. Experimental simulations verified the proposed coding scheme, which may significantly promote the transmission performance of 60 GHz communications. It is also demonstrated that the proposed polar coding scheme will surpass the popular LDPC code, which hence provides a great promise to practical use.

This paper presents a novel human activity recognition method that uses UWB signals to enable a low clutter outdoor environment sensing and recognition of human activities, which can transmit the information and identify human activities simultaneously. Since UWB signals do not require line-of-sight and have very good ability of penetration, the proposed method can enable a low clutter outdoor environment human activities recognition using the UWB signals in wireless communication. Further, it achieves this goal for a through-targets scenario and without requiring seeing devices (e.g., camera, radar). We evaluate the proposed method using UWB signals in a playground, with eight human subjects performing eight different activities. The type of the human activities performed between the transmitter and receiver of UWB communication system can have significant effects on the shape of the received signal waveform. From these time-varying signals, we extract features that are representative of the activities types based on 1-D diagonal slice of fourth-order cumulant within a time window. Then, we use support vector machine (SVM) to realize the human activities identification. Our results show that proposed method can identify and classify a set of eight activities with an average accuracy of 99.2 %.

In this paper, we propose a new approach to detect and recognize the target obscured by foliage based on real data collected by an Ultra-wide band (UWB) radar sensor. The new proposed method is the combination of support vector machine (SVM) and memetic algorithm. SVM is a powerful tool for solving the recognition problem with small sampling, nonlinearity and high dimension. Memetic algorithm is applied to determine the optimal parameters for SVM with highest accuracy and generalization ability. Moreover, the feature vectors for target detection and recognition are obtained from target echo signal that processed by wavelet packet transform (WPT). The results of the experiments indicate that this proposed approach is an effective method for sense-through-foliage target detection and recognition, which has higher recognition accuracy than that of the artificial neural network and the SVM which has randomly extracted parameters.

To realize the direction of arrival (DOA) estimation of wideband Linear Frequency-modulated (LFM) source with local scattering, an approximate model is derived by Taylor series expansion and a novel estimation algorithm using fractional Fourier transform (FrFT) is proposed. New array data models in both the time domain and the fractional Fourier domain are presented and the one-to-one relationship between the location vector in the Energy-concentrated domain and the spatial parameters is given. Then, the conventional Multiple Signal Classification (MUSIC) algorithm is exploited to estimate the spatial parameters of multiple wideband LFM sources with local scattering in the Energy-concentrated domain. Compared with the previous FrFT-MUSIC algorithm based on the assumption of point source model, the proposed algorithm has better performance on location position accuracy and anti-noise property.

This article presents a novel algorithm to estimate the spatial parameters of the coherently distributed wideband linear frequency-modulated (LFM) source model, which is a generalization of the conventional distributed source parameter estimator (DSPE) algorithm. A new array data model is constructed through the energy concentrated property of LFM signal in proper fractional Fourier domain. The central angels and the extension widths of multiple LFM sources can be estimated separately using the proposed algorithm. Compared with previous FrFT-MUSIC algorithm, the proposed algorithm has a better performance on location accuracy and anti-noise property, and can determine the number of incident sources which in this algorithm is allowed to exceed the number of sensors in the array.

Due to the surge in the number of vehicles throughout the world, traffic problems become increasingly severe. Thus, the implementation of lane recognition plays an important role in Intelligent Transportation Systems (ITS). In this paper, we present a RFID based positioning approach for lane recognition which is also practical for scenarios like tunnels or multilayer viaduct where GPS is not available now. As Phase Difference of Arrival is the essential information for localization, two UHF RFID phase extraction schemes have been discussed. Experimental results show that both of the two schemes can achieve an error recognition rate under 6.0 %. The sub-sampling scheme has a better performance than Zero-IF Receiver at the expense of system complexity.

This paper presents a method for target positioning on the map. It provides an intuitive and effective view of the target location. With this method the specific target positioning information can be gotten and the attribute of target can be viewed and analysed easily. The method is implemented using C# windows forms and JavaScript scripting based on Map API. The JavaScript scripting is used to create a page map and accomplish map coordinate covering and other necessary operations. The C# windows forms are used to build the operation and display interface in the VS2010 development environment. This approach is applied to the specific positioning of vessels. The results show that the vessel can be well positioned on the map, and can achieve a good purpose for the analysis of vessels.

Among different types of Wireless Local Area Network (WLAN) positioning systems, fingerprinting-based localization exhibits good accuracy performance, and thus becomes increasingly popular. Fingerprinting-based localization needs extensive calibration effort to construct a Received Signal Strength (RSS) fingerprint database, which hinders the large development of WLAN localization. In this paper, to reconstruct the raw sparse fingerprinting database and reduce manual effort, we propose a novel integrated Propagation Model-based Breakpoint Model Interpolation (PMBMI) and Multidimensional Linear Interpolation (MLI). We carry out extensive experiments in a ground-truth indoor environment to examine the localization accuracy of our proposed approach. Experimental results show that our proposed approach can be applied to reduce the labor cost in off-line phase, as well as guaranteeing high-enough accuracy performance.

In Wireless Local Area Network (WLAN) environment, most of the existed Received Signal Strength (RSS) based localization algorithms rely on RSS data collection at Reference Points (RPs) in off-line phase. The process of RSS data collection is usually time consuming and labor intensive. To solve this problem, we make use of the relationship between the raw RSS sequences and the architecture of ground-truth environment, as well as the correlation among different raw RSS sequences to construct logic graphs. After that, by conducting mapping from the ground-truth graph into logic graph and doing mapping selection, we locate the target in the subarea which the target really belongs to. Since our proposed approach does not require the exact locations of the collected RSS data, a large amount of time and laboring cost for site survey is saved. Experimental results show that our proposed approach can be used to locate target in WLAN environment without site survey on WLAN RSS data.

This paper proposes an automatic cell cycle localization method based on the Latent-Dynamic Conditional Random Fields (LDCRFs) model. Since the LDCRFs model can jointly capture both extrinsic dynamics and intrinsic sub-structure, it can simultaneously model the visual dynamics within one stage and visual transition between adjacent stages in one mitosis sequence. Based on our previous work on candidate mitosis sequence extraction and classification, this paper mainly focuses on the formulation of LDCRFs for cell cycle modeling. Besides, the model learning and inference methods are also presented. The evaluation on C2C12 dataset shows the superiority of the proposed method.

Although the methods based on spatio-temporal interest points have shown promising results for human action recognition, they are not robust in complex scenes especially background clutter, camera motion, occlusions and illumination variations. In this paper, we propose a novel method to classify human actions in complex scenes. We suppress the false detection interest points by detecting salient regions. Furthermore, we encode the features according to their spatio-temporal relationship. Our method is verified on two challenging databases (UCF sports and YouTube), and the experimental results demonstrate that our method achieves better results than previous methods in human action recognition.

Human action recognition is a hot topic in computer vision and pattern recognition. Semantic analysis, as a kind of effective bridge to connect the human brains and computers, has been widely studied over the past several years and a number of methods have been proposed. However, there is no comprehensive study concerning semantic analysis in human action recognition. In this paper, we make a survey on various semantic analysis methods based on an approach-based taxonomy. We choose several representatives from semantic analysis methods and evaluate them on publicly available datasets.

A novel algorithm based on fourth-order cumulants (FOC) was proposed when coherent and incoherent wideband OFDM signals coexist. The proposed algorithm comprises two steps. The first step is to estimate the DOA of incoherent OFDM signals, then a FOC matrix that only contains the information of coherent OFDM signals was constructed by eliminating the contribution of incoherent OFDM signals; the second step is to estimate the DOA of the coherent OFDM signals from the constructed matrix by using a sparse reconstruction method. The simulation results demonstrate the performance of our proposed algorithm.

Network congestion is the determinant of network transmission delay. The QoS differentiated multi-path routing mechanism is proposed in this paper to balance the network load of industrial Ethernet, under the limited network resource condition. ACR algorithm is used to obtain the path sets for data transmission with different QoS requirements. The mathematical model of differentiated multi-path routing algorithm in industrial Ethernet is established and described in detail. Furthermore, key metrics involved in design is analyzed deeply. Difficulties in implementation based on ACR algorithm is pointed out, and the solution is given in specific description.

Link prediction is an important issue to understand the dynamics and evolution mechanisms of complex networks. Traditional link prediction algorithms are based on the topological properties of the underlying network in terms of graph theory. In order to improve the accuracy of link prediction, recent researches increasingly focus on modeling the link behaviors from the latent structure information of the networks. In this paper, we propose a neighborhood-based nonnegative matrix factorization model to solve the problem of link prediction. Our model learns latent feature factors from the overall topological structure combing with local neighborhood structures of the underlying network. Extensive experiments on real-world networks demonstrate the effectiveness and efficiency of our proposed model.

The thesis adopts the theoretical tool of network calculus theory to conduct the performance research on the cascade topology of deterministic Ethernet. Based on the simple star topology, the thesis proposes the cascade topology which is the extension of the simple star topology. The thesis conducts the in-depth study of the network calculus theory and deterministic Ethernet. Firstly, we adopt the deconvolution theory to establish the network calculus model of the cascade topology. Secondly, we establish the simulation model. Lastly, we compare the simulation results with the theory results to analyze the determinacy and real-time of deterministic Ethernet. The conclusion shows that the cascade topology has a good performance.

In this paper, we propose an improved TCP scheme, TCP BRJ, which is capable of adjusting the initial slow-start threshold and congestion window in real time according to the bandwidth estimation in slow-start phase, dividing the network congestion grades based on round-trip time (RTT) jitter in congestion avoidance phase, distinguishing the random packet losses from the congestion packet losses, and reacting accordingly. Simulation results by NS-2 show that TCP BRJ provides more significant performance improvement in throughput, bandwidth utilization and fairness than TCP Reno and TCP Westwood in heterogeneous networks with high random bit-error rate (BER), and shows friendliness towards the widely used algorithm TCP Reno.

Transmission characteristics of a half-filling photonic bandgap fiber Sagnac interferometer are investigated. The temperature responses of the interference dips are further studied. They are fully related to the temperature responses of the phase birefringence and group birefringence. Two of the interference dips have opposite shift direction and different temperature responses. An ultrahigh sensitivity temperature sensor with sensitivity of −20 nm/°C is achieved. In addition, the transmission loss at a certain wavelength changes with temperature changing, and an optical switch with extinction ratio of 45 dB is achieved.

A fusion method based on non-subsampled contourlet transform (NSCT) in compressed sensing was proposed. The method decomposes two or more original images using NSCT, and gets the sparse matrix by the NSCT coefficients sparse representation, and fuses the sparse matrices with the coefficients absolute value maximum scheme. The compressed sample can be received through randomly observed. The fused image is recovered from the reduced samples by solving the optimization. The simulations show that the proposed CS-based image fusion algorithm has the advantages of simple structure and easy implementation, and also can achieve a better fusion performance.

Compressive sensing (CS) has given us a new idea at data acquisition and signal processing. It has proposed some novel solutions in many practical applications. Focusing on the image compressive sensing problem, the paper proposes an algorithm of compressive image sensing based on the multi-resolution analysis. We present the method to decompose the images by nonsubsampled contourlet transform (NSCT) and wavelet transform successively. It means that the images can be sparse represented by more than one basis functions. We named this process as blended basis functions representation. Since the NSCT and wavelet basis functions have complementary advantages in the image multi-resolution analysis, and the signals are more sparse after decomposed by two kinds of basis functions, the proposed algorithm has perceived advantages in comparison with compressive sensing in the wavelet domain which is widely reported by literatures. The simulations show that our method provides promising results.

With the rise of cloud computing, cloud storage has become a challenging issue. It is a huge challenge to design a distributed file architecture to meet the requirements of cloud storage. In this paper, in order to improve the system reliability and performance, we propose a cloud storage architecture with a meta-data service layer. The proposed architecture is a distributed file storage system based on the master-slave architecture, which uses multiple proxy servers of the meta-data server to establish a peer-to-peer meta-data service layer. Each meta-data server and the proxy server, could be an access to service for clients rather than only one fixed access as usual, which can improve the parallel processing performance of a meta-data service layer greatly. Some P2P techniques are used between proxy servers to solve the disadvantages of the master-slave architecture efficiently. Analysis and evaluation are performed to demonstrate that the proposed architecture improves the system reliability and performance greatly.

The spatial layer bit allocation problem of H.264/scalable video coding (SVC) is solved in this paper. The challenge of this problem lies in the fact that the low frequency information in spatial scalable video, which consists mainly of the mean absolute difference (MAD) of the residual texture in the base layer, is not always the suitable reference for inter-layer prediction. To overcome this issue, we first analyze the MAD prediction in the enhancement layer from both temporal and spatial aspects, then derive the Rate–Distortion (R–D) models of the spatial layer analytically, and finally develop a model-based bit allocation algorithm. Experimental results show that the proposed bit allocation algorithm can achieve a coding performance close to the optimal R–D performance based on the full search method, and outperforms two prior H.264/SVC bit allocation algorithms served as benchmarks.

A lot of people die in every year all around the world in the traffic accidents caused by vehicle roadway departure crashes. A lane-detection system is an important part of intelligent transportation to avoid these accidents. In this paper, we novelly present an algorithm which can detect lanes though vanishing points, track lanes though particle filter on each control, and refresh these control points though an scoring algorithm. We can get some pairs of lines though an vanishing point in detecting parts, get some pairs of lines though particle filter on each control point, find the lane in all the pairs of lines though the scoring algorithm, and then refresh these control points by the end-points of each part of the lane. General process is that, we first use a Gabor filter to find the edge in an image. And then, we divide the image into five parts and use Hough transform to find the lines in each part. After that an algorithm based on particle filter and vanishing points is novelly proposed to generate a large number of hypotheses. Then we will use our scoring algorithm to mark every control point/pair of lines to select the control points/pair of lines whose score is highest. At last, a pair of lane is obtained by fitting function. Experimental results on real roads will be presented to prove the effectiveness of the proposed lane detection algorithm.

This invention creatively puts forward a method that we can use a dynamic template combination of pixels for character segmentation in license plates. Firstly, we can obtain the dynamic vehicle monitoring area through the background modeling based on Gaussian mixture model and local license plate by using the edge information of license plate. Secondly, combining structure information, texture information, the rules of arrangement and maximum correlation among characters, we will find segmentation position after sliding this template. Then we can get the best position of license plate segmentation through calculating the number of nonzero pixels in the segmentation position in a binary image. At the end, after extracting the features of the characters, we should use the SVM for training and license plate character recognition [1, 2]. Compared with other solutions in the field, the invention has high accuracy, good robustness and stability in complex environment.

With the development of 3D camera, efficient and effective 3D model retrieval algorithms are highly desired and attracted intensive research attentions. In this paper, we proposed a view-based 3D model retrieval method based on spatial context constrained characteristic view extraction method. First, according to the spatial constraints of views, we cluster all of 2D images, which capture 3D model from different angles. Second, the random-walk algorithm is utilized to update the weight of each view to help us to select the most representative view. Finally, we apply Bayesian model to compute the similarity between query model and candidate 3D model to find the best matching 3D model. Experimental comparisons have been conducted on the ETH and NTU 3D model datasets, and the results have demonstrated the superiority of the proposed method.

Compressed sensing (CS) has drawn quite an amount of attentions as a joint sampling and compression approach. Its theory shows that if a signal is sparse or compressible in a certain transform domain, it can be decoded from much fewer measurements than suggested by the Nyquist sampling theory. In this paper, we propose an unequal-compressed sensing algorithm which combines the compressed sensing theory with the characteristics of the wavelet coefficients. First, the original signal is decomposed by the multi-scale discrete wavelet transform (DWT) to make it sparse. Secondly, we retain the low frequency coefficients; meanwhile, one of the high frequency sub-band coefficients is measured by random Gaussian matrix. Thirdly, the sparse Bayesian learning (SBL) algorithm is used to reconstruct the high frequency sub-band coefficients. What’s more, other high frequency sub-band coefficients can be recovered according to the high frequency sub-band coefficients and the characteristics of wavelet coefficients. Finally, we use the inverse discrete wavelet transform (IDWT) to reconstruct the original signal. Compared with the original CS algorithms, the proposed algorithm has better reconstructed image quality in the same compression ratio. More importantly, the proposed method has better stability for low compression ratio.

Clouds are one of the most important meteorological phenomena related to the hydrological cycle and affect earth radiation balance. Ground-based cloud classification, as a key component of cloud analysis has received great concern in recent years and a number of algorithms have been proposed. However, there is no comprehensive study concerning the different classification methods. In this paper, we first make a survey on various classification methods. Further, we detail the current algorithms for classification and give theoretical analysis. Finally, we evaluate the performance of different algorithms under two ground-based cloud datasets. The experimental findings firmly justify our theoretical analysis.

In this paper, an effective image splicing detection algorithm based on the local statistical features in DCT and DWT domain is proposed. The local ternary pattern (LTP) operator is introduced to characterize the statistical changes of DCT and DWT coefficients caused by image splicing. The LTP histograms are generated from the magnitude components of the block DCT coefficients with varying block sizes and the DWT coefficients in three detail subbands, respectively. All these LTP histograms are concatenated together to form the discriminative feature set for splicing detection. The effectiveness of the proposed detector is evaluated on the Columbia image splicing detection evaluation dataset. Simulation results have shown that the proposed method can perform better than several state-of-the-art methods investigated.

This paper develops an estimation approach for error-in-covariables partially linear model with the help of validation data. A feasible least squares estimator of the parameter is defined by means of local linear kernel estimator. Further, the estimator of non-parametric component is also defined by the Fourier transformation method. A small simulation studies are given and the convergent behaviors for the proposed parametric and nonparametric estimators are shown respectively.

The traditional algorithms are normally suffered from large computation complexity and serious acquisition performance. This paper proposes a fast pseudo noise (PN) phase synchronization algorithm for mobile station downlink channels. Based on overlap-saving method and fast Fourier transform (FFT), the proposed algorithm achieves phase synchronization with less time and resource consumption. Simulation results indicate that the proposed algorithm could significantly decrease PN codes acquisition time cost and is proved to perform well in PN phase synchronization.

With the development of mobile information and intelligent mobile phone, wireless data services have changed in content and form. This paper expounds the function and design of wireless test system based on J2ME platform, using RMS for data persistence, meet the needs of learning anywhere and anytime possible, to achieve efficient and timely communication between teachers and students, provides a new mode for a new mobile education system.

In recent years,to meet the challenge of spectrum sensing with ultra wide band and big data in cooperative and cognitive radio networks,the theory of compressed sensing is introduced in,which can solve the problem of high sampling rate requirement due to Shannon-Nyquist sampling theory. In this paper,considering the property of signals’ cyclostationarity, we innovatively propose a method in OFDM signal detection using sub-Nyquist samples. By doing sparsity analysis combined with detection necessities,we present a partial-scale reconstruction method to reduce the recovery iteration and lower the algorithm complexity. Furthermore,we find out an equivalent cyclic cumulant calculation method for OFDM signals to simplify the calculation and lower the high memory consumption during signal processing. From the simulations we can see the optimized method introduced in effectively eliminates the constraints for compressed detection of OFDM signals and possesses a far-reaching significance in further researches and applications.

In this letter, a novel rectangular waveguide-to-half mode substrate integrated waveguide (HMSIW) transition is presented. The transition is realized by using a trapezoidal-shaped probe to terminate the TE

10

mode of the standard waveguide within a wide frequency band. A back-to-back transition at Ka-band is designed and fabricated. The simulated results show that the proposed structure has less than 0.35 dB insertion loss and a better than 20 dB return loss within a frequency range from 25 to 40 GHz for a back-to-back structure. In addition, there is no need of intermediate transition for this design. The size of the proposed transition is reduced by approximately 81.8 % as compared with the Waveguide-to-HMSIW transition using antipodal fin-line. The proposed transition has the advantages of compact size and sample structure, and it is suitable for the application of HMSIW technology.

Text extraction from various text containers like document, born digital images, real scenes and videos has been a continuous interest in this field for more than a decade. Although a lot of work has been done on printed Mongolian document image analysis, there has little work on Mongolian text extraction from complex images. For the design and evaluation of Mongolian text extraction algorithms and systems, the availability of large-scale dataset is important. This paper first introduces a dataset named CSIMU-MTR which is built by the College of Computer Science of Inner Mongolia University. And then presents benchmark results using two state-of-the-art methods in text detection on this new dataset. The reported results serve as a baseline for evaluating the further works.

The traditional signal detection methods mainly focus on suppressing noise to extract the weak signal. However, stochastic resonance (SR) can enhance the signal component by converting energy from the noise to the signal. Base on the theory of SR, a novel approach to detect weak signal with a short data record is proposed. Ensemble average and cross-correlation operation are applied in this method to improve detection performance. In order to settle the limitation of stochastic resonance to detect large parameters signal, scale transformation stochastic resonance (STSR) is presented. The result of simulation proves the effectiveness of this designed method.

To solve the numerical instability in the recursive process of unscented Kalman filter (UKF), as well as the unsatisfactory performance in case of abrupt changes, a new adaptive target tracking method, called square root unscented Kalman filter based on strong tracking (STF–SRUKF), is presented. On the one hand, inspired by the idea of square-root filter, the square root of the covariance matrix is substituted for the covariance matrix itself in the recursive process, to guarantee numerical stability. On the other hand, based on the idea of strong tracking filter, a time-varied fading factor is introduced into the recursive process, which is helpful to adjust the gain matrix timely, and thus enabling STF–SRUKF more power to deal with sudden changes. Experimental results demonstrate that STF–SRUKF performs well and steadily, especially when target motion changes suddenly.

To estimate resources availability measurement in system integration, a new method is discussed. Firstly it analyzes the change of system resource state and discusses the mechanism of transmission of failures based on resource faults. Secondly the method of resource faults-based health measurement, the representation of classification and formalization for resources in system integration, and the theoretical means of availability metrics of resources and resource platform are proposed. Finally an algorithm for availability measurement of resource integrated platform is proposed.

A compact and broadband rectangular waveguide-to-half mode substrate integrated waveguide (HMSIW) transition based on antisymmetric tapered probes is proposed. The antisymmetric tapered probes are used to change the quasi-TE

0.5,0

mode in the HMSIW to the TE

10

mode in the rectangular waveguide. To verify the proposed transition circuit, a back-to-back transition structure has been designed, fabricated and measured. The measured results show that an insertion loss less than 1.6 dB and a return loss better than 14 dB at 26–40 GHz are obtained for a back-to-back transition. Compared with the waveguide-to-HMSIW transition using antipodal fin-line, the size of the proposed transition circuit is reduced by approximately 71 %. The high performance, compact size and simple structure, enable such transition to be employed in a number of other millimeter-wave applications.

In this paper, the automated detection of tissue specific disease association of long non-coding RNAs (lncRNAs) is modeled as a multi-label classification task, where a single lncRNA transcript may be associated with several diseases in a tissue specific manner. Four algorithms are evaluated and compared in this task. Furthermore, in this article we put emphasis on the fact that secondary structure and the composition features of the lncRNAs dictate their functions that led us to develop a new multi-label feature extraction scheme. Experiments are conducted on a set of 7,566 lncRNA transcripts with 22 tissue labels, and the results provide interesting insights into the quality of the discussed algorithms and the features.

Eriocheir sinensis

is a highly-commercial aquaculture species as an important aquatic product source. The protein-protein interaction network (PIN) of

E. sinensis

has been constructed based on the RNA transcriptional sequencing. Many unknown proteins exist in the PIN, which seriously restricts the further mechanism researches on regulation, immunity and development of

E. sinensis

. In this work, we predicted the functions of the unknown proteins in

E. sinensis

based on the modularity feature of the PIN. The functions of 677 (93 %) proteins were annotated. The analysis of ribosome module indicates that the annotation of proteins and modules provides important references to be studied in the further in vivo experiment as well as the biological capability of the

E. sinensis

PIN.

Relevant modules of a metabolic network can reveal the underlying structure of the metabolic system and hence provide the insight into its function. Modularity is a useful standard for decomposing a complex network. However, it has been shown that simply optimizing modularity failed to find the most natural community structure by overpartition or under partition the network. Here, a novel network decomposition algorithm, called MNetDec, is proposed by integrating the dendrogram and modularity features of a metabolic network. By dealing with MNetDec, the small modules with questionable biological function are eliminated. Moreover, the size of the smallest module and final number of the modules in the decomposition result can be flexibly set by the users based on their own specific needs. The application of this algorithm on brain specific human metabolic network shows that MNetDec can generate biological reasonable modules with high modularity.

This paper realizes the judgment that whether patients have throat polyp by normalization processing, principal component analyzing and Neural Network Classifying the extracted audio data. This implementation replaces the traditional approach to diagnosis of throat polyps. Conventional laryngoscopy need to cutout, clamp or puncture from the patient to remove the lesions to do pathological examinations, which is so hurt to the patient. The test for throat polyp prediction with the neural network classification algorithm are carried out. The results shows that the correct rate of prediction is stable under different number of samples and different random measurement matrices.

Attacking fingerprint-based biometric systems by presenting fake fingers is a serious threat for unattended devices. In this work, we introduce a novel algorithm, by extracting features along the fingerprint curves, to discriminate between fake fingers and real ones on static images. Pairs of mean value and standard deviation are sampled from the ridge, slope and valley of the curves. Then bag-of-words model is used to select cluster centers and form a 128-dimension feature of words’ frequency. We test our method on a dataset collected by Chinese Academy of Science, which contains 960 live fingerprints and 960 fake ones made by silicon. Though the fake fingerprints is too verisimilar to be distinguished by naked eyes, we still get an accuracy of 98.85 %. Because our method is based on single static fingerprint image, it can be freely embedded into existing fingerprint-based biometric systems.

According to the basic principle of Compressive Sensing, a method of implementation of CS theory on DSP by the CCSLink is proposed. In Matlab, we utilize the CCSLink tool to create suitable embedded target to analyze parameter visually and discuss the basic factors which can effect the reconstruction algorithm on the platform of OMAP-L138. By improving and optimizing the algorithm, we accomplish the implementation of the theory of compressive sensing on DSP finally.

Individual Adaptive Filter only allows one input signal through, so doped with a signal source when there are different types of noise or interference, an individual Adaptive Filter is not a good mix of original signal noise and therefore cannot fully restore the initial signal. Combining adaptive filters introduced precisely to address this issue. This article introduces the basic ideas of Adaptive filter design portfolio illustrates combined Adaptive Filter theory. And a combination filter was introduced in the practical application of spread spectrum technology. By simulating the experiments confirmed the combination filter has good filtering effect.

This topic takes STM32F407ZGT6 MCU as the master chip, has designed the wireless intelligent system of fire alarm, which includes 2.4G wireless communication module, smoke transducer, temperature transducer, the audible and visual alarm module, TFT LCD module and GSM communication module. Moreover, the research and application of wireless intelligent system would be discussed further in this topic. It’s convenient and flexible for wireless system to erect and move freely. Also the wireless system doesn’t need large-scale wiring. Currently, the wireless system has replaced the traditional wired network gradually and becomes a new trend in the development of communication technology. It’s so different from the way of traditional wired connection for wireless network that it can reduce the restriction of factors such as topography, terrain and price to the intelligent system. As a result, the intelligent system can be used widely in various fields without restrictions. According to the test, this design has accomplished each function basically and achieved the expected results. Therefore it will show a bright prospect about its application.

The design of spectrum monitoring receiver usually use FPGA (Field Programmable Gate Arrays) + DSP (Digital Signal Processor) hardware model, which requires the baseband data and spectral data under different bandwidths can be real-time transmitted between the FPGA and DSP. According to the characteristics of the receiver, a multi-rate matched universal parallel port (uPP), which is used to transmit data between FPGA and DSP, is implemented in this paper. The port achieves a high-speed data transmission, which is up to 150 MBtye/s with only 20 I/O lines. The adaptation of different input data rates is implemented by a ping-pong buffer in dual-FIFO (First In First Out).

Nonlinear distortion is generated when signals are driven through nonlinear power amplifiers (PAs), and it is a major degradation in orthogonal frequency-division multiplexing (OFDM) like multi-carrier signal transmitter systems. The nonlinearity of PAs is mostly evaluated by three measures, i.e., 1dB compression point (P1dB), the 3rd order intercept point (IP3), and percentage linearity (PL). In this paper, the interrelationship between these three measures of PA nonlinearity are derived by using a truncated complex power series model.

Colors show vivid colorful images displayed by three or more primary colors on each individual pixel. However, the brightness of the color greatly limits the color gamut of the displayer. According to the color mixture principle, the total color gamut shrinks with the increase of brightness. When the maximal brightness, the displayers only show white point of display system. This paper puts forward a theory of the relationship between brightness and color gamut based on multi-primary displayer. The paper simulates and estimates the color gamut boundary of multi-primary colors under the required brightness, which was proved by the experimental results of three primary colors displayer.

While many hydrological data have been collected and produced from hydrological models, the new data from the Office of Hydrologic Development, NOAA has a benefit over others in terms of high resolution both in temporal and spatial resolutions. In this work, the web-based Hydrological Visualization and Analysis System (HyVAS) is developed to help both hydrologists and local people examine and analyze this high resolution hydrological data. The HyVAS provides both temporal and spatial visualization and analysis tools. In total, there are two visualization tools and two analysis tools. Both 2D and 3D graphics have been implemented in these tools to aid users to view and to understand information hidden in the numerical data. In the first stage, this system focuses only on soil moisture visualization and analysis. Later if more data are available and more functions are needed, they can be added because this system does not depend on any environments, operating systems, or commercial software. This web-based Hydrological Visualization and Analysis System is hoped to help the hydrological community visualize and analyze data and extract useful information from the data.

Sichuan province is one of main producing areas of Radix Ophiopogonis which is a valuable herb in Chinese traditional medicine. It is important to monitor the field’s environment parameters during the growing period and maintain the soil moisture value to ensure both the quality and yield of Radix Ophiopogonis. This paper designs and implements the intelligent monitoring wireless network and automatic irrigation system for Radix Ophiopogonis. Our system includes hardware modularity, which separates PSU (Power Supply Unit) from sensor nodes, monitoring network with diversifying field environmental data, and irrigation expert model for Radix Ophiopogonis. Combined with weather data, such as temperature, humidity, and rainfall data from meteorological observatory, environmental data (such as soil humidity and surface temperature) gathered from our network are served as input of the irrigation expert model to make decision, then the result is fed back to SV Node (Solenoid Valve Node) to form a closed loop, which reaches the goal of field monitoring and irrigation control. The result shows that our system can provide moisture accurately for Radix Ophiopogonis, thus realize automatic, real-time and appropriate amount of irrigation. Our system also provides high control accuracy, reliable communication, and is especially suitable for medium-large-size farmland of Medicinal herbs.

This system is mainly based on ATMEL Corporation AT89S52 microcontroller as the core, with using a 12 MHz crystal oscillator to provide the clock signal, and the Hall sensor detects the vehicle speed, timing through software programming, mileage detection, costing the same time the visual image of the LED digital tube display mileage and total costs, so as to achieve the purpose of billing. The system has a total of five buttons (Clear, query/verify, stop, one way/return function selection), corresponding to the button operation can be achieved one way/round-trip pattern selection, stop charging, waiting for the time to query, cleared reset and other functions. Not only does it have quite simple structure, but also it has a bright future with a stable, intuitive display and easy operation.

In this investigation, we proposed a promising digital signal modulation recognition scheme which is inspired by the deep learning. Firstly, the signal discriminations are constructed, which are composed of the full temporal characteristics of digital signals, its frequency spectrum as well as several higher-order spectral characteristics. Subsequently, the deep learning algorithm, with the powerful ability of interpretations and learning, is further suggested to realize modulation recognitions. A major advantage of this new scheme is that it may fully exploit the complete information of digital signals, rather than only utilizing several extracted features. It is verified by experimental simulations that the recognition accuracy of the proposed new scheme is much superior to other traditional recognition methods, which therefore provides an attractive approach to realistic modulation recognition.

In the angle measuring using theodolite, the telescope is driven by the aiming motor to sight the target automatically. The parameters of aiming motor will vary with duty conditions to bring about aiming control system chaotic, which is harmful for the aiming system and the aiming results. The aiming motor control system exist the time-varying load and the inside disturbances, the dynamic model is established and analyzed. The behavior of chaos is proved. Due to terminal sliding control with good robustness, fast dynamic response, finite time convergence and high tracking precision, The finite time proportional-integral (PI) sliding mode structure and control strategy are given, and the system stability is analyzed. The chaotic orbits of the aiming motor control system are stabilized to arbitrary chosen the fixed points and periodic orbits by means of sliding mode method. Simulation results show that finite time PI sliding mode control can realize the stability and accuracy of aiming motor control system, and overcome the negative influence of the chaos for the aiming system.