Search

This paper presents a high-performance coupled BAW/SAW resonator based on ScAlN/Aln thin film hetero acoustic layered (HAL) structure. The influences of the SAW propagation characteristics for different geometrical parameters and configurations are investigated. The SAWs excited in the proposed HAL structure exhibits excellent performance due to the combined advantages of the coupling of BAW and SAW and its energy confinement effect. The optimized Mo/ScAlN/AlN/Mo/SiO 2 /Si structure exhibits a strong piezoelectric coupling (K2) of 21% and a high phase velocity ( $V$ ) of 5100m/ $s$ . And, a Qr is 3548 and Qar is 1348 around the resonance frequency of 4.83GHz and the anti-resonance frequency of 5.32GHz respectively. The coupled BAW/SAW resonator with large K2, high $V$ and high $Q$ provides promising prospects for development of high frequency, broadband and low-loss filters for 5G application.

In this work, a PMUT based on Pt/AlN/Pt/TiO2/SiO2/Si structure with square suspended membrane is proposed and modeled by performing finite element method (FEM) software COMSOL. The vibration mode, resonant frequency and effective electro-mechanical coupling coefficient (k eff 2) of the proposed AlN-based PMUT are analyzed. It is shown that the (0,1) resonant mode of PMUT reaches up to 25.73 MHz when the AlN thickness is 1μm. Moreover, the influences on the sensitivity of geometric parameters, i.e., the side length and thickness of top electrode, and the thickness of AlN thin film, are investigated. Maximum transmitting and receiving sensitivity of 32.79 pm/V and 0.478mV/kPa, respectively, are obtainable with optimum geometric parameters. Investigation results demonstrate that by optimizing the structure of the proposed AlN-based PMUT, k eff 2 and sensitivity are significantly improved, which is beneficial for meets the requirement for application of high resolution ultrasound imaging.

The Root Mean Square current (Irms) has been explored on both N doped un-silicided poly and silicided poly at 40nm technology node. It is found that poly resistance-current (RI) curves show an initially high resistance due to Schottky Rectifying contact. However, poly Irms can be estimated similar to metal Irms. Parameters for poly Irms calculation is provided on the basis of 5°C Joule heating.

In this paper, the received signal of accelerated moving target on three directions is modeled as a cubic phase signal (CPS) for distributed space-borne SAR, and the cubic phase function (CPF) method is used to estimate the parameters of accelerated moving target. The simulation results show that the cubic signal model is reasonable, and CPF provide a better performance for the estimations of velocities and accelerations of moving target for SAR than the method of high order fuzzy function (HAF), especially in the conditions of low SNR.

Bat algorithm (BA) is newly proposed bio-inspired metaheuristic algorithm with the inspiration of the echolocation of bats in nature. Several experimental results have proven to the effectiveness and performance of BA. However, BA may fail to find the global optimal solution occasionally. In this paper, a kind of classical search technology, called variable neighborhood search (VNS), is incorporated into BA as a local search tool. An improved version of BA namely variable neighborhood bat algorithm (VNBA), is thus proposed. In VNBA, the classic BA as a global search tool searches the whole space globally, and this can significantly shrink the search space. Subsequently, VNS as a local search tool is implemented to find the final best solution within the small promising area. After that, the VNBA is benchmarked by sixteen standard benchmark functions. The experimental results imply that VNBA takes the absolute advantage over the basic BA.

Today, plenty of apps are released to enable users to make the best use of their cell phones. Facing the large amount of apps, app retrieval and app recommendation become important, since users can easily use them to acquire their desired apps. To obtain high-quality retrieval and recommending results, it needs to obtain the precise app relationship calculating results. Unfortunately, the recent methods are conducted mostly relying on user's log or app's description, which can only detect whether two apps are downloaded, installed meanwhile or provide similar functions or not. In fact, apps contain many general relationships other than similarity, such as one app needs another app as its tool. These relationships cannot be dug via user's log or app's description. Reviews contain user's viewpoint and judgment to apps, thus they can be used to calculate relationship between apps. To use reviews, this paper proposes an iterative process by combining review similarity and app relationship together. Experimental results demonstrate that via this iterative process, relationship between apps can be calculated exactly. Furthermore, this process is improved in two aspects. One is to obtain excellent results even with weak initialization. The other is to apply matrix product to reduce running time.

Real-time and power management are required for high quality video in mobile environment. This paper presents a GPU(graphics processing unit) based parallel architecture for Multi-view Video (MVC) decoder which reaches these requirements. The 3D video based on stereo or multi-view representation is becoming widely popular. Real-time decoding of such video is an important concern as the number and spatial/temporal resolution of views increase. Significant improvement in video compression capability has been demonstrated by using H.264/Advanced Video Coding (AVC) standard. Multi-view video (MVC) is the extension of H.264/ AVC. In this paper, we proposed MVC decoder architecture based on parallel combination of Cortex-A8 processor and GPU (graphics processing unit). The basic operations are performed by the processor while the motion compensation (MC) feedback loop of the decoder is moved to GPU in order to achieve decoding efficiently. The experimental results show that compared to general implementation, the proposed parallel processing of a particular task in an embedded system can reconstruct the target images with higher quality with reduced processing time and energy saving with almost the same compression performance.

This paper built photovoltaic (PV) grid power system model in MATLAB/Simulink, and researched some factors which can influence the current harmonics of grid for single grid-connected photovoltaic (pv) inverter. These factors include irradiance, environmental temperature, the control method of inverter, the imbalance of the three-phase grid and the voltage harmonics of the grid. This paper also studied the condition that multiple photovoltaic grid inverters can weaken the grid current harmonics in the point of common coupling (PCC).

TinyOS is a key element of the software infrastructure for the research and development involved in realizing wireless sensor networks (WSNs). In this paper, a priority-based scheduling strategy is proposed and implemented. Furthermore, the proposed approach and the original mechanism are tested and compared in the laboratory environment. Test result indicates that the modified strategy greatly improved the system performance for real-time tasks.

This paper describes this development of a test platform of a grid-connection PV station. Essentially this entire platform involves the integration of a Personal computer (PC), a Dc power source, a power analyzer, an AC load and an analog grid to create a grid-connection PV system and to test. This new system boasts of high accuracy measurements coupled with the commercial viability of low cost. In the test system, various testing equipments have been controlled by the computer. This control interface has been developed by utilizing Labview software, which forms the programming backbone of inter-instrument communication via RS232 bus and Ethernet.

Our goal is to establish a simple technique for dynamic human body tracking based on image sequence. In this paper, we present a novel method of template matching and predictive tracking for moving body in clutter background. The method exploits the fact that moving human body changed in approximate rigid parts. First, we use the moving difference to find the region of interest, then we make a template of a target through the morphology algorithm using the skeleton of the object, and by revising the mask combined Kalman predictive tracking method to archive real-time tracking. It has been proved that in clutter background and noisy situation this algorithm is still efficient.

Maintenance is one of the important research subjects of the electric power system. In order to realize cooperative maintenance decision, the system framework of multi-knowledge, multi-method, multi-resource, multi-information based on swarm intelligence is constructed, the method of knowledge modeling is studied. Swarm intelligence is a quality that many individuals can behave intelligently through interaction. The individuals are open to each other, sharing the resource information, managing and dispatching are unified with the platform based on knowledge grid, the serving information is offered for different power plants, manufactures and research units. That is beneficial to enhancing the diagnosis and management of the maintenance decision support system, reducing its costs, and improving its efficiency, thus providing a feasible approach for the realization of condition-based maintenance gradually.

Binary Translation is a migration technique that allows software to run on other machines achieving near native code performance. The paper proposed an abstract machine-based dynamic translation technique in Java processors. The technique employs the “mock execution” of the hardware abstract machine (HAM) to identify and analyze the dependency among Java programs, dynamically translate Java bytecode into tag-based RISC-like instructions. After that, stack folding is combined with the technique to further optimize translated instructions. We used the technique to realize a Java ILP processor. To further describe the technique's availability, we extended the Java processor to design a multithreading Java processor, and explained its some new features.

Most modern file systems provide journaling method instead of traditional file system check routine for fast recovery and consistency guarantee from unexpected crashes. For clustered file systems with shared storage, to improve the performance of journaling, each node gets its own journal space and each journal can be on its own disk for greater parallelism. After a failure, file system consistency is restored quickly by simply re-applying all updates recorded in the failed node's journal. However, for file system operations involving multiple nodes, this method may not be efficient enough as lock mechanism has to be employed in most cases, which may cause poorer performance and more recovery time. By introducing specific journal storage to record the transactions involving multiple nodes, an enhanced lock-free journaling method is presented in this paper, which can improve the performance of journaling operations involving multiple nodes, and then shortens recovery time from crashes. A Linux-based concept-proofing implementation has been applied in self-developed clustered file system.

Locating and tracing torpedo is the key to warship anti-torpedo. Sonar can detect noise of underwater target and get the target's bearing information, but the bearing information is insufficient to locating the torpedo at present. So the method of locating torpedo by two platforms' bearing information is studied. Furthermore, several algorithms are established using EKF, UKF and SUKF methods. The results of simulation demonstrate that SUKF is applicable and has the characteristic of higher accuracy and the stronger convergence for passive locating and tracking torpedo application, which has large observation error. Finally, the application prospect of this method is presented.

A HPN(Hybrid Petri Nets) traffic model, composed of discrete Petri Nets and TdPN(Timed Petri Nets), is proposed for modeling dynamic traffic flows at non-signalized T-intersections in urban traffic networks. This model is used to solve two primary problems. One is how to formulate dynamic traffic flows at intersections. Another is how to simulate their behavior accurately and continuously. The discrete Petri Nets represent the Levels of traffic priority, and the TdPN computes the corresponding time-varying quantities. The simulation results indicate that the model can represent the characteristics of dynamic traffic flows at non-signalized T-intersections and meet the requirements for modeling the urban transportation networks.

A new approach of fast color character extraction was proposed. Clustering algorithm was adopted in our method to differentiate between objective character regions and back- ground regions on the premise that character regions are nearly monochromatic. However, the key point of this approach was how to select suitable elements' features based upon the original image information and character's information; it directly determined extraction results. We brought forward a means for feature selection to resolve clustering's key problem, and proved the means is effective. Furthermore, in order to overcome the non-ignorable computational overhead of clustering, parallel clustering algorithm was designed and realized by CUDA. Experiments illustrated color character could be extracted satisfactorily even if these characters are surrounded by complex background; meanwhile, parallel improved method that executed by assist of GPU was more efficient than former method that executed by CPU only.

In this paper, the fracture problem for a functionally graded coating with a Kelvin-type viscoelastic interface is studied. The functionally graded piezoelectric layer contains an electrically permeable anti-plane crack that is parallel to the interface. By utilizing the integral transformation and singular integral equation techniques, the crack problem is solved numerically. The effects of the physical, graded parameter and geometrical parameters on the stress intensity factor (SIF) are investigated. With time elapsing, the effect of viscoelastic interface becomes weaker and weaker. The effect of the viscoelastic interface on the fracture behavior of the functionally graded piezoelectric coating also depends on the substrate thickness.

Aiming at the complex dynamic feature of large ship, an intelligent control structure based on Library-similar Knowledge-increasable Neural Network Group is presented. This compounded control structure using the dynamic knowledge-increasable learning capability of the neural network groups, solve the problems of online identification and online design of the controller, so that the high precise output tracking control of uncertain nonlinear large ship can be realized. Simulating results show that it is feasible and effective.

Aiming at the complex dynamic feature of large ship, an intelligent control structure based on Library-similar Knowledge-increasable Neural Network Group is presented. This compounded control structure using the dynamic knowledge-increasable learning capability of the neural network groups, solve the problems of online identification and online design of the controller, so that the high precise output tracking control of uncertain nonlinear large ship can be realized. Simulating results show that it is feasible and effective.

Aiming at the complex dynamic feature of large ship, an intelligent control structure based on Library-similar Knowledge-increasable Neural Network Group is presented. This compounded control structure using the dynamic knowledge-increasable learning capability of the neural network groups, solve the problems of online identification and online design of the controller, so that the high precise output tracking control of uncertain nonlinear large ship can be realized. Simulating results show that it is feasible and effective.

This paper deals with the knowledge approximation and rule acquisition based on variable precision rough sets in ordered information systems. The concepts of upward $\beta$-lower approximation, upward $\beta$-upper approximation, downward $\beta$-lower approximation and downward $\beta$-upper approximation based on dominance relation are introduced, and an approach to rule acquisition is also proposed in ordered decision tables with an illustrative example.

Flash memory is a kind of common storage device. Its characteristics of flexibility, low power, and so on offer excellent qualifications for embedded system and mobile system. But flash memory must be written after erasure operation, and the most important thing is that the erasure operation times are very limitable. For assurance of long time availability, data must be distributed over all memory space reasonably and politic, which brings forward challenge for storage system designer. This paper analyses the data structure and physical characteristics of typical flash memory. And a static trigger wear-leveling strategy based on classifying data with trigger condition is brought forward, called STWL. STWL forces these static data to move over all memory space according to the trigger condition so as to avoid some certain data blocks being damaged in advance. An experiment is carried out to simulate this strategy using VHDL. We construct a 4M bytes RAM as flash memory simulation model, a static wear-leveling unit to implement STWL and an excitation generation unit to yield memory store/load operations, As a result, the wear-leveling rate improves. 33% of space recycle times can be reduced and the biggest gap of number of erasing times of data block decreases from 883% to 38%.