The existing TCP mechanisms can not effectively distinguish the cause of packet loss in wireless networks whether it is triggered by congestion or deafness. In order to solve the problem, this paper proposes a Cross-layer Explicit Deafness Notification(CLEDN) in wireless Mesh network. It adopts Media Access Control(MAC) based on busy tones when system uses directional antennas, and an explicit notification frame is designed between IP and TCP, which is passed to source node through the user-defined multi-hop forwarding rule. The mechanism is applied to theoretical deduction and analysis in linear wireless Mesh networks, and numerical analysis verifies its feasibility and validity.
Aiming at the micro-mobility feature of Mobile Ad hoc Network(MANET) nodes, micro-mobility management strategy is presented in the design of MANET gateway which employs the Ad hoc On-demand Distance Vector(AODV) routing protocol. Micro-mobility management is applied to the route sub-layer of network layer in MANET and implemented in AODV routing protocol. Simulations result of NS2 platform shows that the MANET gateway which employs the micro-mobility management strategy can reduce the end-to-end delay and the consumption of control packets of MANET accessing the Internet effectively.
In order to get new file hidden points in drive-level of system, the principle of File System Filter Driver(FSFD) and the structure of driver stack location are analyzed. Through making some changes in driver stack location’s structure and CompletionRoutine, besides modifying the I/O Requst Packet(IRP) delivery method, two methods to hide files are implemented. Hidden files using these methods achieve depth hide. They can not be queried by system or be accessed through normal channels.
This paper proposes the design and implementation of a System on Programmable Chip(SoPC) interface to solve the problems of timing control and bit stream downloading of embedded Field Programmable Gate Array(FPGA) IP core in the evolvable SoPC. The FPGA interface uses asynchronous FIFO and dual port RAM structure, scalable read/write command transmission to achieve asynchronous communication between FPGA IP core and the system. Embedded CPU achieves the on-chip bit stream configuration of FPGA IP core via FPGA interface. Hardware random number generator in the FPGA interface is used to accelerate the evolutionary algorithm. Automatic verification platform and FPGA prototyping platform are used for the verification of FPGA interface to achieve convergence of verification. Test results show that FPGA interface implements the communication between embedded CPU and FPGA IP core successfully, completes on-chip evolution.
A new visual attention model used for rapid perception of complex targets in natural scene is proposed. In the learning process, the model extracts saliency blobs from a given target’s image. Then during the process of attention on a scene image, it adopts a blob searching and merging strategy based on graph matching to guide visual focus to where it looks like the target. The blob searching and merging strategy uses features of learned heterogeneous blobs and their spatial relative positions, which are all recorded during the previous target learning process. Compared with typical bottom-up visual attention model, experiments show that the new method could efficiently introduce feature and structure information of complex target into the process of attention, reduce useless visual focus shifts, and improve the performance of visual attention. The model could be used to locate complex structural targets in natural scene images.
To solve the problem of the data transmission speed mismatch between I/O device and CPU, this paper analyzes the efficiency of various buffer structure based on computer architecture. With QuartusII which is Computer-Aided Design(CAD) using Electronic Design Automation(EDA) software, it designs the asynchronous dual-clock First In First Out(FIFO) buffer, including the buffer structure of the simulation, data recording, and analysis of its effectiveness. Analysis results show that the efficiency of computer system is closely related with that of FIFO. To improve the efficiency of computer system, the efficiency of FIFO is supposed to be improved first. At the same time, the improvement of the efficiency of FIFO is based on the improvement of the efficiency of computer system.
In order to detect the content errors which conform to syntax standard while video contents have changed through 3G wireless networks, an error detection algorithm based on compressed domain and pixel domain spatial correlation for H.264 is proposed. The algorithm uses the similarity of neighboring Direct Current(DC) coefficients in compressed domain and neighboring reconstruct pixels in pixel domain to detect whether there is an error in one macroblock. Experimental result shows that, under different error rate and different streams, the proposed algorithm can detect content errors effectively and improve the reconstruct image quality significantly.
Traditional cache replacement strategy lacks pertinence to the semantic cache, and it can not reuse the data in cache effectively, it affects the performance of the cache. This paper proposes an eXtensible Markup Language(XML) query cache replacement strategy based on semantic contribution value for the characteristics of XML query algebra. It processes cluster analysis according to the cache item of user history query, forecasts the semantic contribution values of future query, and replaces the item which has minimum value of semantic contribution in insufficient space. Experimental results show that this strategy can effectively shorten the inquires time, and improves cache hit than both Least Frequency Used(LFU) and Least Recently Used(LRU).
In order to improve the accuracy of Chinese off-line handwriting recognition, this paper proposes a method based on magnitude and relative phase information of Pyramidal Dual-tree Directional Filter Bank(PDTDFB). Derived Magnitude Distribution(DMD) and relative phase depicted by Von Mises(VM) distribution and Wrapped Cauchy(WC) distribution model are introduced to reflect the features of image in detail, and Resistor-average(RA) distance is applied. According to the idea of Multiple Attribute Decision Making(MADM), the two kinds of features extracted from magnitude and relative phase are fused in the algorithm. Experimental results show that in comparison to Generalized Gaussian Density(GGD) model with PDTDFB, the accuracy of the method increases by 2.2%.
Aiming at longer time and memory pages repeated retransmission in the process of memory pre-copy, this paper optimizes the mechanism of memory pre-copy migration and uses Markov prediction model to improve the algorithm that reckon the working set of memory dirty page, designs an new algorithm that calculate working set of memory dirty page by forecasting the probability of dirty pages. This algorithm calculates probability of being modified next round of iteration using dirty pages history of the operation visits, only the memory pages with a lower probability can be translated. Experimental results show that new algorithm shortens the total time of migration and downtime, and effectively support dynamic migration of Virtual Machine(VM).
In order to reduce the computation of the minimum interclass variance algorithm in the remote sensing image segmentation, the equivalent of the objective functions is proved based on the principles of the minimum variance algorithm and the k-means clustering algorithm. In addition, a new fast minimum variance algorithm based on the k-means optimization is proposed. Experimental results show that it effectively reduces the hunting zone and has a real-time speed to calculate the optimal threshold.