多媒体信号在相对不稳定和带宽有限的无线网络环境下,易出现丢包等传输问题,同时码率受限,影响接收端多媒体信号感官质量。图像压缩感知作为一种结合采样和压缩的信号处理理论,具有编码简单,解码复杂的结构特性,适用于能耗受限的传感器无线网络中的多媒体信号采集与传输问题。但是,现有图像压缩感知技术尚存在压缩效率有限,恢复质量不高等问题,同样易受无线传输环境丢包影响。文中针对图像压缩感知压缩效率和无线传输环境下传输质量受限问题,通过运用图像视觉显著性信息判决技术和路径分集技术,对图像关键区域少量的可分级信息冗余增加,进行非对称的信道保护策略,来保障图像中视觉关注较高区域恢复质量。文中提出的基于显著性信息的压缩感知图像可分级编码方法,在无差错情况下,率失真性能优于传统无显著性信息方法,在丢包网络环境下,优于CS-MDC算法。

Compressive sensing (CS) is a sub-Nyquist sampling way while still enabling exact reconstruction, which is applicable to WMSN. In this paper, based on the characteristic of CS video in WMSN, we proposed a hierarchical objective CS video quality assessment (HOCSVQA) approach to get CS video quality index (CSVQI) from three levels, measurement level, stream level, and packet level, respectively. This approach cannot only keep the convenience and real-time characteristic of objective video assessment, but also reflect the QoE to a certain extent due to the coefficients regressed from subjective video assessment experiments. A set of experiments on subjective CS video quality assessment and another set of verification experiments are designed and settled. The CS video quality index, CSVQI, assessed by the model we proposed maintained a high correlation with data from verification experiments under statistical correlation measure.

压缩感知和矩阵填充是当前的两个研究热点,压缩感知的性能取决于3个要素:信号的稀疏性、压缩感知矩阵的非相干性和重构算法的快速有效性。相应地,矩阵填充性能也取决于3个要素:矩阵的低秩性、矩阵的不相关性和重构算法的快速有效性。文中首先论述了压缩感知和矩阵填充的应用背景,阐述了两者的数学模型,分析了信号的稀疏性和观测矩阵的不相关性对压缩感知性能的影响,研究了矩阵的低秩和不相关性在矩阵填充中的作用,进而对压缩感知和矩阵填充的稀疏性和非相干性进行了对比,总结了压缩感知和矩阵填充的重构算法,介绍了压缩感知和矩阵填充在图像处理中的应用。

矩阵填充是一类稀疏先验下的不适定的反问题.首先阐述了矩阵填充的基本原理,指出只有当待求的矩阵满足不相关特性或秩-受限等距特性时,才有可能精确重构未知矩阵.Jain P等将矩阵的不相关特性与秩-受限等距特性联系起来,提出了仿射-受限等距特性,但没有说明秩-受限等距常数与不相关常数的关系,定理3给出了两者之间的的数值关系,有效的促进了矩阵填充的研究.接下来分析了矩阵填充中常用的几种算法,并在随后的仿真实验中对这几种算法的重构性能做了详细比较.

This paper describes the background of matrix completion firstly,points out the mathematics model of matrix completion,analyzes the low rank property and the incoherence property in matrix completion. Mainly introduces three reconstruction algorithm commonly used in matrix completion: SVT( Singular Value Thresholding) 、ADMiRA( Atomic Decomposition for Minimum Rank Approximation) and SVP( Singular Value Projection),compares their reconstruction performance in this paper. Secondly,we analyze the connection between matrix completion and compressed sensing. Finally we introduce the application of matrix completion in collaborative filtering,system identification,sensor network,image processing,sparse channel estimation,spectrum sensing and multimedia coding and communication.

对经济学方法在无线资源管理中的应用进行了研究,考虑业务、用户、资源等多个域,将无线资源分配看作生产–消费模型,兼顾用户公平性原则,针对不同业务的QoS（quality of service）要求采用不同的资源分配方法,建立了基于社会福利最大化的资源分配模型。采用基于用户柔性业务的调度算法优化所提模型,综合考虑用户效用、网络效益以及运营商收益,实现了基于社会福利最大化的柔性业务资源分配。仿真结果验证了所提算法的优越性。

Matrix completion is the extension of compressed sensing. In compressed sensing, we solve the underdetermined equations using sparsity prior of the unknown signals. However, in matrix completion, we solve the underdetermined equations based on sparsity prior in singular values set of the unknown matrix, which also calls low-rank prior of the unknown matrix. This paper firstly introduces basic concept of matrix completion, analyses the matrix suitably used in matrix completion, and shows that such matrix should satisfy two conditions: low rank and incoherence property. Then the paper provides three reconstruction algorithms commonly used in matrix completion: singular value thresholding algorithm, singular value projection, and atomic decomposition for minimum rank approximation, puts forward their shortcoming to know the rank of original matrix. The Projected Gradient Descent based on Soft Thresholding (STPGD), proposed in this paper predicts the rank of unknown matrix using soft thresholding, and iteratives based on projected gradient descent, thus it could estimate the rank of unknown matrix exactly with low computational complexity, this is verified by numerical experiments. We also analyze the convergence and computational complexity of the STPGD algorithm, point out this algorithm is guaranteed to converge, and analyse the number of iterations needed to reach reconstruction error. Compared the computational complexity of the STPGD algorithm to other algorithms, we draw the conclusion that the STPGD algorithm not only reduces the computational complexity, but also improves the precision of the reconstruction solution.

自1998年互联网工程任务组(IETF)提出下一代互联网标准规范以来,IPv6已经历了十多年的发展.现今已有越来越多的IPv6产品被投入到了开发与应用中.而如何提高不同产品间的互通性和可靠性则成为了一个关键问题.进行协议一致性测试是提高IPv6实现可靠性的一种有效方式.本文就重点针对IPv6邻居发现协议进行了一致性测试分析.本文首先简要分析了IPv6邻居发现协议的主要功能及实现原理,并据此抽象出其有限状态机(FSM)模型.进而结合一种现有基于有限状态机(FSM)的一致性测试序列改进算法生成了该协议的抽象测试序列.本文在最后对得到的测试序列进行了有效性和可靠性分析,分析表明,使用该算法得到的测试序列不仅在序列长度上较传统UIO序列法有了明显的缩短,同时对测试过程中可能发生的输出错误及末状态转换错误也具备良好的检测能力.本文获得的抽象测试序列可对相关IPv6协议开发者提供有效参考.

To reach necessary end-to-end connectivity between the Internet and wireless sensor networks (WSNs), the Internet Engineering Task Force (IETF) IPv6 over low power wireless personal area network (6LowPAN) working group has been established and introduced an adaptation layer for integration of IEEE 802.15.4 physical layer/media access control (PHY/MAC) layers and the upper layers of any Internet protocol (IP)-based networks, such as the Internet. The energy efficiency is one of the most important performance measures in WSNs because most sensor nodes are only battery powered so we should reduce the energy consumption to the lowest to extend the life of nodes. Therefore the determination of MAC frame length should be carefully considered since that the radio frequency (RF) module consumes most the energy of a sensor node meanwhile the MAC protocol is the direct controller of RF module. In this paper, we provide a star-shaped 6LowPAN non-beacon mode with unslotted carrier sense multiple access with collision avoidance (CSMA/CA) mechanism to access to the channel and model the stochastic behavior of a target end node as the M/G/1 queuing system. Analytical expressions for some parameters such as channel busy probability, packet loss probability and energy efficiency are obtained in this paper and our analytical results can clearly show the impact of MAC frame length on the energy efficiency of a target node in both ideal and lossy channel.

A novel framework called distributed compressive video sensing (DCVS), combining distributed video coding (DVC) and compressive sensing (CS), directly capture the raw video data as measurements with low-complexity and low-cost process. It meets the requirements of distributed system very well, because of its resource consumption shifting from encoder to decoder. Nevertheless, the issue of measurements transmission in bit error channel has not been considered yet in the previous work of DCVS. This paper improved the existing DCVS codec scheme by adding the quantization and inverse quantization process, and proposed a parity-based error control (PEC) method. This method is simple enough, and has high coding efficiency. The proposed method is shown to increase video recovery quality greatly under binary symmetric channel.

Based on compressive sampling transmission model, we demonstrate here a method of quality evaluation for the reconstruction images, which is promising for the transmission of unstructured signal with reduced dimension. By this method, the auxiliary information of the recovery image quality is obtained as a feedback to control number of measurements from compressive sampling video stream. Therefore, the number of measurements can be easily derived at the condition of the absence of information sparsity, and the recovery image quality is effectively improved. Theoretical and experimental results show that this algorithm can estimate the quality of images effectively and is in well consistency with the traditional objective evaluation algorithm.

Sparse representation,incorrelate projection and reconstruction are the three elements of compressed sensing,This paper uses Gaussian random matrix as original matrix,and adaptively transforms using the partial positional information of sparse coefficients,then forms a new adaptive measurement matrix.When the compressed sensing matrix projects the sparse coefficients,the small coefficients are more close to zero;at the same time,we decrease the measured values by reducing the columns of measurement matrix,thus the difference between the amount of data transmission using adaptive measurement matrix and the amount of data transmission using Gaussian random measurement is little.The improved performance of compressed sensing employed adaptive measurement matrix embodies in the reconstruction accuracy.When we use iterative hard thresholding as reconstruction algorithm;both theory and experiment verify the performance of adaptive measurement matrix better than Gaussian random measurement matrix.

Matrix completion is the extension of compressed sensing, which uses the prior of low rank to recover original matrix. This paper introduces several reconstruction algorithms (SVT, ADMiRA and SVP) firstly, put forward their shortcomings to know the rank of original matrix, and propose our method to predict the observed samples under unknown rank of original matrix.

In the field of collection,communication and analysis of the information in the wireless sensor networks,the accuracy of the information can be improved,and the node power consumption can be reduced by using cooperative communication techniques. The selection of cooperative partners is an important issue worthy of study.consider making use of the information feedback technology into wireless multimedia sensor networks(WMSN),to feed the Video Stream Quality Index(VSQI) back to sink node through the control channel from geographically distributed sensor nodes for supporting on the selection of collaboration node.In this paper,according to human visual characteristics,a video streaming quality estimation method is proposed,and signal VSQI is designed on the basis of that.Moreover,Compressed Sensing(CS) theory is applied to VSQI compression feedback in the experiment because of the large quantity of feedback information.The random measurement matrix associated with VSQI signal,and reconstruction guidelines is tested in the simulation experiment.The simulation results demonstrate that CS based Compressive feedback method can reduce the processing complexity of sensing node side and enhance the degree of compression,and provide the theoretical support for selection the collaboration node.