Block Partition

14,000,000 Leading Edge Experts on the ideXlab platform

Scan Science and Technology

Contact Leading Edge Experts & Companies

Scan Science and Technology

Contact Leading Edge Experts & Companies

The Experts below are selected from a list of 2994 Experts worldwide ranked by ideXlab platform

Meng-lin Ku - One of the best experts on this subject based on the ideXlab platform.

  • Sparse Spectrum Sensing with Sub-Block Partition for Cognitive Radio Systems
    2014 IEEE 79th Vehicular Technology Conference (VTC Spring), 2014
    Co-Authors: Meng-lin Ku
    Abstract:

    Cognitive users are expected to be capable of exploring spectrum holes over a wide range of frequencies. Motivated by the sparse characteristic of underutilized spectrum, we consider sparse spectrum sensing using compressive sensing techniques for cognitive orthogonal frequency division multiplexing (OFDM) systems. The spectrum sensing problem is formulated as a multi-subcarrier detection problem, solved via the composite hypothesis testing and Neyman-Pearson criterion. Considering the availability of channel state information (CSI) at the cognitive device, two sparse spectrum sensing approaches are proposed for detecting the compressive received signals in time domain. For the purpose of complexity reduction, we further incorporate a sub-Block Partition scheme into the proposed approaches to leverage the spareness of the spectrum occupancy. The proposed approaches enable a flexible tradeoff between the implementation complexity and the sensing accuracy for wideband cognitive radios.

  • VTC Spring - Sparse Spectrum Sensing with Sub-Block Partition for Cognitive Radio Systems
    2014 IEEE 79th Vehicular Technology Conference (VTC Spring), 2014
    Co-Authors: Meng-lin Ku
    Abstract:

    Cognitive users are expected to be capable of exploring spectrum holes over a wide range of frequencies. Motivated by the sparse characteristic of underutilized spectrum, we consider sparse spectrum sensing using compressive sensing techniques for cognitive orthogonal frequency division multiplexing (OFDM) systems. The spectrum sensing problem is formulated as a multi-subcarrier detection problem, solved via the composite hypothesis testing and Neyman-Pearson criterion. Considering the availability of channel state information (CSI) at the cognitive device, two sparse spectrum sensing approaches are proposed for detecting the compressive received signals in time domain. For the purpose of complexity reduction, we further incorporate a sub-Block Partition scheme into the proposed approaches to leverage the spareness of the spectrum occupancy. The proposed approaches enable a flexible tradeoff between the implementation complexity and the sensing accuracy for wideband cognitive radios.

Zhilin Zhang - One of the best experts on this subject based on the ideXlab platform.

  • Recovery of Block sparse signals using the framework of Block sparse Bayesian learning
    2012 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP), 2012
    Co-Authors: Zhilin Zhang
    Abstract:

    In this paper we study the recovery of Block sparse signals and extend conventional approaches in two important directions; one is learning and exploiting intra-Block correlation, and the other is generalizing signals' Block structure such that the Block Partition is not needed to be known for recovery. We propose two algorithms based on the framework of Block sparse Bayesian learning (bSBL). One algorithm, directly derived from the framework, requires a priori knowledge of the Block Partition. Another algorithm, derived from an expanded bSBL framework using the generalization method, can be used when the Block Partition is unknown. Experiments show that they have superior performance to state-of-the-art algorithms.

Jian Song - One of the best experts on this subject based on the ideXlab platform.

  • Block Sparse Bayesian Learning-Based NB-IoT Interference Elimination in LTE-Advanced Systems
    IEEE Transactions on Communications, 2017
    Co-Authors: Fang Yang, Jian Song
    Abstract:

    Narrowband Internet-of-Things (NB-IoT) is one of the emerging 5G technologies, but might introduce narrowband interference (NBI) to existing broadband systems, such as long-term evolution advanced (LTE-A) systems. Thus, the mitigation of the NB-IoT interference to LTE-A is an important issue for the harmonic coexistence and compatibility between 4G and 5G. In this paper, a newly emerged sparse approximation technique, Block sparse Bayesian learning (BSBL), is utilized to estimate the NB-IoT interference in LTE-A systems. The Block sparse representation of the NBI is constituted through the proposed temporal differential measuring approach, and the BSBL theory is utilized to recover the practical Block sparse NBI. A BSBL-based method, Partition estimated BSBL, is proposed. With the aid of the estimated Block Partition beforehand, the Bayesian parameters are obtained to yield the NBI estimation. The intra-Block correlation (IBC) is considered to facilitate the recovery. Moreover, exploiting the inherent structure of the identical IBC matrix, another method of informative BSBL is proposed to further improve the accuracy, which does not require prior estimation of the Block Partition. Reported simulation results demonstrate that the proposed methods are effective in canceling the NB-IoT interference in LTE-A systems, and significantly outperform other conventional methods.

Howie H. Huang - One of the best experts on this subject based on the ideXlab platform.

  • HPEC - Fast Stochastic Block Partition for Streaming Graphs
    2018 IEEE High Performance extreme Computing Conference (HPEC), 2018
    Co-Authors: Ahsen J. Uppal, Howie H. Huang
    Abstract:

    The graph Partition problem continues to be challenging, particularly for streaming graph data. Although optimal graph Partitioning is NP-hard, stochastic methods can provide approximate solutions in reasonable time. However, such methods are optimized for static, not dynamic graph data. In this paper, we describe a new efficient algorithm we have developed for stochastic Block Partitioning on time-varying, streaming graph data. Our algorithm is a refinement of the baseline algorithm of the IEEE HPEC Graph Challenge [1]. Our incremental algorithm efficiently updates its previous internal state as new pieces are streamed in, and generates a complete Partition at every time step. Compared to the naive baseline which performs a complete Partitioning from scratch at every time step, our algorithm offers speedups between 1.96x for $\mathbf{N}=500$ and 3.56x for $\mathbf{N}=20\mathbf{k}$ overall, for a graph streamed over 10 parts, with similar accuracy. At the margin, the speedup in processing time for additional streaming pieces over the baseline is between 7.1x for $\mathbf{N}=500$ to 25.1x for $\mathbf{N}=20\mathbf{k}$ .

  • Fast Stochastic Block Partition for Streaming Graphs
    2018 IEEE High Performance extreme Computing Conference (HPEC), 2018
    Co-Authors: Ahsen J. Uppal, Howie H. Huang
    Abstract:

    The graph Partition problem continues to be challenging, particularly for streaming graph data. Although optimal graph Partitioning is NP-hard, stochastic methods can provide approximate solutions in reasonable time. However, such methods are optimized for static, not dynamic graph data. In this paper, we describe a new efficient algorithm we have developed for stochastic Block Partitioning on time-varying, streaming graph data. Our algorithm is a refinement of the baseline algorithm of the IEEE HPEC Graph Challenge [1]. Our incremental algorithm efficiently updates its previous internal state as new pieces are streamed in, and generates a complete Partition at every time step. Compared to the naive baseline which performs a complete Partitioning from scratch at every time step, our algorithm offers speedups between 1.96x for N=500 and 3.56x for N=20k overall, for a graph streamed over 10 parts, with similar accuracy. At the margin, the speedup in processing time for additional streaming pieces over the baseline is between 7.1x for N=500 to 25.1x for N=20k.

  • HPEC - Scalable stochastic Block Partition
    2017 IEEE High Performance Extreme Computing Conference (HPEC), 2017
    Co-Authors: Ahsen J. Uppal, Guy Swope, Howie H. Huang
    Abstract:

    The processing of graph data at large scale, though important and useful for real-world applications, continues to be challenging, particularly for problems such as graph Partitioning. Optimal graph Partitioning is NP-hard, but several methods provide approximate solutions in reasonable time. Yet scaling these approximate algorithms is also challenging. In this paper, we describe our efforts towards improving the scalability of one such technique, stochastic Block Partition, which is the baseline algorithm for the IEEE HPEC Graph Challenge [1]. Our key contributions are: improvements to the parallelization of the baseline bottom-up algorithm, especially the Markov Chain Monte Carlo (MCMC) nodal updates for Bayesian inference; a new top-down divide and conquer algorithm capable of reducing the algorithmic complexity of static Partitioning and also suitable for streaming Partitioning; a parallel single-node multi-CPU implementation and a parallel multi-node MPI implementation. Although our focus is on algorithmic scalability, our Python implementation obtains a speedup of 1.65× over the fastest baseline parallel C++ run at a graph size of 100k vertices divided into 8 subgraphs on a multi-CPU single node machine. It achieves a speedup of 61× over itself on a cluster of 4 machines with 256 CPUs for a 20k node graph divided into 4 subgraphs, and 441× speedup over itself on a 50k node graph divided into 8 subgraphs on a multi-CPU single node machine.

  • Scalable stochastic Block Partition
    2017 IEEE High Performance Extreme Computing Conference (HPEC), 2017
    Co-Authors: Ahsen J. Uppal, Guy Swope, Howie H. Huang
    Abstract:

    The processing of graph data at large scale, though important and useful for real-world applications, continues to be challenging, particularly for problems such as graph Partitioning. Optimal graph Partitioning is NP-hard, but several methods provide approximate solutions in reasonable time. Yet scaling these approximate algorithms is also challenging. In this paper, we describe our efforts towards improving the scalability of one such technique, stochastic Block Partition, which is the baseline algorithm for the IEEE HPEC Graph Challenge [1]. Our key contributions are: improvements to the parallelization of the baseline bottom-up algorithm, especially the Markov Chain Monte Carlo (MCMC) nodal updates for Bayesian inference; a new top-down divide and conquer algorithm capable of reducing the algorithmic complexity of static Partitioning and also suitable for streaming Partitioning; a parallel single-node multi-CPU implementation and a parallel multi-node MPI implementation. Although our focus is on algorithmic scalability, our Python implementation obtains a speedup of 1.65× over the fastest baseline parallel C++ run at a graph size of 100k vertices divided into 8 subgraphs on a multi-CPU single node machine. It achieves a speedup of 61× over itself on a cluster of 4 machines with 256 CPUs for a 20k node graph divided into 4 subgraphs, and 441× speedup over itself on a 50k node graph divided into 8 subgraphs on a multi-CPU single node machine.

Xiaolong Li - One of the best experts on this subject based on the ideXlab platform.

  • improved reversible visible watermarking based on adaptive Block Partition
    International Workshop on Digital Watermarking, 2017
    Co-Authors: Guangyuan Yang, Wenfa Qi, Xiaolong Li
    Abstract:

    Visible watermarking is a useful technique to perceptually protect the copyright while the reversible technique can help losslessly recover the original image. A reversible visible image watermarking scheme based on difference-expansion and adaptive Block Partition is presented in this paper. First, the cover image is divided into non-overlapped \(k \times k\) sized Blocks. Then, an adaptive visual effect factor for each Block is calculated by non-watermarked Blocks and estimated watermarked Blocks in its neighborhood to embed the visible watermark. Since the information in watermarking region is not used, authorized users can exactly recover the original cover image without the availability of the watermark in the recovery process. Afterwards, one watermark bit is embedded into each Block based on the conventional difference-expansion method. To reduce the exceeding number which denotes the pixel whose values is larger than 255 or less than 0 generated in watermark bit embedding procedure, an adaptive Block Partition strategy is utilized in the proposed method. Experimental results show that compared with the related work, the proposed method can greatly reduce the exceeding numbers and the visual effect is better at the same time.

  • IWDW - Improved Reversible Visible Watermarking Based on Adaptive Block Partition
    Digital Forensics and Watermarking, 2017
    Co-Authors: Guangyuan Yang, Wenfa Qi, Xiaolong Li
    Abstract:

    Visible watermarking is a useful technique to perceptually protect the copyright while the reversible technique can help losslessly recover the original image. A reversible visible image watermarking scheme based on difference-expansion and adaptive Block Partition is presented in this paper. First, the cover image is divided into non-overlapped \(k \times k\) sized Blocks. Then, an adaptive visual effect factor for each Block is calculated by non-watermarked Blocks and estimated watermarked Blocks in its neighborhood to embed the visible watermark. Since the information in watermarking region is not used, authorized users can exactly recover the original cover image without the availability of the watermark in the recovery process. Afterwards, one watermark bit is embedded into each Block based on the conventional difference-expansion method. To reduce the exceeding number which denotes the pixel whose values is larger than 255 or less than 0 generated in watermark bit embedding procedure, an adaptive Block Partition strategy is utilized in the proposed method. Experimental results show that compared with the related work, the proposed method can greatly reduce the exceeding numbers and the visual effect is better at the same time.