Robust adaptive beamforming for large-scale arrays

For a large-scale adaptive array, heavy computational load and high-rate data transmission are two challenges in the implementation of an adaptive digital beamforming system. Moreover, the large-scale array becomes extremely sensitive to array imperfections. First, based on a restructured recursive linearly constrained minimum variance algorithm and a gradient-based optimization method, a new robust recursive linearly constrained minimum variance (RRLCMV) algorithm is proposed in this paper. The computational load of the RRLCMV algorithm is on the order of o(N), which is less than that of the conventional gradient-based robust adaptive algorithm. Then, a new efficient parallel robust recursive linearly constrained minimum variance (PRRLCMV) adaptive algorithm is proposed by appropriately partitioning the RRLCMV algorithm into a number of operational modules. It can be easily executed in a distributed-parallel-processing fashion, sequentially and in parallel. As a result, the PRRLCMV algorithm provides an effective solution that can alleviate the bottleneck of high-rate data transmission and reduce the computational cost. Finally, an implementation scheme of the PRRLCMV algorithm based on a distributed-parallel-processing system is also proposed. The simulation results demonstrate that the new PRRLCMV algorithm can significantly reduce the degradation due to various array errors.     

二维多源测向的实时处理算法及实现

研究用二维MUSIC算法，解决阵列处理二维多源测向的快速实现问题。利用阵列流形具有的对称性，减少阵列流形和空间谱的计算量。在判定出信号源个数后，有选择地使用信号子空间或噪声子空间去计算空间谱，进一步减少计算量。采用2片ADI公司的高性能DSP芯片（ADSP21160），用并行处理方法实现了二维多源测向的实时处理。     

基于多核DSP激光成像雷达数据处理系统

采用多核DSP设计了一个用于地面目标检测的激光雷达实时图像处理系统。在详细分析算法各模块资源消耗量的基础上,完成了硬件电路设计,实现了以主辅拓扑结构为框架的软件并行处理系统开发。在系统实现时,先将图像进行分区,并合理地将分区后的图像分配到四个DSP核中进行处理。最后,将并行系统进一步扩展到双核和六核,并与单核系统进行性能比较。对算法运算时间的测试结果表明,系统处理一帧图像仅需50 ms达到了实时性要求。结果表明,对于固定负载的处理系统,单纯地通过增加并行的核数来提高加速比的幅度是有限的。当增加并行的核数已不能明显地提高计算效率时,在系统设计中应着重减少每个核串行运算的负载量。     

Reduced Complexity Algorithm for Spreading Sequence Design

This paper describes a novel gradient descent algorithm for the constrained optimization of spreading sequence design in the uplink of CDMA systems that use codeword adaptation. We prove that the proposed algorithm converges to the optimal sequence. In addition, the paper demonstrates the convergence of the parallel distributed optimization which is confirmed by simulations. Parallel optimization ensures faster convergence and thus reduces computational load. The performance of the parallel algorithm is improved as the number of active users increases.      

Scheduling algorithm for stochastic job stream cipher service based on load balancing

Business stream parallel processing system face the situation of the diversity of cipher service requests,the cross of serial mode and parallel mode,the intercross of different business data flow,and the demand of high speed and high reliability in security field.In order to improve the parallel processing efficiency of multi-cryptographic algorithm,load balancing was used as scheduling objective.Based on hierarchical hardware scheduling method (HHS-ACDID),considering the processing node’s storage capacity and processing speed,a load balancing scheduling algorithm was approved to support non related tasks and related tasks at the same time,which achieves the high speed cipher processing throughput.Simulation results show that the algorithm can complete dynamic scheduling of data stream system and get better load balancing effect.Compared with HHS-ACDID,the efficiency of the algorithm is improved by about 12%.     

并行射线跟踪算法及其在城市电波预测的应用

射线跟踪算法的计算量较大,耗时较长.针对这一问题提出并研究了对等模式和主从模式两类并行射线跟踪算法,结合MPI并行运算函数库,在局域网计算机簇中,实现了城市复杂微蜂窝环境的电波预测.结果表明,并行算法在精度相同的情况下,大大缩减了计算机运行时间.而且,主从模式在异类网络中具有更好的并行加速增益和负载均衡.     

Design and analysis of parallel file downloading algorithms in peer-to-peer networks

It is well known that the method of parallel downloading can be used to reduce file download times in a peer-to-peer (P2P) network. There has been little investigation on parallel download and chunk allocation for source peers with random service capacities. The main contribution of this paper is to address the problem of efficient parallel file download in P2P networks with random service capacities. A precise analysis of the expected download time is given when the service capacity of a source peer is a random variable. A general framework is developed for analyzing the expected download time of a parallel download and chunk allocation algorithm, and is applied to the analysis of several algorithms. Two chunk allocation algorithms for parallel download are proposed. It is observed that the performance of parallel download can be significantly improved by using the method of probing high-capacity peers. One such algorithm is proposed and its expected parallel download time is analyzed. The performance of these parallel file download algorithms in P2P networks with random service capacities are compared. The above parallel download algorithms are extended to multiple file download by dividing source peers into clusters. It is noticed that there is an important issue of optimal parallelism which minimizes the combined effect of intracluster and intercluster overhead of parallel download and load imbalance.     

云环境下一种基于任务类型匹配的负载均衡策略研究

针对云计算环境下大量并行计算节点容易产生计算节点之间的负载不均问题,本文提出了一种基于任务类型匹配的负载均衡方案。该方案针对任务集中的多种不同长度的子任务类型情况进行判定,并对当前主流的Max-Min和Min-Min两种启发式负载均衡算法进行分析,综合其优缺点,并针对任务集的类型采用不同的算法进行任务调度。实验结果表明在该负载均衡的策略下,提出的方案具有比单一应用Max-Min或者Min-Min算法具有更好的负载均衡特性和更短的完成时间。     

Parallel algorithm for image rendering and its implementation

The authors present a parallel algorithm for fast rendering. A set of zones which can be concurrently processed is identified; this set is split into a number of subsets equal to the number of available processors. Each processor renders, in parallel with the others, a zone of its set     

An efficient decoding algorithm for QO-STBCs based on iterative interference cancellation

We propose an efficient decoding algorithm targeted for quasi-orthogonal space-time block codes which can be viewed as an iterative extension of the parallel detection algorithm using iterative interference cancellation. The essence of the algorithm is in the realization of the fact that the number of candidate symbols at the end of each interference cancellation iteration reduces very quickly, and that no more than two iterations are required.     

8/4-4分载式并联六维力传感器建模与分析

为增大传感器量程,适应大负载要求,基于8/4-4并联压电六维力传感器,提出了一种8/4-4分载式并联结构的六维力传感器。基于大力分载原理,在8/4-4并联压电六维力传感器的基础上,各个支链采用压电陶瓷作为敏感元件,增加中心承载梁,使其分载传感器的大部分外负载广义六维力。建立8/4-4分载式并联六维力传感器的数学模型,并推导其解耦算法。通过ADAMS算例仿真与理论模型进行对比,六维力传感器的数学模型及解耦算法的误差不超过0.14%,验证了其准确性。     

用于非圆信号二维测向的扩展Root-MUSIC算法

利用双平行线阵的阵列结构,提出了用于非圆信号二维测向的扩展root-MUSIC(ERM)算法.ERM算法估计得到的俯仰角与方位角一一对应,自动配对,无需搜索,其可测向信号数大于子阵阵元数,方位测角精度优于同样基于双平|亍线阵的二维测向波达方向矩阵法(DOAM),俯仰测角精度在低信噪比下优于DOAM算法,俯仰角兼并时算法仍有效.     

Optimum Allocation of Components in a Special 2-Port Network

A 2-port network is designed with i.i.d. components and has a) i.i.d. series stages, each with parallel components or b) i.i.d parallel strings, each with series components (dual network); components can fail in either an open or short mode. The network-success requirement is that its input-output terminals be neither open nor short. A computer algorithm is given which determines a) the number of required components from the available number of components, b) the optimum number of stages or parallel strings, and c) the optimum number of components per stage or string, to maximize system reliability. The case of an unlimited number of components is considered in connection with the design of the network to meet a prescribed level of reliability.     

A new parallel binary image shrinking algorithm

A new parallel binary image shrinking algorithm that can be considered as an improvement of Levialdi's (1972) parallel shrinking algorithm and can be used in many image labeling algorithms as a basic operation in order to reduce the storage requirements for local memory and speed up the labeling process is presented. This new parallel shrinking algorithm shrinks an nxn binary image to an image with no black pixels in O(n) parallel steps with a multiplicative constant of 1.5, preserving the 8-connectivity in the shrinking process.     

A distributed algorithm for fault simulation of combinatorial and asynchronous sequential digital designs, utilizing circuit partitioning, on loosely-coupled parallel processors

Fault simulation constitutes an indispensable tool in ensuring the correctness and quality of manufactured digital designs. Traditional uniprocessor based algorithms for fault simulation have been observed to be agonizingly slow for today's large and complex digital designs. More recently, a few researchers introduce an approach, as evident in the literature, wherein the fault set is partitioned and the digital design fault simulated for each of the fault subset on separate processors of a parallel processor system. The approach is limited in that it continues to utilize the traditional uniprocessor-based algorithm and the performance results are not encouraging. This paper introduces, perhaps for the first time, a distributed algorithm that is capable of fault simulating both combinational and asynchronous sequential digital designs on parallel processors. An underlying assumption of the algorithm is that the digital design, under fault simulation, is partitioned by the user. In this approach, referred to as NODIFS in this paper, every component in the circuit is modeled as an asynchronous, concurrent, entity that is fault simulated as soon as appropriate signal transitions and fault lists are asserted at its input ports. The circuit partitioning is such that components of every partition are allocated to a unique processor of the parallel processor system. Consequently, a number of components may be concurrently fault simulated on multiple processors in NODIFS implying significant increase in throughput. This approach promises (i) very high throughput because of its ability, in principle, to utilize the maximal inherent parallelism, and (ii) scalability. The algorithm is novel in that the overall task of decision-making i.e., fault simulation of the circuit, is distributed into a number of natural, independent, and concurrent entities that execute asynchronously to utilize maximal parallelism. NODIFS's success is the result of the asynchronous distributed discrete-event simulation algorithm, YADDES, and a new approach to fault simulation. The notion of scalability implies that where the problem size increases, the algorithm continues to apply and, by increasing the number of computational engines proportionately, the performance of the algorithm will continue to increase. Furthermore, NODIFS is a natural choice for fault simulation of digital designs at the behavior-level — an eventual reality, wherein the ratio of the computational to communication load for the behavior models may approach a significantly large value. This paper also reports on an implementation of NODIFS on both the ARMSTRONG parallel processor system at Brown University and the performance results indicate significant increase in the speedup for a few representative example digital designs. It is stressed that the representative digital designs serve to support the mathematical validity of the algorithm along with the proof of correctness and not to demonstrate a high-performance, commercial, industrial-quality fault simulator.     

Parallel canonical recoding

The author introduces a parallel algorithm for generating the canonical signed-digit expansion of an n-bit number in O(log n) time using O(n) gates. The algorithm is similar to the computation of the carries in a carry look-ahead circuit. It is also proven that if the binary number x+[x/2] is given, then the canonical signed-digit recoding of x can be computed in O(1) time using O(n) gates     

一种有效的Winograd付里叶变换并行算法

本文提出一种有效的Winograd付里叶变换并行算法,该算法具有处理器间负载平衡,并行计算效率高等特点。     

基于JADE和平行因子分解的欠定混合盲辨识算法

针对源信号个数未知的欠定混合盲源分离问题,本文提出了一种基于特征矩阵联合近似对角化（Joint Approximate Diagonalization of Eigenmatrices, JADE）和平行因子分解的欠定混合盲辨识算法,该算法不需要源信号满足稀疏性要求,仅在源信号满足相互独立和最多一个高斯信号的条件下,通过将JADE算法中的样本四阶协方差矩阵叠加成三阶张量,再对此三阶张量进行平行因子分解来完成源信号数和混合矩阵的估计,由于平行因子分解的唯一辨识性在欠定条件下仍然满足,该算法能够解决欠定盲源分离问题。并对该欠定混合盲辨识算法进行了深入的分析。通过仿真实验,计算估计矩阵与混合矩阵的平均相关误差,结果表明本文提出的算法在适定和欠定混合时均具有很好的辨识效果,而且实现简单,可满足实际应用的要求。      

异构无线网络中基于业务转移和接入控制的混合负载均衡

 针对现有异构无线网络负载均衡方法未能综合考虑重载网络业务转移和新业务接入控制的问题,提出了一种混合负载均衡算法.该算法首先根据各小区负载水平和终端移动性,将重载小区的适量业务向重叠覆盖的轻载小区转移;其次通过资源预留和强占优先的接入控制策略,为不同优先级的新到业务提供有差别的服务.仿真结果表明,本文算法在保证系统资源利用率的同时,保障了实时与非实时业务的QoS,并相对于参考算法有效降低了系统阻塞率和业务切换概率.     

DWDM波长路由网络光链路负载均衡的波长路由算法

提出了一种应用于密集波分复用(DWDM)波长路由网络(WRON)中光链路负荷均衡的思想,并将其应用于优化Dijkstra算法的权值,同时将优化Dijkstra算法用于遗传算法求得了在不同的负荷条件下波长下限的网络所需波长数目。并将优化前后的算法分别对美国自然科学基金(NSF)网络的最优波长分配进行数值分析,发现基于负荷均衡思想的优化Dijkstra算法能够对网络的性能有很大提高：当遗传代数为20代时,采用优化Dijkstra算法阻塞率降低了约36％;当波长使用数为7个时,降低网络阻塞率10％。