In this paper we present some novel algorithms for scheduling hierarchical signal flow graphs in the domain of high-level synthesis. There are several key contributions of this paper. First, we develop a novel extensi...
详细信息
In this paper we present some novel algorithms for scheduling hierarchical signal flow graphs in the domain of high-level synthesis. There are several key contributions of this paper. First, we develop a novel extension of the force directed scheduling problem which naturally handles loops and conditionals by coming up with a scheme of scheduling hierarchical signal flow graphs. Second, we develop three new parallel algorithms for the scheduling problem. Third, our parallel algorithms are portable across a wide range of parallel platforms. We report results on a set of high-level synthesis benchmarks on 8-processor SGI Challenge and a network of 4 SUN SPARCstation5 work stations. Finally, while some parallel algorithms for VLSI CAD reported by earlier researchers have reported a loss of qualities of results, our parallel algorithms produce exactly the same results as the sequential algorithms on which they are based.
Edge detection is an important process in image segmentation, object recognition, template matching, etc. It computes gradients in both horizontal and vertical directions of the image at each pixel position to find th...
详细信息
Edge detection is an important process in image segmentation, object recognition, template matching, etc. It computes gradients in both horizontal and vertical directions of the image at each pixel position to find the image boundaries. The conventional edge detectors take significant time to detect the edges in the image. To reduce the computational time, this paper proposes parallel algorithms for edge detection with Sobel, Prewitt and Robert first order derivatives using a Shared Memory - Single Instruction Multiple Data (SM - SIMD) parallel architecture. From the experimental results, it is inferred that the proposed parallel algorithms for edge detection are faster than the conventional methods.
The degree-constrained minimum spanning tree (d-MST) problem attempts to find a minimum spanning tree with an added constraint that no nodes in the tree have a degree larger than a specified integer d. It is known tha...
详细信息
The degree-constrained minimum spanning tree (d-MST) problem attempts to find a minimum spanning tree with an added constraint that no nodes in the tree have a degree larger than a specified integer d. It is known that computing the d-MST is NP-hard for every d in the range 2 /spl les/ d /spl les/ (n - 2), where n denotes the total number of nodes. Several approximate algorithms (heuristics) have been proposed in the literature. We have previously proposed three approximate algorithms, IR, TC-RNN, and TC-NNC, for solving the d-MST problem, the last two (TC-RNN and TC-NNC) take advantage of nearest neighbors and their properties. Our experimental results showed that both the TC-RNN and TC-NNC algorithms consistently produce spanning trees with a smaller weight (better quality-of-solution) than that of IR, but using slightly longer execution time. We propose a new heap traversal technique that further improves the time efficiency of TC-RNN and TC-NNC. Our experiments using randomly generated, weighted graphs as inputs show that the TC-NNC algorithm outperforms the other two approximate algorithms in terms of the execution time and quality-of-solution.
Spatial join queries are usually processed using a two-step approach, i.e., the filter-and-refinement strategy. The first step eliminates most of spatial objects that are unlikely to contribute to the query results, a...
详细信息
Vehicle routing problems involve the navigation of one or more vehicles through a network of locations. Locations have associated handling times as well as time windows during which they are active. The arcs connectin...
详细信息
Vehicle routing problems involve the navigation of one or more vehicles through a network of locations. Locations have associated handling times as well as time windows during which they are active. The arcs connecting locations have time costs associated with them. In this paper, we consider two different problems in single vehicle routing. The first is to find least time cost routes between all pairs of nodes in a network for navigating vehicles; we call this the all pairs routing problem. We show that there is an O(log/sup 2/ n) time parallel algorithm using a polynomial number of processors for this problem on a CREW PRAM. We next consider the problem in which a vehicle services all locations in a network. Here, locations can be passed through at any time but only serviced during their time window. The general problem is NP-complete under even fairly stringent restrictions but polynomial algorithms have been developed for some special cases. In particular, when the network is a line, there is no time cost in servicing a location, and all time windows are unbounded at either their lower or upper end, O(n/sup 2/) algorithms have been developed. We show that under the same conditions, we can reduce this problem to the all pairs routing problem and therefore obtain an O(log/sup 2/ n) time parallel algorithm on a CREW PRAM.
In the research, neural nets, or to be more precise, multilayer perceptrons (MLP), are used for the nonlinear phonetic mapping of the acoustic features. These phonetic outputs are used, subsequently, for the final pro...
详细信息
In the research, neural nets, or to be more precise, multilayer perceptrons (MLP), are used for the nonlinear phonetic mapping of the acoustic features. These phonetic outputs are used, subsequently, for the final probabilistic computation of the hidden Markov models (HMM). The speech recognition system was implemented and tested on real speech, low-pass filtered with a cut-off frequency of 3.8 kHz, and sampled at 8 kHz.< >
The existing grounding grid corrosion diagnosis algorithms have large amount of calculation, they take up lots of time and are inconvenient in the practical application. In this paper, a parallel algorithm based on Lo...
详细信息
The existing grounding grid corrosion diagnosis algorithms have large amount of calculation, they take up lots of time and are inconvenient in the practical application. In this paper, a parallel algorithm based on Local Area Network (LAN) is put forward to solve above problems. The master sub-process sends the samples needed to be calculated to the slave sub-processes on the computers of the LAN. The slave sub-processes calculate the samples at the same time and send the results back to the master sub-process. The final result will be figured out by the master sub-process after all the results being gathered. The parallel algorithms of grounding grids testability evaluation, uncertain branches estimation and monitoring pole allocation optimization are discussed, respectively. An experimental grounding grid with sixty branches is used as an example to validate the proposed parallel algorithms. The test is based on a LAN with five computers. The results show that the proposed methods are feasible and the calculation time can be shortened remarkably.
This paper presents interactive software which shows and analyzes through visual simulation some sequential and parallel algorithms. It was study the memory access for PRAM model and some significant algorithms for PR...
详细信息
ISBN:
(纸本)9789604742011
This paper presents interactive software which shows and analyzes through visual simulation some sequential and parallel algorithms. It was study the memory access for PRAM model and some significant algorithms for PRAM model. It was analyze also some sorting algorithms and some graph algorithms in both variants sequential and parallel. The software was implemented in Java. It was also performed a comparative study between a classic sequential algorithm and a parallel algorithm in terms of execution times.
This paper proposes two parallel algorithms called an even region parallel algorithm (ERPA) and an even strip parallel algorithm (ESPA) respectively for extracting ridge and ravine geometric features of a surface. The...
详细信息
This paper proposes two parallel algorithms called an even region parallel algorithm (ERPA) and an even strip parallel algorithm (ESPA) respectively for extracting ridge and ravine geometric features of a surface. The parallel programs were implemented on a GCcl-1/64 T805 transputer based parallel machine with maximum 64 transputers. The performance of these two algorithms are reported and analyzed in respect of a load balance problem and communication overheads. The efficiency and speed-up versus the number of transputers used and the problem size chosen are shown and discussed.< >
暂无评论