Constant-time parallel algorithms for image labeling on areconfigurable network of processors

A constant-time algorithm for labeling the connected components of an N×N image on a reconfigurable network of N³ processors is presented. The main contribution of the algorithm is a novel constant-time technique for determining the minimum-labeled PE in each component. The number of processors used by the algorithm can be reduced to N/sup 2+(1/d/), for any 1⩽d⩽log N, if O(d) time is allowed     

Toward an efficient and scalable feature selection approach for internet traffic classification

There is significant interest in the network management and industrial security community about the need to identify the “best” and most relevant features for network traffic in order to properly characterize user behaviour and predict future traffic. The ability to eliminate redundant features is an important Machine Learning (ML) task because it helps to identify the best features in order to improve the classification accuracy as well as to reduce the computational complexity related to the construction of the classifier. In practice, feature selection (FS) techniques can be used as a preprocessing step to eliminate irrelevant features and as a knowledge discovery tool to reveal the “best” features in many soft computing applications. In this paper, we investigate the advantages and disadvantages of such FS techniques with new proposed metrics (namely goodness, stability and similarity). We continue our efforts toward developing an integrated FS technique that is built on the key strengths of existing FS techniques. A novel way is proposed to identify efficiently and accurately the “best” features by first combining the results of some well-known FS techniques to find consistent features, and then use the proposed concept of support to select a smallest set of features and cover data optimality. The empirical study over ten high-dimensional network traffic data sets demonstrates significant gain in accuracy and improved run-time performance of a classifier compared to individual results produced by some well-known FS techniques.     

Network-coding based event diffusion for wireless networks using semi-broadcasting

Publish/subscribe is a well known and powerful distributed programming paradigm with many potential applications. Publish/subscribe content dissemination techniques based on opportunistic networking and network coding-based epidemic routing are key techniques for optimizing network resources, simplifying network architecture, and providing a platform for realizing innovative networking applications and service.In this paper we consider the central problem of any pub/sub implementation, namely the problem of event dissemination, in the case of a wireless mesh network. We propose a new dissemination strategy based on the notion of semi-broadcast. In a semi-broadcast based protocol the actual content is disseminated in two phases. In the first phase only a fraction of the content is broadcasted (pushed) over the network and stored inside any node, whereas in the second phase the missed part is retried (pulled) on demand from other nodes. Thanks to network coding the partial content stored in each node at the end of the first phase is a set of random linear combinations over the whole content. This allows a very efficient recovery strategy as the missed part is found in nearby nodes with a high probability.The benefit of this approach is that only the interested subscribers, which can vary in number and position over time, can engage the pulling phase.We propose several protocols based on non-trivial forwarding mechanisms that employ network coding as a central tool for supporting adaptive event dissemination while exploiting the broadcast nature of wireless transmissions and guided to the semi-broadcast principle. We show a considerable enhancement in term of total flooding costs and full decoding rates by a self parameter control deployment during the dissemination procedure.     

Transmit antenna selection of correlated MIMO multiuser cognitive radio networks in Nakagami‐m fading channels

In this paper, we examine the impact of antenna correlation on transmit antenna selection with receive maximal ratio combining (TAS/MRC) in multiple‐input multiple‐output multiuser underlay cognitive radio network (MIMO‐MCN) over a Nakagami‐m fading environment. The secondary network under consideration consists of a single source and M destinations equipped with multiple correlated antennas at each node. The primary network composed of L primary users, each of which is equipped with multiple correlated antennas. For the considered underlay spectrum sharing paradigm, the transmission power of the proposed secondary system is limited by the peak interference limit on the primary network and the maximum transmission power at the secondary network. In particular, we derive exact closed‐form expressions for the outage probability and average symbol error rate of the proposed secondary system. To gain further insights, simple asymptotic closed‐form expressions for the outage probability and symbol error rate are provided to obtain the achievable diversity order and coding gain of the system. In addition, the impact of antenna correlation on the secondary user ergodic capacity has been investigated by deriving closed‐form expressions for the secondary user capacity. The derived analytical formulas herein are supported by numerical and simulation results to clarify the main contributions. Copyright © 2016 John Wiley & Sons, Ltd.     

A Generalized Saturation Throughput Analysis for IEEE 802.11e Contention-Based MAC

The main service differentiation mechanism introduced by the IEEE 802.11e standard is described in terms of prioritized contention access. Analyzing the performance and service differentiation ability of each of the priority features of the 802.11e standard is an important task, and considerable research effort has recently been dedicated to this subject. In this article, we present a generalized and comprehensive analysis of the prioritized access scheme of the 802.11e standard. Our model corrects and extends the existing models and considers all priority features of the 802.11e Medium Access Control (MAC) protocol. We consider an arbitrary number of priority levels, and provide a detailed parametric analysis of the MAC behavior for the main priority parameters of inter frame spacing, contention window size, and transmission opportunity. We validate the accuracy of our analysis with simulation experiments.

Parallel constant-time connectivity algorithms on a reconfigurablenetwork of processors

This short note presents constant-time algorithms for labeling the connected components of an image on a network of processors with a wide reconfigurable bus. The algorithms are based on a processor indexing scheme which employs constant-weight codes. The use of such codes enables identifying a single representative processor for each component in a constant number of steps. The proposed algorithms can label an N×N image in O(1) time using N² processors, which is optimal. Furthermore, the proposed techniques lead to an O(logN/loglogN)-time image labeling algorithm on a network of N² processors with a reconfigurable bus of width log N bits. It is shown that these techniques on be applied to labeling an undirected N-vertex graph represented by an adjacency matrix     

Processor-time optimal parallel algorithms for digitized images on mesh-connected processor arrays

We present processor-time optimal parallel algorithms for several problems onn ×n digitized image arrays, on a mesh-connected array havingp processors and a memory of sizeO(n ²) words. The number of processorsp can vary over the range [1,n ^3/2] while providing optimal speedup for these problems. The class of image problems considered here includes labeling the connected components of an image; computing the convex hull, the diameter, and a smallest enclosing box of each component; and computing all closest neighbors. Such problems arise in medium-level vision and require global operations on image pixels. To achieve optimal performance, several efficient data-movement and reduction techniques are developed for the proposed organization.     

Optical CDMA for All-Optical Sub-Wavelength Switching in Core GMPLS Networks

Generalized multi-protocol label switching (GMPLS) is a multipurpose control-plane paradigm that extends the MPLS scheme allowing switching without recognizing packet boundaries. In this paper, we present a novel extension that exploits a new physical layer for switching in optical GMPLS. The proposed extension is achieved through adding an optical code switching layer, or code switch capable (CSC) layer, to the existing label mapping layers. Our proposal enables finer granularity at sub-wavelength level in all-optical GMPLS core switches, resulting in significant enhancements to traffic isolation capabilities for all-optical GMPLS core switches. We employ mathematical analysis to derive performance bounds for the proposed scheme, from both the labeling capacity and network throughput points of view. We use our analytical model to derive several optimum operating points for the network, and show that our techniques significantly improve the overall performance of all-optical core networks     

Performance Analysis of Amplify-and-Forward Relay Systems with Interference-Limited Destination in Various Rician Fading Channels

In this paper, we investigate the performance of a dual-hop fixed-gain amplify-and-forward relay system in the presence of co-channel interference at the destination node. Different fading scenarios for the desired user and interferers’ channels are assumed in this study. We consider the Rician/Nakagami- \(m\) , the Rician/Rician, and the Nakagami- \(m\) /Rician fading environments. In our analysis, we derive accurate approximations for the outage probability and symbol error probability (SEP) of the considered scenarios. The generic independent non-identically distributed (i.n.d.) case of interferers’ channels is considered for the Rician/Nakagami- \(m\) scenario; whereas, the independent identically distributed (i.i.d.) case is studied for the Rician/Rician and the Nakagami- \(m\) /Rician environments. Furthermore, to get more insights on the considered systems, high signal-to-noise ratio (SNR) asymptotic analysis of the outage probability and SEP is derived for special cases of the considered fading scenarios. Monte-Carlo simulations and numerical examples are presented in order to validate the analytical and asymptotic results and to illustrate the effect of interference and other system parameters on the system performance. Results show that the different fading models of interferers’ channels have the same diversity order and that the interference degrades the system performance by only reducing the coding gain. Furthermore, findings show that the case where the fading parameter of the desired user first hop channel is better than that of the second hop gives better performance compared to the vise versa case, especially, at low SNR values; whereas, both cases almost behave the same at high SNR values where the performance of the system is dominated by the interference affecting the worst link. Finally, results show the big gap in system performance due to approximating the Rician fading distribution with the Nakagami- \(m\) distribution which is an indication on the inaccuracy of making such approximations in systems like the considered.     

Efficient network folding techniques for routing permutations inVLSI

Network folding is a technique for realizing permutations on N elements using interconnection networks with M input (and output) terminals, where M相似文献