Fast distributed approximation algorithms for vertex cover and set cover in anonymous networks

We present a distributed algorithm that finds a maximal edge packing in O(Δ + log* W) synchronous communication rounds in a weighted graph, independent of the number of nodes in the network; here Δ is the maximum degree of the graph and W is the maximum weight. As a direct application, we have a distributed 2-approximation algorithm for minimum-weight vertex cover, with the same running time. We also show how to find an f-approximation of minimum-weight set cover in O(f2k2 + fk log* W) rounds; here k is the maximum size of a subset in the set cover instance, f is the maximum frequency of an element, and W is the maximum weight of a subset. The algorithms are deterministic, and they can be applied in anonymous networks.

Optimisation problems in wireless sensor networks

This thesis studies optimisation problems related to modern large-scale distributed systems, such as wireless sensor networks and wireless ad-hoc networks. The concrete tasks that we use as motivating examples are the following: (i) maximising the lifetime of a battery-powered wireless sensor network, (ii) maximising the capacity of a wireless communication network, and (iii) minimising the number of sensors in a surveillance application. A sensor node consumes energy both when it is transmitting or forwarding data, and when it is performing measurements. Hence task (i), lifetime maximisation, can be approached from two different perspectives. First, we can seek for optimal data flows that make the most out of the energy resources available in the network; such optimisation problems are examples of so-called max-min linear programs. Second, we can conserve energy by putting redundant sensors into sleep mode; we arrive at the sleep scheduling problem, in which the objective is to find an optimal schedule that determines when each sensor node is asleep and when it is awake. In a wireless network simultaneous radio transmissions may interfere with each other. Task (ii), capacity maximisation, therefore gives rise to another scheduling problem, the activity scheduling problem, in which the objective is to find a minimum-length conflict-free schedule that satisfies the data transmission requirements of all wireless communication links. Task (iii), minimising the number of sensors, is related to the classical graph problem of finding a minimum dominating set. However, if we are not only interested in detecting an intruder but also locating the intruder, it is not sufficient to solve the dominating set problem; formulations such as minimum-size identifying codes and locating–dominating codes are more appropriate. This thesis presents approximation algorithms for each of these optimisation problems, i.e., for max-min linear programs, sleep scheduling, activity scheduling, identifying codes, and locating–dominating codes. Two complementary approaches are taken. The main focus is on local algorithms, which are constant-time distributed algorithms. The contributions include local approximation algorithms for max-min linear programs, sleep scheduling, and activity scheduling. In the case of max-min linear programs, tight upper and lower bounds are proved for the best possible approximation ratio that can be achieved by any local algorithm. The second approach is the study of centralised polynomial-time algorithms in local graphs – these are geometric graphs whose structure exhibits spatial locality. Among other contributions, it is shown that while identifying codes and locating–dominating codes are hard to approximate in general graphs, they admit a polynomial-time approximation scheme in local graphs.

Voluntary Energy Harvesting Relays and Selection in Cooperative Wireless Networks

The use of energy harvesting (EH) nodes as cooperative relays is a promising and emerging solution in wireless systems such as wireless sensor networks. It harnesses the spatial diversity of a multi-relay network and addresses the vexing problem of a relay's batteries getting drained in forwarding information to the destination. We consider a cooperative system in which EH nodes volunteer to serve as amplify-and-forward relays whenever they have sufficient energy for transmission. For a general class of stationary and ergodic EH processes, we introduce the notion of energy constrained and energy unconstrained relays and analytically characterize the symbol error rate of the system. Further insight is gained by an asymptotic analysis that considers the cases where the signal-to-noise-ratio or the number of relays is large. Our analysis quantifies how the energy usage at an EH relay and, consequently, its availability for relaying, depends not only on the relay's energy harvesting process, but also on its transmit power setting and the other relays in the system. The optimal static transmit power setting at the EH relays is also determined. Altogether, our results demonstrate how a system that uses EH relays differs in significant ways from one that uses conventional cooperative relays.

Composite MR image reconstruction and unaliasing for general trajectories using neural networks

In rapid parallel magnetic resonance imaging, the problem of image reconstruction is challenging. Here, a novel image reconstruction technique for data acquired along any general trajectory in neural network framework, called ``Composite Reconstruction And Unaliasing using Neural Networks'' (CRAUNN), is proposed. CRAUNN is based on the observation that the nature of aliasing remains unchanged whether the undersampled acquisition contains only low frequencies or includes high frequencies too. Here, the transformation needed to reconstruct the alias-free image from the aliased coil images is learnt, using acquisitions consisting of densely sampled low frequencies. Neural networks are made use of as machine learning tools to learn the transformation, in order to obtain the desired alias-free image for actual acquisitions containing sparsely sampled low as well as high frequencies. CRAUNN operates in the image domain and does not require explicit coil sensitivity estimation. It is also independent of the sampling trajectory used, and could be applied to arbitrary trajectories as well. As a pilot trial, the technique is first applied to Cartesian trajectory-sampled data. Experiments performed using radial and spiral trajectories on real and synthetic data, illustrate the performance of the method. The reconstruction errors depend on the acceleration factor as well as the sampling trajectory. It is found that higher acceleration factors can be obtained when radial trajectories are used. Comparisons against existing techniques are presented. CRAUNN has been found to perform on par with the state-of-the-art techniques. Acceleration factors of up to 4, 6 and 4 are achieved in Cartesian, radial and spiral cases, respectively. (C) 2010 Elsevier Inc. All rights reserved.

Interaction Energy Based Protein Structure Networks

The three dimensional structure of a protein is formed and maintained by the noncovalent interactions among the amino acid residues of the polypeptide chain These interactions can be represented collectively in the form of a network So far such networks have been investigated by considering the connections based on distances between the amino acid residues Here we present a method of constructing the structure network based on interaction energies among the amino acid residues in the protein We have investigated the properties of such protein energy based networks (PENs) and have shown correlations to protein structural features such as the clusters of residues involved in stability formation of secondary and super secondary structural units Further we demonstrate that the analysis of PENs in terms of parameters such as hubs and shortest paths can provide a variety of biologically important information such as the residues crucial for stabilizing the folded units and the paths of communication between distal residues in the protein Finally the energy regimes for different levels of stabilization in the protein structure have clearly emerged from the PEN analysis

Work/Family Reconciliation: Corporate Management, Family Policies, and Gender Equality in the Finnish Context

Work/family reconciliation is a crucial question for both personal well-being and on societal level for productivity and re-production throughout the Western world. This thesis examines work/family reconciliation on societal and organisational level in the Finnish context. The study is based on an initial framework, developing it further and analysing the results with help of it. The methodology of the study is plural, including varying epistemological emphasis and both quantitative and qualitative methods. Policy analysis from two different sectors is followed by a survey answered by 113 HR-managers, and then, based on quantitative analyses, interviews in four chosen case companies. The central findings of the thesis are that there indeed are written corporate level policies for reconciling work and family in companies operating in Finland, in spite of the strong state level involvement in creating a policy context in work/family reconciliation. Also, the existing policies vary in accessibility and use. The most frequently used work/family policies still are the statutory state level policies for family leave, taking place when a baby is born and during his or her first years. Still, there are new policies arising, such as a nurse for an employee’s child who has fallen ill, that are based on company activity only, which shows in both accessibility and use of the policy. Reasons for developing corporate level work/family policies vary among the so-called pro-active and re-active companies. In general, family law has a substantial effect for developing corporate level policies. Also headquarter gender equality strategies as well as employee demands are important. In regression analyses, it was found that corporate image and importance in recruitment are the foremost reasons for companies to develop policies, not for example the amount of female employees in the company. The reasons for policy development can be summarized into normative pressures, coercive pressures and mimetic pressures, in line with findings from institutional theory. This research, however, includes awareness of different stakeholder interests and recognizes that institutional theory needs to be complemented with notions of gender and family, which seem to play a part in perceived work/family conflict and need for further work/family policies both in managers’ personal lives and on the organisational level. A very central finding, demanding more attention, is the by HR managers perceived change in values towards work and commitment towards organisation at the youngest working generation, Generation Y. This combined with the need for key personnel has brought new challenges to companies especially in knowledge business and will presumably lead to further development of flexible practices in organisations. The accessibility to this flexibility seems to, however, be even more dependent on the specific knowledge and skills of the employee. How this generation will change the organisations remains to be seen in further research.

Discriminative learning of Bayesian networks via factorized conditional log-likelihood

We propose an efficient and parameter-free scoring criterion, the factorized conditional log-likelihood (ˆfCLL), for learning Bayesian network classifiers. The proposed score is an approximation of the conditional log-likelihood criterion. The approximation is devised in order to guarantee decomposability over the network structure, as well as efficient estimation of the optimal parameters, achieving the same time and space complexity as the traditional log-likelihood scoring criterion. The resulting criterion has an information-theoretic interpretation based on interaction information, which exhibits its discriminative nature. To evaluate the performance of the proposed criterion, we present an empirical comparison with state-of-the-art classifiers. Results on a large suite of benchmark data sets from the UCI repository show that ˆfCLL-trained classifiers achieve at least as good accuracy as the best compared classifiers, using significantly less computational resources.

Circuit layout through an analogy with neural networks

Standard-cell design methodology is an important technique in semicustom-VLSI design. It lends itself to the easy automation of the crucial layout part, and many algorithms have been proposed in recent literature for the efficient placement of standard cells. While many studies have identified the Kerninghan-Lin bipartitioning method as being superior to most others, it must be admitted that the behaviour of the method is erratic, and that it is strongly dependent on the initial partition. This paper proposes a novel algorithm for overcoming some of the deficiencies of the Kernighan-Lin method. The approach is based on an analogy of the placement problem with neural networks, and, by the use of some of the organizing principles of these nets, an attempt is made to improve the behavior of the bipartitioning scheme. The results have been encouraging, and the approach seems to be promising for other NP-complete problems in circuit layout.

Quality-of-Transmission-Aware Manycasting Over Optical Burst-Switched Networks

Many next-generation distributed applications, such as grid computing, require a single source to communicate with a group of destinations. Traditionally, such applications are implemented using multicast communication. A typical multicast session requires creating the shortest-path tree to a fixed number of destinations. The fundamental issue in multicasting data to a fixed set of destinations is receiver blocking. If one of the destinations is not reachable, the entire multicast request (say, grid task request) may fail. Manycasting is a generalized variation of multicasting that provides the freedom to choose the best subset of destinations from a larger set of candidate destinations. We propose an impairment-aware algorithm to provide manycasting service in the optical layer, specifically OBS. We compare the performance of our proposed manycasting algorithm with traditional multicasting and multicast with over provisioning. Our results show a significant improvement in the blocking probability by implementing optical-layer manycasting.

Identification of collagenolytic MMP networks as potential biomarkers in progressive chronic periodontitis subjects and MMP-8 null allele model

Chronic periodontitis results from a complex aetiology, including the formation of a subgingival biofilm and the elicitation of the host s immune and inflammatory response. The hallmark of chronic periodontitis is alveolar bone loss and soft periodontal tissue destruction. Evidence supports that periodontitis progresses in dynamic states of exacerbation and remission or quiescence. The major clinical approach to identify disease progression is the tolerance method, based on sequential probing. Collagen degradation is one of the key events in periodontal destructive lesions. Matrix metalloproteinase (MMP)-8 and MMP-13 are the primary collagenolytic MMPs that are associated with the severity of periodontal inflammation and disease, either by a direct breakdown of the collagenised matrix or by the processing of non-matrix bioactive substrates. Despite the numerous host mediators that have been proposed as potential biomarkers for chronic periodontitis, they reflect inflammation rather than the loss of periodontal attachment. The aim of the present study was to determine the key molecular MMP-8 and -13 interactions in gingival crevicular fluid (GCF) and gingival tissue from progressive periodontitis lesions and MMP-8 null allele mouse model. In study (I), GCF and gingival biopsies from active and inactive sites of chronic periodontitis patients, which were determined clinically by the tolerance method, and healthy GCF were analysed for MMP-13 and tissue inhibitor of matrix metalloproteinases (TIMP)-1. Chronic periodontitis was characterised by increased MMP-13 levels and the active sites showed a tendency of decreased TIMP-1 levels associated with increments of MMP-13 and total protein concentration compared to inactive sites. In study (II), we investigated whether MMP-13 activity was associated with TIMP-1, bone collagen breakdown through ICTP levels, as well as the activation rate of MMP-9 in destructive lesions. The active sites demonstrated increased GCF ICTP levels as well as lowered TIMP-1 detection along with elevated MMP-13 activity. MMP-9 activation rate was enhanced by MMP-13 in diseased gingival tissue. In study (III), we analysed the potential association between the levels, molecular forms, isoenzyme distribution and degree of activation of MMP-8, MMP-14, MPO and the inhibitor TIMP-1 in GCF from periodontitis progressive patients at baseline and after periodontal therapy. A positive correlation was found for MPO/MMP-8 and their levels associated with progression episodes and treatment response. Because MMP-8 is activated by hypochlorous acid in vitro, our results suggested an interaction between the MPO oxidative pathway and MMP-8 activation in GCF. Finally, in study (IV), on the basis of the previous finding that MMP-8-deficient mice showed impaired neutrophil responses and severe alveolar bone loss, we aimed to characterise the detection patterns of LIX/CXCL5, SDF-1/CXCL12 and RANKL in P. gingivalis-induced experimental periodontitis and in the MMP-8-/- murine model. The detection of neutrophil-chemoattractant LIX/CXCL5 was restricted to the oral-periodontal interface and its levels were reduced in infected MMP-8 null mice vs. wild type mice, whereas the detection of SDF-1/CXCL12 and RANKL in periodontal tissues increased in experimentally-induced periodontitis, irrespectively from the genotype. Accordingly, MMP-8 might regulate LIX/CXCL5 levels by undetermined mechanisms, and SDF-1/CXCL12 and RANKL might promote the development and/or progression of periodontitis.

Algorithms for Exact Structure Discovery in Bayesian Networks

Bayesian networks are compact, flexible, and interpretable representations of a joint distribution. When the network structure is unknown but there are observational data at hand, one can try to learn the network structure. This is called structure discovery. This thesis contributes to two areas of structure discovery in Bayesian networks: space--time tradeoffs and learning ancestor relations. The fastest exact algorithms for structure discovery in Bayesian networks are based on dynamic programming and use excessive amounts of space. Motivated by the space usage, several schemes for trading space against time are presented. These schemes are presented in a general setting for a class of computational problems called permutation problems; structure discovery in Bayesian networks is seen as a challenging variant of the permutation problems. The main contribution in the area of the space--time tradeoffs is the partial order approach, in which the standard dynamic programming algorithm is extended to run over partial orders. In particular, a certain family of partial orders called parallel bucket orders is considered. A partial order scheme that provably yields an optimal space--time tradeoff within parallel bucket orders is presented. Also practical issues concerning parallel bucket orders are discussed. Learning ancestor relations, that is, directed paths between nodes, is motivated by the need for robust summaries of the network structures when there are unobserved nodes at work. Ancestor relations are nonmodular features and hence learning them is more difficult than modular features. A dynamic programming algorithm is presented for computing posterior probabilities of ancestor relations exactly. Empirical tests suggest that ancestor relations can be learned from observational data almost as accurately as arcs even in the presence of unobserved nodes.

Mechanical and swelling properties of HTPB-based copolyurethane networks

Copolyurethanes of hydroxy terminated polybutadiene (HTPB) and ISRO–Polyol (ISPO), an indigenously developed castor-oil based polyol, have been prepared using toluene diiso-cyanate and hexamethylenediisocyanate. The mechanical strength and swelling characteristics of the copolyurethanes cured with trimethylol propane and triethanolamine have been studied to evolve improved solid propellant binders. By varying the ratios of the two hydroxy pre-polymers, chain extenders, and crosslinkers, copolyurethanes having a wide range of tensile strength and elongation could be obtained. Many of these systems are desirable for their use as propellant binders. The results have been explained in terms of the measured crosslink densities and other swelling properties. © 1993 John Wiley & Sons, Inc.