In this paper, we address the challenge of performing effective joint wireless data-energy transfers in distributed mobile energy-harvesting networks. In principle, wireless power transfer resembles energy harvesting, however, it exhibits its own special features. We develop a holistic, backpressure-inspired technique, describing the evolution of each node’s queue-battery state, and we define an optimization problem with the objective of improving the balance of energy. We implement a dual Lagrange multipliers solution and determine the key variables influencing the system’s behavior. We investigate the overall energy transfer from the periphery to the core network in cases of traffic-stressed core nodes, and through analysis and simulation, we demonstrate the theoretical and practical potentials of this framework and its potential use for greener and self-sustainable networks.

In various IoT applications it is essential that the autonomous connected devices have the most up-to-date information, either to be read or posted. In this work, we consider a general setting of a wireless multihop network with multiple data flows and study the interactions of backpressure routing, congestion control, clean energy harvesting and Age-of-Information (AoI). We propose a heuristic scheme that is based on optimal source data rates and routing decisions, enhanced with heuristically determined AoI-based features in the form of a weight scaling method. The feasibility of the solution in terms of satisfying queue stability is proven, and the obtained upper bound on the queue lengths depends on the quotient of the max versus the min weight value. Numerical evaluations point out the improvements of the heuristic scheme in terms of delivering fresh information with priority, as well as the tradeoffs between AoI and optimality.

The emerging technology of Wireless Power Trans-fer (WPT) has enabled mobile devices to replenish their batteries and increase their lifetime by exchanging energy with other devices in vicinity. In this paper, we study the problem of peer-to-peer WPT in a network of battery-constrained devices for which we aim to provide a fair energy allocation via mutual exchanges. In this respect, devices of lowest battery level remain functional for longer time, while respecting and satisfying a given set of wireless charging constraints. By taking into consideration the skewed energy availability in the network, as well as the loss induced by wireless energy transfer, we formulate and analyze three energy allocation schemes based on the concepts of lexicographic optimization and algorithmic graph theory, which, under different optimization criteria, aim to extend network lifetime. The performance of the proposed schemes is evaluated and compared in terms of energy efficiency and balancing quality through modeling and simulation over synthetic networks.

The Fifth Generation (5G) mobile communications together with software defined networking (SDN) and network function virtualization (NFV) are expected to enable a wide range of vertical use-cases. Different vertical industries with diverse service streams and sets of requirements should leverage the advanced capabilities of 5G networks through a single infrastructure to support the desired Quality of Service/Experience (QoS/QoE). In this paper, we focus on the Transport vertical and we study four novel service categories, each one consisting of one or more related scenarios, within the framework of the 5G Health, Aquaculture and Transport (5G-HEART) 5G PPP Phase 3 project. The first pass analysis of the envisioned vehicular services and their underlying operation, combined with the mapping of the mostly high-level functional user requirements to quantitative network Key Performance Indicators (KPIs) via a thorough and concise methodology, is essential for future testing with real pilots. Furthermore, our work paves the way towards efficient network slicing by exploring the interrelations between the identified KPIs and the respective target values that must be simultaneously satisfied over the same physical network infrastructure, in the context of the three 5G generic services.

Traffic matrices (TMs) contain information that is essential for network management, traffic engineering, and anomaly detection. However, constructing a TM through direct traffic measurements has a high administrative and computational cost. A more feasible approach is to estimate the TM from the easily obtainable link load measurements. In this paper, we address the issue of traffic matrix estimation (TME) from link loads using a deep generative model – namely, a variational autoencoder (VAE) – to solve the respective ill-posed inverse problem. In particular, we train the VAE with historical data (previously observed TMs) and we leverage the trained decoder to transform TME into a minimization problem in the latent space, which in turn can be solved by employing a gradient-based optimizer. Furthermore, the trained decoder can be used for traffic matrix synthesis, i.e., for generating synthetic TM examples that have “similar” properties to the samples of the training set. Finally, we explore the incremental optimization of the sequence of objectives constructed from the sequence of decoders that we obtain at different stages of the VAE training. The performance of the proposed methods is evaluated using a publicly available dataset of actual traffic matrices recorded in a real backbone network.

In this paper, we study the problem of caching at the network edge by taking into account the impact of recommendations on user content requests. We consider a heterogeneous caching network with small cells and mobile users who can offload traffic from the core network by delivering data via Device-to-Device (D2D) communication. Given the user mobility pattern, we derive for each user the expected waiting time to encounter cache-enabled devices and we propose two schemes in order to select the user equipment that will cache content and participate in the offloading. Expressing the user Quality of Experience (QoE) as a function of user-content relevance and its expected delivery delay, we formulate the problem of content placement and recommendations in caching networks as a user QoE maximization problem, which is known to be NP-hard. In order to address it, we provide two heuristic algorithms, the first focusing on user-content relevance and the second focusing on the content delivery delay. We evaluate the performance of our algorithms through simulation over synthetic datasets. The obtained results are compared with a state-of-the-art polynomial-time approximation algorithm and show that the proposed algorithms balance better the trade-off between solution quality and execution time.

In this work, we consider the joint behavior of an information diffusion process integrated with a recommender system (RecSys) over an online social network (OSN), where the typical users' resilience to information varies, leading to potential information overloads. We assume that each user has a threshold over the information that she can process in a meaningful way, and exceeding it could lead to user dissatisfaction or the user remaining idle. In order to efficiently tackle this issue, while considering complex user constraints, we consider two types of users' capacity to information, that is, capacity for distinct items and capacity for duplicate items. In this setting, we aim to allocate the items in the OSN in order to maximize users' total relevance to the former while ensuring that no user exceeds any type of capacity. We show that this problem is NP-complete and present various heuristic methods to address it. A novel framework, called socially constrained recommendations (SCoRe), is developed for the final assignment of items to users, consisting of a two-step procedure. We present and evaluate two different approaches for each step and discuss the usability of SCoRe for the efficient diffusion of items in the network while respecting the users' constraints.

In this article, we consider the problem of recommendations to the users of an online social network (OSN), through an information diffusion aware recommender system (IDARS). We map the assignment of recommendations in influence networks to a problem of selecting anl-cover of the minimum total cost, which is defined to be a set of assignments such that each user in the OSN is recommended of at leastldifferent items at the minimum defined cost. This corresponds to a special case of the minimum weighted partition set cover problem, which is a generalization of the minimum weighted set cover problem, both of which are proven to be NP-hard. We formulate a corresponding integer programming problem and we apply a linear programming (LP)-based branch and bound (BnB) methodology for its solution. We also propose a greedy algorithm, denoted as CoveR, which we show to be an O((Δ/δ)·H(Δ))-approximation for thel-coverage problem, where Δ and δ are the maximum and minimum degree of an influence network, respectively, and H(Δ) is the Δ th harmonic number. We investigate CoveR's performance through extensive simulations on both synthetic and real networks, which indicate that the quality of its solution is comparable to the one obtained by the BnB method, while at the same time outperforms other information diffusion-aware recommendation heuristics.

In this paper, we address the network tomography problems of inferring the multicast routing tree topology and estimating core link performance characteristics (i.e., loss rate, jitter) based on end-to-end measurements from a source node to a set of destination nodes. We extend the agglomerative hierarchical clustering algorithm that works in a bottom-up manner and iteratively joins siblings (i.e., nodes with the same parent) by incorporating the concepts of reciprocal nearest neighbors and nearest neighbors chains. We employ two alternative ways for calculating the required distance matrix of terminal nodes. One based on additive tree metrics and another utilizing several normalized dissimilarity measures on the binary sequences of received/lost probes maintained at each node. Finally, we evaluate the performance of the proposed algorithm in terms of estimation accuracy and correctness of the inferred logical routing tree over real network topologies constructed in an open testbed of the Fed4FIREP1us federation.

In this paper, we focus on the development and realistic evaluation of a resource allocation approach for cognitive radios implemented with Software Defined Radio (SDR) technology over two testbeds of the ORCA federation. Our cross-layer approach is based on a Markov Random Field (MRF) framework realizing a distributed computation among the secondary nodes of cognitive radios. This is the first implementation and real experimental evaluation of such a mechanism. New SDR functions for implementing various cognitive radio functionalities, such as spectrum sensing, distributed node synchronization, etc., were developed in GNU Radio from scratch. We demonstrated the feasibility of the MRF-based resource allocation approach and quantified various performance metrics of interest, such as allocation fairness and collision percentage. Through this development, several design principles of broader interest for SDR emerged (e.g., for spectrum sensing and collision detection design), while salient features of our framework requiring further research and development were discovered (e.g., need for parallel implementation).

Mathematical models for malware propagation

The proliferation of complex communication networks (CCNs) and their importance for maintaining social coherency nowadays have urgently elevated the need for protecting networking infrastructures from malicious software attacks. In this paper, we propose a Markov Random Field (MRF) based spatio-stochastic framework for modeling the macroscopic behavior of a CCN under random attack, where malicious threats propagate through direct interactions and follow the Susceptible-Infected-Susceptible infection paradigm. We exploit the MRF framework for analytically studying the propagation dynamics in various types of CCNs, i.e., lattice, random, scale-free, small-world and multihop graphs, in a holistic manner. By combining Gibbs sampling with simulated annealing, we study the behavior of the above systems for various topological and malware related parameters with respect to the general random attacks considered. We demonstrate the effectiveness of the MRF framework in capturing the evolution of SIS malware propagation and use it to assess the robustness of synthetic and real CCNs with respect to the involved parameters. It is found that random networks are more robust, followed by scale-free, regular and small-world, while multihop emerge as the most vulnerable of all.

In this paper we propose a topology control based approach for addressing the coverage problem in a planar region containing convex opaque obstacles. Such environments represent typical cases of realistic wireless sensor networks and Internet-of-Things deployments. Assuming the devices have the capability to modify their sensing ranges, our goal is to maximize the area covered by randomly dispersed sensors, while reducing their sensing energy consumption as much as possible despite the presence of convex obstacles. To address the former, we introduce a relevant framework capitalizing on the notion of the visibility polygon and propose two algorithms, a centralized (and a randomized version thereof) and a distributed one, which aim to maximize the ratio of covered area to consumed energy, while ensuring a minimum coverage percentage. Through analysis and simulation we demonstrate that the proposed schemes achieve energy efficient coverage, outperforming the plain assignment of maximum sensing range across the network.

Twitter constitutes an accessible platform for studying and experimenting with the dynamics of information dissemination. By exploiting this and using real data, in this paper, we study the temporal dynamics of topic-specific information spread in Twitter, where we assume that each topic corresponds to a hashtag. We develop an epidemic model for information spread in Twitter and we validate it using real data for several hashtags chosen so as to cover a variety of characteristics. Contrary to the existing works in literature, which define the informed Twitter users as those who have produced/reproduced tweets with a specific hashtag, our model considers as informed a superset of Twitter users who have seen/produced/reproduced tweets with a specific hashtag. Thus, it does not underestimate the extent of information propagation in the network. The evaluation results indicate a satisfactory performance of the proposed epidemic model for all hashtag types examined; while more importantly, they allow studying the impact of several factors, such as the need of time-varying infection rates depending on the hashtag type.

In this paper we introduce and demonstrate new recommendation algorithms for large-scale online systems, such as e-shops and cloud services. The proposed algorithms are based on the combination of network embedding in hyperbolic space with greedy routing, exploiting properties of hyperbolic metric spaces. Contrary to the existing recommender systems that rank products in order to propose the highest ranked ones to the users, our proposed recommender system creates a progressive path of recommendations towards a final (known or inferred) target product using greedy routing over networks embedded in hyperbolic space. Thus, it prepares the user by intermediate recommendations for maximizing the chances that he/she accepts the recommendation of the target product(s). This casts the problem of locating a suitable recommendation as a path problem, where leveraging on the efficiency of greedy routing in graphs embedded in hyperbolic spaces and exploiting special network structure, if any, pays dividends. Two variants of our recommendation approach are provided, namely Hyperbolic Recommendation-Known Destination (HRKD), Hyperbolic Recommendation-Unknown Destination (HRUD), when the target product is known or unknown, respectively. We demonstrate how the proposed approach can be used for producing efficient recommendations in online systems, along with studying the impact of the several parameters involved in its performance via proper emulation of user activity over suitably defined graphs.

In this paper, we address the problem of distributed resource management for secondary users in Cognitive Radio Networks (CRNs), through a topology aware and frequency agile cross-layer approach. We exploit the theory of spatial processes and propose a Markov Random Field (MRF) based framework, which enables secondary CRN users to achieve efficient and viable mechanisms in the lower protocol stack layers by exchanging local only information. Specifically, through Gibbs sampling secondary users can optimize in a distributed and parallel manner their channel allocation, medium access and routing without resolving to otherwise computationally demanding optimization approaches. Through analysis and simulation we exhibit the efficacy of the proposed framework and show that a semi-parallel implementation can significantly reduce the required overhead cost compared to the sequential Gibbs sampling approach, while retaining a very close performance to the latter. We also study the emerging trade-offs by demonstrating the performance benefits in terms of channel assignment, medium access and data flow.

Contemporary traffic demands call for efficient infrastructures capable of sustaining increasing volumes of social communications. In this work, we focus on improving the properties of wireless multihop networks with social features through network evolution. Specifically, we introduce a framework, based on inverse Topology Control (iTC), for distributively modifying the transmission radius of selected nodes, according to social paradigms. Distributed iTC mechanisms are proposed for exploiting evolutionary network churn in the form of edge/node modifications, without significantly impacting available resources. We employ continuum theory for analytically describing the proposed top-down approach of infusing social features in physical topologies. Through simulations, we demonstrate how these mechanisms achieve their goal of reducing the average path length, so as to make a wireless multihop network scale like a social one, while retaining its original multihop character. We study the impact of the proposed topology modifications on the operation and performance of the network with respect to the average throughput, delay, and energy consumption of the induced network.

Secondary ad hoc users of Cognitive Radio networks experience constant fluctuations of their link quality due to dynamic traffic behavior of their primary network counterparts. Several cognitive-aware MAC and routing protocols have been proposed in order to counterfeit properly such spectrum variations. In this paper, we introduce a vertical time-based control mechanism in the traditional protocol stack, aiming at properly balancing the inherent trade-off between instant local adaptations at MAC layer, against slower, but more globally aware rerouting reaction mechanisms. Based on localized and independent node decisions, an optimal feedback policy is proposed in order to exploit the most appropriate protocol layer in each case and provide per link channel assignments, such that flow rate requirements are satisfied with the least cost. We use Markov theory to examine the coexistence of primary and secondary networks, while the optimal decision policy is derived via a Markov Decision Process (MDP) and linear programming. Analysis and simulations are used for performance evaluation and together they confirm that the proposed time-based cross-layer feedback strategy contributes to higher network performance regarding flow rate requirements, sensing and rerouting costs.

Cognitive Radio (CR) techniques have been developed to allow ad hoc users to communicate with each other by exploiting the licensed bands of primary systems without disturbing the entrenched users. In this work, we take current approaches one step ahead and combine Topology Control (TC) techniques with CR technology, in order to improve operation and performance, even under the stringent and non-uniform arrangements of CR networks (CRNs). We propose a novel node-degree based Topology Control approach, denoted by Enhanced Cognitive Nearest Random Neighbor (e-CNRN), for multi-channel CRNs, aiming at maintaining network connectivity and adapting to environmental changes such as primary user activity and channel conditions. Compared with TC protocols in conventional ad hoc networks, e-CNRN requires only minimal local information and is specially designed to perform under non-uniform node arrangements, rendering e-CNRN a generic and robust distributed TC approach, especially suitable for multichannel CRNs as well. In addition, we leverage e-CNRN for distributively establishing a virtual common control channel for multi-channel CRNs. Through analysis and simulations we validate that e-CNRN guarantees network connectivity, while achieving efficient power control.

In this paper the problem of enhancing trust establishment in multi-hop wireless networks is addressed. Exploiting small-world features and based on preferential attachment and initial node trust values, we design inverse Topology Control methods that achieve to reduce the mean hop-distance between two nodes and increase the average trust value of the shortest paths. Based on continuum theory a mathematical framework is developed for the overall socially-motivated trust-based network churn mechanism. Analytical and simulation results exhibit the effectiveness of the proposed approaches for enhancing physical topologies, increasing the average trust path values, and thus further securing future communications systems.

Today's self-managing systems would ideally be able to adapt themselves (their internal structure or behavior), as well as to autonomously participate in larger, self-organizing systems. Analogously, the enterprises or other socio-economic systems autonomously manage themselves - they make decisions on how to adapt their structure and behavior, and how to organize with other entities in the environment. To connect internal self-adaptive with external self-organizational behavior, an enterprise is “aware” of itself and of its environment, and acts according to this awareness. This position paper proposes to address the challenges of a complex distributed self-managing system by making entities in such a system able to adapt themselves similarly to how companies manage themselves in socio-economic systems. To enable the knowledge transfer between these two fields, the paper proposes to utilize symbolic models which will be used by self-managing systems for knowledge representation and reasoning. This will make such systems in a way also self-aware and enable both self-adaptive and self-organizing capabilities. The paper discusses research directions to make this approach possible.

In this paper we study the propagation of malicious software in wireless ad hoc networks under a probabilistic framework. We design topology control algorithms for the development of effective attack strategies by a malicious mobile node, based on the risk function metric, which indicates the network's vulnerability. Our approach takes on the attacker's perspective, in order to investigate the extent of its attack potentials, which in turn could be used for the effective design of network countermeasures. Our performance evaluation results demonstrate that the proposed risk-based topology control algorithms and respective attack strategies effectively balance the tradeoffs between the potential network damage and the attacker's lifetime, and as a result significantly outperform any other flat and threshold-based approaches.

In this paper we study the spreading of malicious software over ad hoc networks, where legitimate nodes are prone to propagate the infections they receive from an attacker or their already infected neighbors. Considering the susceptible-infected-susceptible (SIS) node infection paradigm we propose a probabilistic model, based on the theory of closed queueing networks, that aims at describing the aggregated behavior of the system when attacked by a malicious node. Due to its nature the model is able to deal more effectively with the stochastic behavior of the attacker and the inherent probabilistic nature of the wireless environment. The proposed model is able to describe accurately the asymptotic behavior of malware-propagative ad hoc networking environments, where the number of nodes is large. Using the Norton equivalent of the closed queueing network, we obtain analytical results for its steady state behavior, which in turn can be used to identify the critical parameters affecting the operation of the network.

The spread of active attacks has become a frequent cause of vast systems breakdown in modern communication networks. In this paper, we first present a probabilistic modeling framework for the propagation of an energy-constrained mobile threat in a wireless ad hoc network. The introduced formulation is used to identify and evaluate different attack strategies and approaches, which in turn can help in the development of efficient countermeasures for such attacks. Through modeling and simulation, we evaluate the impact of various parameters associated with the operational characteristics of the mobile attack node - such as transmission radius, mobility, energy - on an outbreak spreading and the evolution of the network. Furthermore, we introduce a new metric which indicates the overall infection-capability of each attack strategy and characterize their ability to harm the network according to this metric