We study Constrained Online Convex Optimization with Memory (COCO-M), where both the loss and the constraints depend on a finite window of past decisions made by the learner. This setting extends the previously studied unconstrained online optimization with memory framework and captures practical problems such as the control of constrained dynamical systems and scheduling with reconfiguration budgets. For this problem, we propose the first algorithms that achieve sublinear regret and sublinear cumulative constraint violation under time-varying constraints, both with and without predictions of future loss and constraint functions. Without predictions, we introduce an adaptive penalty approach that guarantees sublinear regret and constraint violation. When short-horizon and potentially unreliable predictions are available, we reinterpret the problem as online learning with delayed feedback and design an optimistic algorithm whose performance improves as prediction accuracy improves, while remaining robust when predictions are inaccurate. Our results bridge the gap between classical constrained online convex optimization and memory-dependent settings, and provide a versatile learning toolbox with diverse applications.

While traditional handovers (THOs) have served as a backbone for mobile connectivity, they increasingly suffer from failures and delays, especially in dense deployments and high-frequency bands. To address these limitations, 3GPP introduced Conditional Handovers (CHOs) that enable proactive cell reservations and user-driven execution. However, both handover (HO) types present intricate trade-offs in signaling, resource usage, and reliability. This paper presents unique, countrywide mobility management datasets from a top-tier mobile network operator (MNO) that offer fresh insights into these issues and call for adaptive and robust HO control in next-generation networks. Motivated by these findings, we propose CONTRA, a framework that, for the first time, jointly optimizes THOs and CHOs within the O-RAN architecture. We study two variants of CONTRA: one where users are a priori assigned to one of the HO types, reflecting distinct service or user-specific requirements, as well as a more dynamic formulation where the controller decides on-the-fly the HO type, based on system conditions and needs. To this end, it relies on a practical meta-learning algorithm that adapts to runtime observations and guarantees performance comparable to an oracle with perfect future information (universal no-regret). CONTRA is specifically designed for near-real-time deployment as an O-RAN xApp and aligns with the 6G goals of flexible and intelligent control. Extensive evaluations leveraging crowdsourced datasets show that CONTRA improves user throughput and reduces both THO and CHO switching costs, outperforming 3GPP-compliant and Reinforcement Learning (RL) baselines in dynamic and real-world scenarios.

We revisit the Follow the Regularized Leader (FTRL) framework for Online Convex Optimization (OCO) over compact sets, focusing on achieving dynamic regret guarantees. Prior work has highlighted the framework’s limitations in dynamic environments due to its tendency to produce "lazy" iterates. However, building on insights showing FTRL’s ability to produce "agile" iterates, we show that it can indeed recover known dynamic regret bounds through optimistic composition of future costs and careful linearization of past costs, which can lead to pruning some of them. This new analysis of FTRL against dynamic comparators yields a principled way to interpolate between greedy and agile updates and offers several benefits, including refined control over regret terms, optimism without cyclic dependence, and the application of minimal recursive regularization akin to AdaFTRL. More broadly, we show that it is not the "lazy" projection style of FTRL that hinders (optimistic) dynamic regret, but the decoupling of the algorithm’s state (linearized history) from its iterates, allowing the state to grow arbitrarily. Instead, pruning synchronizes these two when necessary.

Handovers (HOs) are the cornerstone of modern cellular networks for enabling seamless connectivity to a vast and diverse number of mobile users. However, as mobile networks become more complex with more diverse users and smaller cells, traditional HOs face significant challenges, such as prolonged delays and increased failures. To mitigate these issues, 3GPP introduced conditional handovers (CHOs), a new type of HO that enables the preparation (i.e., resource allocation) of multiple cells for a single user to increase the chance of$\mathbf{H O}$success and decrease the delays in the procedure. Despite its advantages, CHO introduces new challenges that must be addressed, including efficient resource allocation and managing signaling/communication overhead from frequent cell preparations and releases. This paper presents a novel framework aligned with the O-RAN paradigm that leverages meta-learning for CHO optimization, providing robust dynamic regret guarantees and demonstrating at least 180% superior performance than other 3GPP benchmarks in volatile signal conditions.

The efficient execution of inferences at the edge is becoming increasingly critical for communication systems that are expected to provide users with fast and accurate mobile data analytics. These inference tasks are inherently latency-sensitive and computationally demanding, whereas edge nodes are limited by energy budgets and heterogeneous resources. This paper studies how a set of edge nodes can collaborate in executing demanding streaming inference tasks to optimize their aggregate performance. Such collaborative task exchange schemes enable the sharing of scarce computing resources and machine learning models (which perform the inferences), and constitute a scalable approach to this intricate problem. We formulate this exchange process as an online convex optimization problem, and design a dynamic task assignment algorithm, which is proven to have optimality guarantees even when the network and service parameters (resources and task properties) are unknown and vary arbitrarily over time. The algorithm aims to maximize inference accuracy while minimizing overall task latency and energy (including for data transfers), and simultaneously ensures that collaborating nodes do not suffer imbalanced energy costs. Through a series of data-driven experiments, we quantify the cooperation benefits under different weight combinations and validate the convergence and adaptability of the proposed learning algorithm across diverse conditions, including variations of system parameters as well as heterogeneity across nodes and tasks.

Open RAN systems, with their virtualized base stations (vBSs), offer increased flexibility and reduced costs, vendor diversity, and interoperability. However, optimizing the allocation of radio resources in such systems raises new challenges due to the volatile vBSs operation, and the dynamic network conditions and user demands they are called to support. Leveraging the novel O-RAN multi-tier control architecture, we propose a new set of resource allocation threshold policies with the aim of balancing the vBSs’ performance and energy consumption in a robust and provably optimal fashion. To that end, we introduce an online learning algorithm that operates under minimal assumptions and without requiring knowledge of the environment, hence being suitable even for “challenging” environments with non-stationary or adversarial demands and conditions. We also develop a meta-learning scheme that utilizes other available algorithmic schemes, e.g., tailored for more “easy” environments, by choosing dynamically the best-performing algorithm; thus enhancing the system’s effectiveness. We prove that the proposed solutions achieve sub-linear regret (zero optimality gap), and characterize their dependence on the main system parameters. The performance of the algorithms is evaluated with real-world data from a testbed, in stationary and adversarial conditions, indicating energy savings of up to 64.5% compared with several state-of-the-art benchmarks.

Parallel Split Learning (SL) allows resource-constrained devices that cannot participate in Federated Learning (FL) to train deep neural networks (NNs) by splitting the NN model into parts. In particular, such devices (clients) may offload the processing task of the largest model part to a computationally powerful helper, and multiple helpers may be employed and work in parallel. In hybrid federated and split learning (HFSL), on the other hand, devices can participate in the training process through any of the two protocols (SL and FL), depending on the system's characteristics. This could considerably reduce the maximum training time over all clients (makespan), especially in highly heterogeneous scenarios. In this paper, we study the joint problem of the training protocol selection, client-helper assignments, and scheduling decisions, to minimize the training makespan. We prove this problem is NP-hard and propose two solution methods: one based on the decomposition of the problem by leveraging its inherent symmetry, and a second fully scalable one. Through numerical evaluations using our testbed's measurements, we build a solution strategy comprising these methods. Moreover, this strategy finds a near-optimal solution and achieves a shorter makespan than the baseline schemes by up to 71%.

Collaborative execution of inference tasks by extreme-edge nodes can effectively address the challenge of scarce resources in the Next Generation IoT and 6G networks. In this paper, we study how such nodes can coordinate the execution of streaming inferences to jointly optimize their task performance (accuracy and latency) and energy consumption. We formulate this process as an online learning problem, and design an online task assignment algorithm, which is proven to provide optimality guarantees even when the network parameters (node resources and task properties) are unkown and subject to arbitrary variations over time. Further, we validate the performance of the proposed algorithm using data-driven simulations of representative scenarios and compare it with non-cooperative benchmarks.

Cellular networks rely on handovers (HOs) as a fundamental element to enable seamless connectivity for mobile users. A comprehensive analysis of HOs can be achieved through data from Mobile Network Operators (MNOs); however, the vast majority of studies employ data from measurement campaigns within confined areas and with limited end-user devices, thereby providing only a partial view of HOs. This paper presents the first countrywide analysis of HO performance, from the perspective of a top-tier MNO in a European country. We collect traffic from approximately 40M users for 4 weeks and study the impact of the radio access technologies (RATs), device types, and manufacturers on HOs across the country. We characterize the geo-temporal dynamics of horizontal (intra-RAT) and vertical (inter-RATs) HOs, at the district level and at millisecond granularity, and leverage open datasets from the country's official census office to associate our findings with the population. We further delve into the frequency, duration, and causes of HO failures, and model them using statistical tools. Our study offers unique insights into mobility management, highlighting the heterogeneity of the network and devices, and their effect on HOs.

Split learning (SL) has been recently proposed as a way to enable resource-constrained devices to train multi-parameter neural networks (NNs) and participate in federated learning (FL). In a nutshell, SL splits the NN model into parts, and allows clients (devices) to offload the largest part as a processing task to a computationally powerful helper. In parallel SL, multiple helpers can process model parts of one or more clients, thus, considerably reducing the maximum training time over all clients (makespan). In this paper, we focus on orchestrating the workflow of this operation, which is critical in highly heterogeneous systems, as our experiments show. In particular, we formulate the joint problem of client-helper assignments and scheduling decisions with the goal of minimizing the training makespan, and we prove that it is NPhard. We propose a solution method based on the decomposition of the problem by leveraging its inherent symmetry, and a second one that is fully scalable. A wealth of numerical evaluations using our testbed’s measurements allow us to build a solution strategy comprising these methods. Moreover, we show that this strategy finds a near-optimal solution, and achieves a shorter makespan than the baseline scheme by up to 52.3%.

The design of effective online caching policies is an increasingly important problem for content distribution networks, online social networks and edge computing services, among other areas. This paper proposes a new algorithmic toolbox for tackling this problem through the lens ofoptimisticonline learning. We build upon the Follow-the-Regularized-Leader (FTRL) framework, which is developed further here to include predictions for the file requests, and we design online caching algorithms for bipartite networks with pre-reserved or dynamic storage subject to time-average budget constraints. The predictions are provided by a content recommendation system that influences the users viewing activity and hence can naturally reduce the caching network's uncertainty about future requests. We also extend the framework to learn and utilize the best request predictor in cases where many are available. We prove that the proposed optimistic learning caching policies can achievesub-zeroperformance loss (regret) for perfect predictions, and maintain the sub-linear regret bound$O(\sqrt{T})$, which is the best achievable bound for policies that do not use predictions, even for arbitrary-bad predictions. The performance of the proposed algorithms is evaluated with detailed trace-driven numerical tests.

Recommendations are employed by Content Providers (CPs) of streaming services in order to boost user engagement and their revenues. Recent works suggest that nudging recommendations towards cached items can reduce operational costs in the caching networks, e.g., Content Delivery Networks (CDNs) or edge cache providers in future wireless networks. However, cache-friendly recommendations could deviate from users' tastes, and potentially affect the CP's revenues. Motivated by real-world business models, this work identifies the misalignment of the financial goals of the CP and the caching network provider, and presents a network-economic framework for recommendations. We propose a cooperation mechanism leveraging the Nash bargaining solution that allows the two entities to jointly design the recommendation policy. We consider different problem instances that vary on the extent these entities are willing to share their cost and revenue models, and propose two cooperative policies, CCR and DCR, that allow them to make decisions in a centralized or distributed way. In both cases, our solution guarantees reaching a fair and Pareto optimal allocation of the cooperation gains. Moreover, we discuss the extension of our framework towards caching decisions. A wealth of numerical experiments in realistic scenarios show the policies lead to significant gains for both entities.

Virtualized Radio Access Networks (vRANs) are fully configurable and can be implemented at a low cost over commodity platforms to enable network management flexibility. In this paper, a novel vRAN reconfiguration problem is formulated to jointly reconfigure the functional splits of the base stations (BSs), locations of the virtualized central units (vCUs) and distributed units (vDUs), their resources, and the routing for each BS data flow. The objective is to minimize the long-term total network operation cost while adapting to the varying traffic demands and resource availability. In the first step, testbed measurements are performed to study the relationship between the traffic demands and computing resources, which reveals high variance and depends on the platform and its load. Consequently, finding the perfect model of the underlying system is non-trivial. Therefore, to solve the proposed problem, a deep reinforcement learning (RL)-based framework is proposed and developed using model-free RL approaches. Moreover, the problem consists of multiple BSs sharing the same resources, which results in a multi-dimensional discrete action space and leads to a combinatorial number of possible actions. To overcome this curse of dimensionality, action branching architecture, which is an action decomposition method with a shared decision module followed by neural network is combined with Dueling Double Deep Q-network (D3QN) algorithm. Simulations are carried out using an O-RAN compliant model and real traces of the testbed. Our numerical results show that the proposed framework successfully learns the optimal policy that adaptively selects the vRAN configurations, where its learning convergence can be further expedited through transfer learning even in different vRAN systems. It also offers significant cost savings by up to 59% of a static benchmark, 35% of Deep Deterministic Policy Gradient with discretization, and 76% of non-branching D3QN.

The virtualization of wireless networks enables new services to access network resources made available by the Network Operator (NO) through a Network Slicing market. The different service providers (SPs) have the opportunity to lease the network resources from the NO to constitute slices that address the demand of their specific network service. The goal of any SP is to maximize its service utility and minimize costs from leasing resources while facing uncertainties of the prices of the resources and the users' demand. In this paper, we propose a solution that allows the SP to decide its online reservation policy, which aims to maximize its service utility and minimize its cost of reservation simultaneously. We design the Optimistic Online Learning for Reservation (OOLR) solution, a decision algorithm built upon the Follow-the-Regularized Leader (FTRL), that incorporates key predictions to assist the decision-making process. Our solution achieves a$\mathcal{O}(\sqrt{T})$regret bound where$T$represents the horizon. We integrate a prediction model into the OOLR solution and we demonstrate through numerical results the efficacy of the combined models' solution against the FTRL baseline.

Virtualized base stations (vBS) can be implemented in diverse commodity platforms and are expected to bring unprecedented operational flexibility and cost efficiency to the next generation of cellular networks. However, their widespread adoption is hampered by their complex configuration options that affect in a non-traditional fashion both their performance and their power consumption requirements. Following an in-depth experimental analysis in a bespoke testbed, we characterize the vBS power cost profile and reveal previously unknown couplings between their various control knobs. Motivated by these findings, we develop a Bayesian learning framework for the orchestration of vBSs and design two novel algorithms: (i) BP-vRAN, which employs online learning to balance the vBS performance and energy consumption, and (ii) SBP-vRAN, which augments our optimization approach with safe controls that maximize performance while respecting hard power constraints. We show that our approaches are data-efficient, i.e., converge an order of magnitude faster than state-of-the-art Deep Reinforcement Learning methods, and achieve optimal performance. We demonstrate the efficacy of these solutions in an experimental prototype using real traffic traces.

We consider the general problem of online convex optimization with time-varying budget constraints in the presence of predictions for the next cost and constraint functions, that arises in a plethora of network resource management problems. A novel saddle-point algorithm is designed by combining a Follow-The-Regularized-Leader iteration with prediction-adaptive dynamic steps. The algorithm achieves$\mathcal O(T^{(3-\beta)/4})$regret and$\mathcal O(T^{(1+\beta)/2})$constraint violation bounds that are tunable via parameter$\beta \!\in \![1/2,1$) and have constant factors that shrink with the predictions quality, achieving eventually$\mathcal O(1)$regret for perfect predictions. Our work extends the seminal FTRL framework for this new OCO setting and outperforms the respective state-of-the-art greedy-based solutions which naturally cannot benefit from predictions, without imposing conditions on the (unknown) quality of predictions, the cost functions or the geometry of constraints, beyond convexity.

Future mobile networks need to support intelligent services which collect and process data streams at the network edge, so as to offer real-time and accurate inferences to users. However, the widespread deployment of these services is hindered by the unprecedented energy cost they induce to the network, and by the difficulties in optimizing their end-to-end operation. To address these challenges, we propose a Bayesian learning framework for jointly configuring the service and the Radio Access Network (RAN), aiming to minimize the total energy consumption while respecting accuracy and latency service requirements. Using a fully-fledged prototype with a software-defined base station (vBS) and a GPU-enabled edge server, we profile a typical video analytics service and identify new performance trade-offs and optimization opportunities. Accordingly, we tailor the proposed learning framework to account for the (possibly varying) network conditions, user needs, and service metrics, and apply it to a range of experiments with real traces. Our findings suggest that this approach effectively adapts to different hardware platforms and service requirements, and outperforms state-of-the-art benchmarks based on neural networks.

The design of Open Radio Access Network (O-RAN) compliant systems for configuring the virtualized Base Stations (vBSs) is of paramount importance for network operators. This task is challenging since optimizing the vBS scheduling procedure requires knowledge of parameters, which are erratic and demanding to obtain in advance. In this paper, we propose an online learning algorithm for balancing the performance and energy consumption of a vBS. This algorithm provides performance guarantees under unforeseeable conditions, such as non-stationary traffic and network state, and is oblivious to the vBS operation profile. We study the problem in its most general form and we prove that the proposed technique achieves sub-linear regret (i.e., zero average optimality gap) even in a fast-changing environment. By using real-world data and various trace-driven evaluations, our findings indicate savings of up to 74.3% in the power consumption of a vBS in comparison with state-of-the-art benchmarks.

An increasing number of mobile applications rely on Machine Learning (ML) routines for analyzing data. Executing such tasks at the user devices saves the energy spent on transmitting and processing large data volumes at distant cloud-deployed servers. However, due to memory and computing limitations, the devices often cannot support the required resource-intensive routines and fail to accurately execute such tasks. In this work, we address the problem of edge-assisted analytics in resource-constrained systems by proposing and evaluating a rigorous selective offloading framework. The devices execute their tasks locally and outsource them to cloudlet servers only when they predict a significant performance improvement. We consider the practical scenario where the offloading gains and resource costs are time-varying; and propose an online optimization algorithm that maximizes the service performance without requiring to know this information. Our approach relies on an approximate dual subgradient method combined with a primal-averaging scheme, and works under minimal assumptions about the system stochasticity. We fully implement the proposed algorithm in a wireless testbed and evaluate its performance using a state-of-the-art image recognition application, finding significant performance gains and cost savings.

Network slicing markets have the potential to increase significantly the utilization of virtualized network resources and facilitate the low-cost deployment of over-the-top services. However, their success is conditioned on the service providers (SPs) being able to bid effectively for the virtualized resources. In this paper, we consider a hybrid advance-reservation and spot slice market and study how the SPs should reserve resources to maximize their services’ performance while not violating a time-average budget threshold. We consider this problem in its general form where the SP demand and slice prices are time-varying and revealed only after the reservations are decided. We develop a learning-based framework, using the theory of online convex optimization, that allows the SP to employ a no-regret reservation policy, i.e., achieve the same performance with an oracle that has full access to all future demand and prices. We extend the framework to the scenario where the SP decides dynamically its slice orchestration and hence needs to learn the performance-maximizing resource composition; and we further develop a mixed-time scale scheme that allows the SP to leverage spot-market information that is revealed between successive reservations. The proposed learning framework is evaluated using representative simulation scenarios that highlight its efficacy as well as the impact of key system and algorithm parameters.

The development of Multi-access edge computing (MEC) has resulted from the requirement for supporting next generation mobile services, which need high capacity, high reliability and low latency. The key issue in such MEC architectures is to decide which edge nodes will be employed for serving the needs of the different end users. Here, we take a fresh look into this problem by focusing on the minimization of migration events rather than focusing on maximizing usage of resources. This is important because service migrations can create significant service downtime to applications that need low latency and high reliability, in addition to increasing traffic congestion in the underlying network. This paper introduces a priority induced service migration minimization (PrISMM) algorithm, which aims at minimizing service migration for both high and low priority services, through the use of Markov decision process, learning automata and combinatorial optimization. We carry out extensive simulations and produce results showing its effectiveness in reducing the mean service downtime of lower priority services and the mean admission time of the higher priority services.

Supporting Edge AI services is one of the most exciting features of future mobile networks. These services involve the collection and processing of voluminous data streams, right at the network edge, so as to offer real-time and accurate inferences to users. However, their widespread deployment is hampered by the energy cost they induce to the network. To overcome this obstacle, we propose a Bayesian learning framework for jointly configuring the service and the Radio Access Network (RAN), aiming to minimize the total energy consumption while respecting desirable accuracy and latency thresholds. Using a fully-fledged prototype with a software-defined base station (BS) and a GPU-enabled edge server, we profile a state-of-the-art video analytics AI service and identify new performance trade-offs. Accordingly, we tailor the optimization framework to account for the network context, the user needs, and the service metrics. The efficacy of our proposal is verified in a series of experiments and comparisons with neural network-based benchmarks.

Millions of sensors, cameras, meters, and other edge devices are deployed in networks to collect and analyse data. In many cases, such devices are powered only by Energy Harvesting (EH) and have limited energy available to analyse acquired data. When edge infrastructure is available, a device has a choice: to perform analysis locally or offload the task to other resource-rich devices such as cloudlet servers. However, such a choice carries a price in terms of consumed energy and accuracy. On the one hand, transmitting raw data can result in a higher energy cost in comparison to the required energy to process data locally. On the other hand, performing data analytics on servers can improve the task's accuracy. Additionally, due to the correlation between information sent by multiple devices, accuracy might not be affected if some edge devices decide to neither process nor send data and preserve energy instead. For such a scenario, we propose a Deep Reinforcement Learning (DRL) based solution capable of learning and adapting the policy to the time-varying energy arrival due to EH patterns. We leverage two datasets, one to model energy an EH device can collect and the other to model the correlation between cameras. Furthermore, we compare the proposed solution performance to three baseline policies. Our results show that we can increase accuracy by 15% in comparison to conventional approaches while preventing outages.

Recommender systems have been established as a key component of video streaming services, shaping up to 80% of content requests. Hence, recommendations are employed by the Content Providers (CPs) of these services to increase the viewing time and their revenues. Furthermore, it has been recently suggested that recommendations could be a means to reduce the operational costs of the Content Delivery Networks (CDNs) when they are related to already cached items, i.e., when they are cache-friendly. Clearly, these conflicting objectives, i.e., increasing revenue for the CP and reducing costs for the CDN, can create tensions between the two entities, and hence, prevent the full utilization of recommendations. In this work, we propose a model for capturing these tradeoffs, and an economic mechanism, based on the Nash bargaining solution, for reconciling the potentially conflicting objectives of the CP and the CDN. Our scheme enables the CP and CDN to jointly design the recommendations in a way that balances the revenue gains and cost savings, ensuring a fair and Pareto optimal split of the accrued benefits for both entities. Our numerical experiments in realistic scenarios show that the proposed scheme leads to important financial gains of up to 30%.

Network virtualization is intended to be a key element of new generation networks. However, it is no clear how the implantation of this new paradigm will affect the power consumption of the network. To shed light on this relatively unexplored topic, we evaluate and analyze the power consumption of virtualized Base Station (vBS) experimentally. In particular, we measure the power consumption associated with uplink transmissions as a function of different variables such as traffic load, channel quality, modulation selection, and bandwidth. We find interesting tradeoffs between power savings and performance and propose two linear mixed-effect models to approximate the experimental data. These models allow us to understand the power behavior of the vBS and select power-efficient configurations. We release our experimental dataset hoping to foster further efforts in this research area.

Emerging network slicing markets promise to boost the utilization of expensive network resources and to unleash the potential of over-the-top services. Their success, however, is conditioned on the service providers (SPs) being able to bid effectively for the virtualized resources. In this paper we consider a hybrid advance-reservation and spot slice market and study how the SPs should reserve slices in order to maximize their performance while not exceeding their budget. We consider this problem in its general form, where the SP demand and slice prices are time-varying and revealed only after the reservations are decided. We develop a learning-based framework, using the theory of online convex optimization, that allows the SP to employ a no-regret reservation policy, i.e., achieve the same performance with a hypothetical policy that has knowledge of future demand and prices. We extend our framework for the scenario the SP decides dynamically its slice orchestration, where it additionally needs to learn which resource composition is performance - maximizing; and we propose a mixed-time scale scheme that allows the SP to leverage any spot-market information revealed between its reservations. We evaluate our learning framework and its extensions using a variety of simulation scenarios and following a detailed parameter sensitivity analysis.

Radio Access Network Virtualization (vRAN) will spearhead the quest towards supple radio stacks that adapt to heterogeneous infrastructure: from energy-constrained platforms deploying cells-on-wheels (e.g., drones) or battery-powered cells to green edge clouds. We perform an in-depth experimental analysis of the energy consumption of virtualized Base Stations (vBSs) and render two conclusions: (i) characterizing performance and power consumption is intricate as it depends on human behavior such as network load or user mobility; and (ii) there are many control policies and some of them have non-linear and monotonic relations with power and throughput. Driven by our experimental insights, we argue that machine learning holds the key for vBS control. We formulate two problems and two algorithms: (i) BP-vRAN, which uses Bayesian online learning to balance performance and energy consumption, and (ii) SBP-vRAN, which augments our Bayesian optimization approach with safe controls that maximize performance while respecting hard power constraints. We show that our approaches are data-efficient and have provably performance, which is paramount for carrier-grade vRANs. We demonstrate the convergence and flexibility of our approach and assess its performance using an experimental prototype.

Video analytics constitute a core component of many wireless services that require processing of voluminous data streams emanating from handheld devices. Multi-Access Edge Computing (MEC) is a promising solution for supporting such resource-hungry services, but there is a plethora of configuration parameters affecting their performance in an unknown and possibly time-varying fashion. To overcome this obstacle, we propose an Automated Machine Learning (AutoML) framework for jointly configuring the service and wireless network parameters, towards maximizing the analytics' accuracy subject to minimum frame rate constraints. Our experiments with a bespoke prototype reveal the volatile and system/data-dependent performance of the service, and motivate the development of a Bayesian online learning algorithm which optimizes on-the-fly the service performance. We prove that our solution is guaranteed to find a near-optimal configuration using safe exploration, i.e., without ever violating the set frame rate thresholds. We use our testbed to further evaluate this AutoML framework in a variety of scenarios, using real datasets.

Data centers are increasingly using high-speed circuit switches to cope with the growing demand and reduce operational costs. One of the fundamental tasks of circuit switches is to compute a sparse collection of switching configurations to support a traffic demand matrix. Such a problem has been addressed in the literature with variations of the approach proposed by Birkhoff in 1946 to decompose a doubly stochastic matrix exactly. However, the existing methods are heuristic and do not have theoretical guarantees on how well a collection of switching configurations (i.e., permutations) can approximate a traffic matrix (i.e., a scaled doubly stochastic matrix). In this paper, we revisit Birkhoff’s approach and make three contributions. First, we establish the first theoretical bound on the sparsity of Birkhoff’s algorithm (i.e., the number of switching configurations necessary to approximate a traffic matrix). In particular, we show that by using a subset of the admissible permutation matrices, Birkhoff’s algorithm obtains an$\epsilon $-approximate decomposition with at most$O(\log (1 / \epsilon))$permutations. Second, we propose a new algorithm,Birkhoff+, which combines the wealth of Frank-Wolfe with Birkhoff’s approach to obtain sparse decompositions in a fast manner. And third, we evaluate the performance of the proposed algorithm numerically and study how this affects the performance of a circuit switch. Our results show thatBirkhoff+is superior to previous algorithms in terms of throughput, running time, and number of switching configurations.

The advent of elastic Content Delivery Networks (CDNs) enable Content Providers (CPs) to lease cache capacity on demand and at different cloud and edge locations in order to enhance the quality of their services. This article addresses key challenges in this context, namely how to invest an available budget in cache space in order to match spatio-temporal fluctuations of demand, wireless environment and storage prices. Specifically, we jointly consider dynamic cache rental, content placement, and request-cache association in wireless scenarios in order to provide just-in-time CDN services. The goal is to maximize the an aggregate utility metric for the CP that captures both service benefits due to caching and fairness in servicing different end users. We leverage the Lyapunov drift-minus-benefit technique and Jensen's inequality to transform our infinite horizon problem into hour-by-hour subproblems which can be solved without knowledge of future file popularity and transmission rates. For the case of non-overlapping small cells, we provide an optimal subproblem solution. However, in the general overlapping case, the subproblem becomes a mixed integer non-linear program (MINLP). In this case, we employ a randomized cache lease method to derive a scalable solution. We show that the proposed algorithm guarantees the theoretical performance bound by exploiting the submodularity property of the objective function and pick-and-compare property of the randomized cache lease method. Finally, via real dataset driven simulations, we find that the proposed algorithm achieves 154% utility compared to similar static cache storage-based algorithms in a representative urban topology.

Virtualized radio access networks (vRAN) are emerging as a key component of wireless cellular networks, and it is therefore imperative to optimize their architecture. vRANs are decentralized systems where the Base Station (BS) functions can be split between the edge Distributed Units (DUs) and Cloud computing Units (CUs); hence they have many degrees of design freedom. We propose a framework for optimizing the number and location of CUs, the function split for each BS, and the association and routing for each DU-CU pair. We combine a linearization technique with a cutting-planes method to expedite theexactproblem solution. The goal is to minimize the network costs and balance them with the criterion of centralization, i.e., the number of functions placed at CUs. Using data-driven simulations we find that multi-CU vRANs achieve cost savings up to 28% and improve centralization by 77%, compared to single-CU vRANs. Interestingly, we see non-trivial trade-offs among centralization and cost, which can be aligned or conflicting based on the traffic and network parameters. Our work sheds light on the vRAN design problem from a new angle, highlights the importance of deploying multiple CUs, and offers a rigorous optimization tool for balancing costs and performance.

This paper studies Online Convex Optimization (OCO) problems where the constraints have additive perturbations that (i) vary over time and (ii) are not known at the time to make a decision. Perturbations may not be i.i.d. generated and can be used, for example, to model a time-varying budget or time-varying requests in resource allocation problems. Our goal is to design a policy that obtains sublinear regret and satisfies the constraints in the long-term. To this end, we present an online primal-dual proximal gradient algorithm that has $O(T^\epsilon \vee T^{1-\epsilon})$ regret and $O(T^\epsilon)$ constraint violation, where $\epsilon \in [0,1)$ is a parameter in the learning rate. The proposed algorithm obtains optimal rates when $\epsilon = 1/2$, and can compare against a stronger comparator (the set of fixed decisions in hindsight) than previous work.

Delivering massive video content while consuming low energy at the user devices is an issue of increasing importance, both for the users and operators. In this paper we consider the problem of minimum-energy video delivery when the users are served by LTE multicast transmissions which are assisted by Wi-Fi networks. We formulate the optimization problem where we jointly decide the user association and spectrum allocation to multicast groups, and the Wi-Fi access point (AP) association and airtime allocation to users, in order to deliver a requested video quality. The problem is NP-hard and we propose a cutting planes algorithm, based on Benders decomposition, to expedite its exact solution. We evaluate our proposal using a wealth of simulation experiments based on 3GPP parameters and measurement studies. Our findings show that coordinated decisions on both networks reduce the devices power consumption by up to approximately 50%.

360° video streaming is rapidly progressing towards the multimedia industry, offering an interactive user experience but also challenging telco operators due to the huge volume of the video data and the user channel dynamics. We propose a software defined transport layer proxy overlay architecture, that builds on top of existing standard technologies, to facilitate hybrid multicast/unicast 360° video delivery as a service-particularly combining LTE multicast and WiFi unicast transmissions. In addition we holistically define the emergent optimization problem, where we jointly consider network association, multicast/unicast scheduling, and spectrum/bandwidth management in both networks, taking also into consideration user specific preferences, e.g., the popularity of the video tiles. The proposed framework is evaluated through a series of simulation based experiments following 3GPP standard parameters and real 360° video traces. Our findings reveal significantly increase in the user utility, particularly when the wireless spectrum is limited.

Real-time object recognition is becoming an essential part of many emerging services, such as augmented reality, which require accurate inference in a timely fashion with low delay. We consider an edge-assisted object recognition system that can be configured in ways that have diverse impacts on these key performance criteria. Our goal is to design an online algorithm that learns the optimal configuration of the system by observing the outcomes of configurations applied in the past. We leverage the structure of the problem and combine a Gaussian process with a multi-armed bandit framework to efficiently solve the problem at hand. Our results indicate that our solution makes better configuration choices compared to other bandit algorithms, resulting in lower regret.

A large number of emerging IoT applications rely on machine learning routines for analyzing data. Executing such tasks at the user devices improves response time and economizes network resources. However, due to power and computing limitations, the devices often cannot support such resource-intensive routines and fail to accurately execute the analytics. In this work, we propose to improve the performance of analytics by leveraging edge infrastructure. We devise an algorithm that enables the IoT devices to execute their routines locally; and then outsource them to cloudlet servers, only if they predict they will gain a significant performance improvement. It uses an approximate dual subgradient method, making minimal assumptions about the statistical properties of the system's parameters. Our analysis demonstrates that our proposed algorithm can intelligently leverage the cloudlet, adapting to the service requirements.

We develop an edge-assisted object recognition system with the aim of studying the system-level trade-offs between end-to-end latency and object recognition accuracy. We focus on developing techniques that optimize the transmission delay of the system and demonstrate the effect of image encoding rate and neural network size on these two performance metrics. We explore optimal trade-offs between these metrics by measuring the performance of our real time object recognition application. Our measurements reveal hitherto unknown parameter effects and sharp trade-offs, hence paving the road for optimizing this key service. Finally, we formulate two optimization problems using our measurement-based models and following a Pareto analysis we find that careful tuning of the system operation yields at least 33% better performance for real time conditions, over the standard transmission method.

We study the important and challenging problem of virtualized radio access network (vRAN) design in its most general form. We develop an optimization framework that decides the number and deployment locations of central/cloud units (CUs); which distributed units (DUs) each of them will serve; the functional split that each BS will implement; and the network paths for routing the traffic to CUs and the network core. Our design criterion is to minimize the operator's expenditures while serving the expected traffic. To this end, we combine a linearization technique with a cutting-planes method in order to expedite the exact solution of the formulated problem. We evaluate our framework using real operational networks and system measurements, and follow an exhaustive parameter-sensitivity analysis. We find that the benefits when departing from single-CU deployments can be as high as 30% for our networks, but these gains diminish with the further addition of CUs. Our work sheds light on the vRAN design from a new angle, highlights the importance of deploying multiple CUs, and offers a rigorous framework for optimizing the costs of Multi-CUs vRAN.

We study the problem of in-network execution of data analytic services using multi-grade VNF chains. The nodes host VNFs offering different and possibly time-varying gains for each stage of the chain, and our goal is to maximize the analytics performance while minimizing the data transfer and processing costs. The VNFs' performance is revealed only after their execution, since it is data-dependent or controlled by third-parties, while the service requests and network costs might also vary with time. We devise an operation algorithm that learns, on the fly, the optimal routing policy and the composition and length of each chain. Our algorithm combines a lightweight sampling technique and a Lagrange-based primal-dual iteration, allowing it to be scalable and attain provable optimality guarantees. We demonstrate the performance of the proposed algorithm using a video analytics service, and explore how it is affected by different system parameters. Our model and optimization framework is readily extensible to different types of networks and services.

We propose an online policy that schedules the transmission and processing of data analytic tasks in an Internet of Things (IoT) network. The tasks are executed with different precision at the (possibly heterogeneous) nodes; the network is subject to link bandwidth and node processing capacity changes; and the task requests vary following unknown statistics. For this general IoT scenario, we formulate a resource allocation problem towards maximizing the aggregate tasks precision, and design a dynamic solution policy by combining the FrankWolfe and dual subgradient algorithms. Our policy (FWDS) is guaranteed to converge within bounded distance from the optimal solution, while ensuring interference-free transmissions, and being oblivious to network and/or traffic load changes. We use a wireless testbed and a state-of-the-art face recognition application to implement FWDS and compare it with static or dynamic (maxweight-type) competitor policies. Our findings verify that FWDS pushes the envelope of network control algorithms by handling time-varying objectives and possibly non-i.i.d. network/load statistics, with smaller complexity than its competitors.

Mobile network operators today need to deliver new types of multimedia content, such as 360° video, to an ever-growing population of users. This creates a pressing need for network mechanisms that can support such demanding services in an economically-efficient fashion. This work focuses on applications that deliver a video file concurrently to multiple users which can utilize different radio technologies, namely cellular and Wi-Fi networks. We propose a mechanism for the orchestration of LTE multicast and Wi-Fi unicast transmissions by optimizing jointly: the design of multicast groups, the spectrum and bandwidth management in both networks, and the video encoding quality for each user (or, group). We consider two key performance criteria: minimizing the utilized LTE spectrum resources, and maximizing the delivered video quality. We formulate the respective optimization problems, prove they are NP-hard, and design an exact algorithm for solving them in near-real time. We employ a wealth of simulation experiments, using 3GPP-compliant parameters, that compare our approach with state-of-the-art benchmarks. Our findings suggest that, in representative scenarios, this multi-RAT orchestration can save up 55% LTE radio resources and increase up to 46% the delivered video quality.

An increasing number of mobile applications (abbrev. apps), like Pokemon Go and Snapchat, reward the users who physically visit some locations tagged as POIs (places-of-interest) by the apps. We study the novel POI-based collaboration between apps and venues (e.g., restaurants). On the one hand, an app charges a venue and tags the venue as a POI. The POI tag motivates users to visit the venue, which potentially increases the venue's sales. On the other hand, the venue can invest in the app-related infrastructure, which enables more users to use the app and further benefits the app's business. The apps' existing POI tariffs cannot fully incentivize the venue's infrastructure investment, and hence cannot lead to the most effective app-venue collaboration. We design an optimal two-part tariff, which charges the venue for becoming a POI, and subsidizes the venue every time a user interacts with the POI. The subsidy design efficiently incentivizes the venue's infrastructure investment, and we prove that our tariff achieves the highest app's revenue among a general class of tariffs. Furthermore, we derive some counter-intuitive guidelines for the POI-based collaboration. For example, a bandwidth-consuming app should collaborate with a low-quality venue (users have low utilities when consuming the venue's products).

We study the problem of wireless edge caching when file popularity is unknown and possibly non-stationary. A bank of J caches receives file requests and a utility is accrued for each request depending on the serving cache. The network decides dynamically which files to store at each cache and how to route them, in order to maximize total utility. The request sequence is assumed to be drawn from an arbitrary distribution, capturing time-variance, temporal and spatial locality of requests. For this challenging setting, we propose the Bipartite Supergradient Caching Algorithm (BSCA) which provably exhibits no regret (RT/T → 0). That is, as the time horizon T increases, BSCA achieves (at least) the same utility with the cache configuration that we would have chosen knowing all future requests. The learning rate of the algorithm is characterized by its regret expression RT= O( √JT ), which is independent of the file library size. For the single-cache case, we prove that this is the lowest attainable bound. BSCA requires at each step J projections on intersections of boxes and simplices, for which we propose a tailored algorithm. Our model is the first that draws a connection between the network caching problem and Online Convex Optimization, and we demonstrate its generality by discussing various practical extensions and presenting a tracedriven comparison with state-of-the-art competitors.

This paper introduces a novel caching analysis that, contrary to prior work, makes no modeling assumptions for the file request sequence. We cast the caching problem in the framework of Online Linear optimization (OLO), and introduce a class of minimum regret caching policies, which minimize the losses with respect to the best static configuration in hindsight when the request model is unknown. These policies are very important since they are robust to popularity deviations in the sense that they learn to adjust their caching decisions when the popularity model changes. We first prove a novel lower bound for the regret of any caching policy, improving existing OLO bounds for our setting. Then we show that the Online Gradient Ascent (OGA) policy guarantees a regret that matches the lower bound, hence it is universally optimal. Finally, we shift our attention to a network of caches arranged to form a bipartite graph, and show that the Bipartite Subgradient Algorithm (BSA) has no regret.

This paper studies the problem of allocating band-width and computation resources to data analytics tasks in Internet of Things (IoT) networks. IoT nodes are powered by batteries, can process (some of) the data locally, and the quality grade or performance of how data analytics tasks are carried out depends on where these are executed. The goal is to design a resource allocation algorithm that jointly maximizes the network lifetime and the performance of the data analytics tasks subject to energy constraints. This joint maximization problem is challenging with coupled resource constraints that induce non-convexity. We first show that the problem can be mapped to an equivalent convex problem, and then propose an online algorithm that provably solves the problem and does not require any a priori knowledge of the time-varying wireless link capacities and data analytics arrival process statistics. The algorithm's optimality properties are derived using an analysis which, to the best of our knowledge, proves for the first time the convergence of the dual subgradient method with time-varying sets. Our simulations seeded by real IoT device energy measurements, show that the network connectivity plays a crucial role in network lifetime maximization, that the algorithm can obtain both maximum network lifetime and maximum data analytics performance in addition to maximizing the joint objective, and that the algorithm increases the network lifetime by approximately 50% compared to an algorithm that minimizes the total energy consumption.

We study the problem of cooperative execution of IoT data analytics at the edge nodes. We propose an auction mechanism that allows the nodes to trade their tasks in a way that optimizes execution accuracy and delay, without requiring information about the nodes' accuracy/delay priorities, and yields the optimal task outsourcing policy. Testbed-driven evaluation shows gains in both accuracy and delay compared to heuristic/greedy policies.

The proliferation of advanced mobile devices has enabled the wide-spread adoption of computation-intensive mobile applications. Nevertheless, the execution of these applications is still constrained by the limited battery and processing capacity of the devices. These limitations have led to the mobile cloud/edge computation offloading paradigm, and several such architectures have been proposed in recent years. However, we currently lack a clear understanding of the benefits of these diverse offloading solutions in terms of computation delay and energy cost savings. In this paper, we implement a hierarchical mobile-edge-cloud computing system that is comprised of a mobile device, an edge server and a cloud server, and conduct a series of experiments to measure the device's energy consumption, as well as the computation and transmission delay for different tasks. Our experiments reveal an interesting relation between the mobile's CPU clock frequency and the transmission delay for the offloaded tasks. Based on this finding, we propose a dynamic greedy algorithm that selects the CPU frequency, the tasks to be offloaded, and the network path (cellular or WiFi), in order to reduce the total energy cost and execution delay. We fully implement this multi-tier architecture and verify experimentally that the proposed algorithm can save up to 50% of the device battery energy, at the expense of transmission delay. Our results motivate the design of offloading policies that optimize jointly the CPU clock and network capacity.

WiFi is increasingly used by carriers for opportunistically offloading the cellular network infrastructure or even for increasing their revenue through WiFi-only plans and WiFi on-demand passes. Despite the importance and momentum of this technology, the current deployment of WiFi access points (APs) by the carriers follows mostly a heuristic approach. In addition, the prevalent free-of-charge WiFi access policy may result in significant opportunity costs for the carriers, as this traffic could yield non-negligible revenue. In this paper, we study the problem of optimizing the deployment of WiFi APs and pricing the WiFi data usage with the goal of maximizing carrier profit. Addressing this problem is a prerequisite for the efficient integration of WiFi to next-generation carrier networks. Our framework considers various demand models that predict how traffic will change in response to alteration in price and AP locations. We present both optimal and approximate solutions and reveal how key parameters shape the carrier profit. Evaluations on a dataset of WiFi access patterns indicate that WiFi can indeed help carriers reduce their costs while charging users about 50% lower than the cellular service.

Distributed caching architectures have been proposed for bringing content close to requesters, and the key problem is to design caching algorithms for reducing content delivery delay, which determines to an extent the user Quality of Experience (QoE). This problem obtains an interesting new twist with the advent of advanced layered-video encoding techniques such as Scalable Video Coding. In this paper, we show that the problem of finding the caching configuration of video encoding layers that minimizes delivery delay for a network operator is NP-Hard, and we establish a pseudopolynomial-time optimal solution by using a connection with the multiple-choice knapsack problem. Next, we design caching algorithms for multiple network operators that cooperate by pooling together their co-located caches, in an effort to aid each other, so as to avoid large delays due to fetching content from distant servers. We derive an approximate solution to this cooperative caching problem by using a technique that partitions the cache capacity into amounts dedicated to own and other operators' caching needs. Trace-driven evaluations demonstrate up to 25 percent reduction in delay over existing caching schemes. As a side benefit, our algorithms achieve smoother playback for video streaming applications, with fewer playback stalls and higher decoded quality.

Nowadays, there is a fast-paced shift from legacy telecommunication systems to novel software-defined network (SDN) architectures that can support on-the-fly network reconfiguration, therefore, empowering advanced traffic engineering mechanisms. Despite this momentum, migration to SDN cannot be realized at once especially in high-end networks of Internet service providers (ISPs). It is expected that ISPs will gradually upgrade their networks to SDN over a period that spans several years. In this paper, we study the SDN upgrading problem in an ISP network: which nodes to upgrade and when we consider a general model that captures different migration costs and network topologies, and two plausible ISP objectives: 1) the maximization of the traffic that traverses at least one SDN node, and 2) the maximization of the number of dynamically selectable routing paths enabled by SDN nodes. We leverage the theory of submodular and supermodular functions to devise algorithms with provable approximation ratios for each objective. Using real-world network topologies and traffic matrices, we evaluate the performance of our algorithms and show up to 54% gains over state-of-the-art methods. Moreover, we describe the interplay between the two objectives; maximizing one may cause a factor of 2 loss to the other. We also study the dual upgrading problem, i.e., minimizing the upgrading cost for the ISP while ensuring specific performance goals. Our analysis shows that our proposed algorithm can achieve up to 2.5 times lower cost to ensure performance goals over state-of-the-art methods.

Virtualized Radio Access Network (vRAN) architectures constitute a promising solution for the densification needs of 5G networks, as they decouple Base Stations (BUs) functions from Radio Units (RUs) allowing the processing power to be pooled at cost-efficient Central Units (CUs). vRAN facilitates the flexible function relocation (split selection), and therefore enables splits with less stringent network requirements compared to state-of-the-art fully Centralized (C-RAN) systems. In this paper, we study the important and challenging vRAN design problem. We propose a novel modeling approach and a rigorous analytical framework, FluidRAN, that minimizes RAN costs by jointly selecting the splits and the RUs-CUs routing paths. We also consider the increasingly relevant scenario where the RAN needs to support multi-access edge computing (MEC) services, that naturally favor distributed RAN (D-RAN) architectures. Our framework provides a joint vRAN/MEC solution that minimizes operational costs while satisfying the MEC needs. We follow a data-driven evaluation method, using topologies of 3 operational networks. Our results reveal that (i) pure C-RAN is rarely a feasible upgrade solution for existing infrastructure, (ii) FluidRAN achieves significant cost savings compared to D-RAN systems, and (iii) MEC can increase substantially the operator's cost as it pushes vRAN function placement back to RUs.

Fog architectures at the network edge are becoming a popular research trend to provide elastic resources and services to end-users, where the processing capacity resides at the network periphery as opposed to traditional data-centers. Despite their momentum, the control plane of these architectures remains complex and challenging to implement. To enhance control capability, in this work, we propose to use Software Defined Networking. SDN moves the control logic off data plane devices and onto external network entities, the controllers. We provide a proof-of-concept implementation of a multi-controller edge system and measure traffic delay and overheads. The results reveal the sensitivity of delay to the location of controllers and the magnitude of inter-controller and controller-node overheads. Guided by the above, we model the problem of determining the placement of controllers in the edge network. Using linearization and supermodular function techniques, we present approximation solutions which perform close to optimal and better than state-of-the-art methods.

Many mobile applications (abbrev. apps) reward the users who physically visit some locations tagged as POIs (places-of-interest) by the apps. In this paper, we study the POI-based collaboration between apps and venues (e.g., restaurants and cafes). On the one hand, an app charges a venue and tags the venue as a POI, which attracts users to visit the venue and potentially increases the venue's sales. On the other hand, the venue can invest in the app-related infrastructure (e.g., Wi-Fi networks and smartphone chargers), which enhances the users' experience of using the app. However, the existing POI pricing schemes of the apps (e.g., Pokemon Go and Snapchat) cannot incentivize the venue's infrastructure investment, and hence cannot achieve the most effective app-venue collaboration. We model the interactions among an app, a venue, and users by a three-stage Stackelberg game, and design an optimal two-part pricing scheme for the app. This scheme has a charge-with-subsidy structure: the app first charges the venue for becoming a POI, and then subsidizes the venue every time a user interacts with the POI. Compared with the existing pricing schemes, our two-part pricing better incentivizes the venue's investment, attracts more users to interact with the POI, and achieves a much larger app revenue. We analyze the impacts of the app's and venue's characteristics on the app's optimal revenrevenueue, and show that the apps with small and large congestion effects should collaborate with opposite types of venues.

The emergence of delay sensitive and computationally demanding data analytic applications has burdened the core network with huge data transfers and increased computation load. Furthermore, the increasing number of Internet of Things deployments rely significantly on the execution of such applications. We propose an architecture where devices collaboratively execute data analytic tasks in order to improve their execution delay and accuracy. This is possible by exploiting the aggregate computation capabilities of the abundance of small devices. We design an optimization framework where the nodes decide where their data analytic tasks will be executed, in order to jointly optimize their average execution delay and accuracy, while respecting power consumption constraints. We propose a distributed dual ascent solution to the formulated convex problem, so that the nodes can make the outsourcing decisions by exchanging local information. The results indicate that the nodes can achieve better performance when collaborating than when they locally compute the tasks, depending on the network load.

With elastic CDNs, content providers can rent cache space on demand at different cloud locations in order to enhance their offered quality of service (QoS). This paper addresses a key challenge in this context, namely how to invest an available budget in cache space in order to match spatio-temporal fluctuations of file demand and storage price. Specifically, we consider jointly dynamic cache rental, file placement, and request-cache association in a wireless scenario in order to provide a just-in-time CDN service. The objective is to maximize the benefit in average download delay obtained by the rented caches, while ensuring that the time-average rental cost is less than a fixed budget. We leverage a Lyapunov drift-minus-benefit technique to transform our infinite horizon problem into day-by-day subproblems which can be solved without knowledge of distant future file popularity and transmission rates. For the case of non-overlapping small cells (also wired case) we provide an efficient subproblem solution, referred to as JCC. However, in the general overlapping case, the subproblem becomes a mixed integer non-linear program (MINLP). In this case, we employ a dual decomposition method to derive a scalable solution, namely the JCCA algorithm. Finally, via extensive simulations, we reveal that the proposed JCCA algorithm attains 82.66 % higher delay benefit than existing static cache storage-based algorithms when available average cache budget is 20% of entire file library; moreover, the benefit becomes higher as the average cache budget gets tighter.

Virtualized radio access network (vRAN) architectures and multiple-access edge computing (MEC) systems constitute two key solutions for the emerging Tactile Internet applications and the increasing mobile data traffic. Their efficient deployment, however, requires a careful design tailored to the available network resources and user demand. In this paper, we propose a novel modeling approach and a rigorous analytical framework, MEC-vRAN joint design problem (MvRAN), that minimizes vRAN costs and maximizes MEC performance. Our framework selects jointly the base-station function splits, the fronthaul routing paths, and the placement of MEC functions. We follow a data-driven evaluation method, using topologies of three operational networks and experiments with a typical face-recognition MEC service. Our results reveal that MvRAN achieves significant cost savings (up to 2.5 times) compared to non-optimized centralized RAN or decentralized RAN systems, and MEC pushes the vRAN functions to radio units and hence can increase substantially the network cost.

This paper has the following ambitious goal: to convince the reader that content caching is an exciting research topic for the future communication systems and networks. Caching has been studied for more than 40 years, and has recently received increased attention from industry and academia. Novel caching techniques promise to push the network performance to unprecedented limits, but also pose significant technical challenges. This tutorial provides a brief overview of existing caching solutions, discusses seminal papers that open new directions in caching, and presents the contributions of this special issue. We analyze the challenges that caching needs to address today, also considering an industry perspective, and identify bottleneck issues that must be resolved to unleash the full potential of this promising technique.

Welcome to the second part of the IEEE JSAC special issue on Caching for Communication Systems and Networks. The goal of this special issue is to present the multiple facets of caching, from information theory to networking and services, and explore the role of memory in communications. This is a very timely topic due to recent technological and theoretical advances summarized in the tutorial paper that appears in the first part of the issue[1].

Fog architectures at the network edge are becoming a popular research trend to provide elastic resources and services to end-users, where the processing capacity resides at the network periphery as opposed to traditional data-centers. Despite their momentum, the control plane of these architectures remains complex and challenging to implement. To enhance control capability, in this paper, we propose to use software defined networking (SDN). SDN moves the control logic off data plane devices and onto external network entities, the controllers. We provide a proof-of-concept implementation of a multi-controller edge system and measure traffic delay and overheads. The results reveal the sensitivity of delay to the location of controllers and the magnitude of inter-controller and controller-node overheads. Guided by the above, we model the problem of determining the placement of controllers in the edge network. Using linearization and supermodular function techniques, we present approximation solutions which perform close to optimal and substantially better than state-of-the-art methods. Finally, we analyze the interplay between various performance and reliability objectives.

Participatory sensing is a paradigm through which mobile device users (or participants) collect and share data about their environments. The data captured by participants is typically submitted to an intermediary (the service provider) who will build a service based upon this data. For a participatory sensing system to attract the data submissions it requires, its users often need to be incentivized. However, as an environment is constantly changing (for example, an accident causing a buildup of traffic and elevated pollution levels), the value of a given data item to the service provider is likely to change significantly over time, and therefore an incentivization scheme must be able to adapt the rewards it offers in real‐time to match the environmental conditions and current participation rates, thereby optimizing the consumption of the service provider’s budget. This paper presents adaptive reward allocation (ARA), which uses the Lyapunov Optimization method to provide adaptive reward allocation that optimizes the consumption of the service provider’s budget. ARA is evaluated using a simulated participatory sensing environment with experimental results showing that the rewards offered to participants are adjusted so as to ensure that the data captured matches the dynamic changes occurring in the sensing environment and takes the response rate into account while also seeking to optimize budget consumption.

Nowadays, there is a fast-paced shift from legacy telecommunication systems to novel Software Defined Network (SDN) architectures that can support on-the-fly network reconfiguration, therefore, empowering advanced traffic engineering mechanisms. Despite this momentum, migration to SDN cannot be realized at once especially in high-end cost networks of Internet Service Providers (ISPs). It is expected that ISPs will gradually upgrade their networks to SDN over a period that spans several years. In this paper, we study the SDN upgrading problem in an ISP network: which nodes to upgrade and when. We consider a general model that captures different migration costs and network topologies, and two plausible ISP objectives; first, the maximization of the traffic that traverses at least one SDN node, and second, the maximization of the number of dynamically selectable routing paths enabled by SDN nodes. We leverage the theory of submodular and supermodular functions to devise algorithms with provable approximation ratios for each objective. Using real-world network topologies and traffic matrices, we evaluate the performance of our algorithms and show up to 54% gains over state-of-the-art methods. Moreover, we describe the interplay between the two objectives; maximizing one may cause a factor of 2 loss to the other.

Motivated by the recent efforts in extending long term evolution (LTE) to the unlicensed spectrum, we propose a novel spectrum sharing framework for the coopetition (i.e., cooperation and competition) between LTE and Wi-Fi in the unlicensed band. Basically, the LTE network can choose to work in one of the two modes: in the competition mode, it randomly accesses an unlicensed channel, and interferes with the Wi-Fi access point using the same channel; in the cooperation mode, it onloads the Wi-Fi users' traffic in exchange for the exclusive access of the corresponding channel. We design a second-price reverse auction mechanism, which enables the LTE provider and the Wi-Fi access point owners (APOs) to effectively negotiate the operation mode. Specifically, the LTE provider is the auctioneer (buyer), and the APOs are the bidders (sellers) who compete to sell the rights of onloading the APOs' traffic to the LTE provider. In Stage I of the auction, the LTE provider announces a reserve rate, which is the maximum data rate that it is willing to allocate to the APOs in the cooperation mode. In Stage II of the auction, the APOs submit their bids, which indicate the data rates that they would like the LTE provider to offer in the cooperation mode. We show that the auction involves allocative externalities, i.e., the cooperation between the LTE provider and one APO benefits other APOs who are not directly involved in this cooperation. We characterize the APOs' unique equilibrium bidding strategies in Stage II, and analyze the LTE provider's optimal reserve rate in Stage I. Numerical results show that our framework improves the payoffs of both the LTE provider and the APOs comparing with a benchmark scheme. In particular, our framework increases the LTE provider's payoff by 70% on average, when the LTE provider has a large throughput and a small data rate discounting factor. Moreover, our framework leads to a close-to-optimal social welfare under a large LTE throughput.

The surging global mobile data traffic challenges the economic viability of cellular networks and calls for innovative solutions to reduce the network congestion and improve user experience. In this context, user-provided networks (UPNs), where mobile users share their Internet access by exploiting their diverse network resources and needs, turn out to be very promising. Heterogeneous users with advanced handheld devices can form connections in a distributed fashion and unleash dormant network resources at the network edge. However, the success of such services heavily depends on users' willingness to contribute their resources, such as network access and device battery energy. In this paper, we introduce a general framework for UPN services and design a bargaining-based distributed incentive mechanism to ensure users' participation. The proposed mechanism determines the resources that each user should contribute in order to maximize the aggregate data rate in UPN, and fairly allocate the benefit among the users. The numerical results verify that the service can always improve users' performance, and such improvement increases with the diversity of the users' resources. Quantitatively, it can reach an average 30% increase of the total served traffic for a typical scenario even with only six mobile users.

Recent technological advances have rendered storage a readily available resource, yet there exist few examples that use it for enhancing network performance. We revisit in-network storage and we evaluate its usage as an additional degree of freedom in network optimization. We consider the network design problem of maximizing the volume of end-to-end transferred data and we derive storage allocation (placement) solutions. We show that different storage placements have different impact on the performance of the network and we introduce a systematic methodology for the derivation of the optimal one. Accordingly, we provide a framework for the joint optimization of routing and storage control (usage) in dynamic networks for the case of a single commodity transfer. The derived policies are based on time-expanded graphs and ensure maximum performance improvement with minimum possible storage usage. We also study the respective multiple commodity problem, where the network link capacities and node storage resources are shared by the different commodities. A key advantage of our methodology is that it employs algorithms that are applicable to both centralized as well as to distributed execution in an asynchronous fashion, and thus, no tight synchronization is required among the various involved storage and routing devices in an operational network. We also present an extensive performance evaluation study using the backbone topology and actual traffic traces from a large European Internet Service Provider, and a number of synthetic network topologies. Our results show that indeed our approach offers significant improvements in terms of delivery time and transferred traffic volume.

Distributed caching architectures have been proposed for bringing content close to requesters and the key problem is to design caching algorithms for reducing content delivery delay. The problem obtains an interesting new twist with the advent of advanced layered-video encoding techniques such as Scalable Video Coding (SVC). We show that the problem of finding the caching configuration of video encoding layers that minimizes average delay for a network operator is NP-Hard, and we establish a pseudopolynomial-time optimal solution using a connection with the multiple-choice knapsack problem. We also design caching algorithms for multiple operators that cooperate by pooling together their co-located caches, in an effort to aid each other, so as to avoid large delays due to downloading content from distant servers. We derive an approximate solution to this cooperative caching problem using a technique that partitions the cache capacity into amounts dedicated to own and others' caching needs. Numerical results based on real traces of SVC-encoded videos demonstrate up to 25% reduction in delay over existing (layer-agnostic) caching schemes, with increasing gains as the video popularity distribution gets steeper, and cache capacity increases.

WiFi data offloading provides a promising auxiliary to alleviate network congestion by diverting traffic from the cellular infrastructure onto WiFi access points (APs). Despite the importance and momentum of this method, the current deployment of APs by the carriers follows mostly a heuristic approach. In addition, the prevalent free-of-charge WiFi access approach may result in significant opportunity costs for the carriers as this traffic could yield non-negligible revenues. In this paper, we propose and study the problem of optimizing the deployment of WiFi offloading infrastructure, and pricing the offloading service with the goal of maximizing carrier profits. Addressing this problem is a prerequisite for the efficient integration of WiFi technology to next generation of cellular systems and the development of carrier-grade offloading solutions. Our framework considers a fundamental, intuitive model of carrier costs and revenues, and two demand models that predict how traffic will change in response to alteration in the price and the set of deployed APs. We present both analytical and approximate solutions for this intricate problem, and reveal how key network parameters shape the offloading benefits. Using a dataset of WiFi access patterns collected from real users, we evaluate the impact of offloading for different regional markets around the world. We find that in mature markets WiFi can help carriers reduce their costs, while charging users up to 50% lower than the cellular service. The gains are higher for small "virtual carriers" who resell other's mobile data services (up to a factor of 2). However, in less mature markets where the AP deployment or access costs are higher, deploying APs can actually lead to a net loss for the carrier. Our evaluation code is publicly available for the benefit of research community.

Motivated by the recent proliferation of advanced handheld devices and the unprecedented growth of mobile data traffic, this paper proposes the concept of Mobile edge-Networks (MeNs), a solution that leverages the end-user devices to enhance the performance of emerging 5G systems. MeNs enable mobile users to collaborate with each other and address in a bottom-up fashion key problems in wireless systems, such as poor channel conditions. We design a dynamic cooperation policy that determines transmission parameters of the network in a utility-optimal fashion, ensuring that no user performs worse than she would without cooperation and that the benefits from the collaboration are shared among the users.

Motivated by the recent efforts in extending LTE to the unlicensed spectrum, we propose a novel spectrum sharing framework for the coopetition (i.e., cooperation and competition) between LTE and Wi-Fi in the unlicensed band. Basically, the LTE network chooses to work in one of the two modes: in the competition mode, it randomly accesses an unlicensed channel, and interferes with a Wi-Fi access point; in the cooperation mode, it onloads a Wi-Fi access point's traffic in exchange for the full access of the corresponding channel. Because the LTE network works in an interference-free manner in the cooperation mode, it can achieve a much larger total data rate (comparing to the competition mode) to serve both its own users and the Wi-Fi users under proper channel conditions. To achieve the maximum potential of this novel coopetition framework, we design a reverse auction mechanism, where the LTE provider is the auctioneer (buyer), and the Wi-Fi access point owners (APOs) are the bidders who compete to sell their channels to the LTE provider. An APO's bid indicates the data rate that it would like the LTE provider to offer in the cooperation mode. We show that the auction involves the allocative externalities, i.e., the cooperation between the LTE provider and an APO benefits other APOs who are not directly involved in this cooperation. As a result, a particular APO's bidding strategy is affected by its belief about other APOs' bidding strategies. This makes our analysis much more challenging than that of the standard second-price auction, where bidding truthfully is a weakly dominant strategy. We characterize the APOs' unique equilibrium bidding strategies, and analyze the LTE provider's optimal reserve rate that maximizes its payoff for a general APO type distribution. Our analysis shows that only when the LTE throughput exceeds a threshold, the LTE provider will choose a reasonably large reserve rate to cooperate with the APOs; otherwise, it will restrict the reserve rate to a small value and work in the competition mode.

The liberalization of wireless communication services markets and the subsequent competition among network operators, is expected to foster optimal utilization of the scarce wireless spectrum and ensure the provision of cost-efficient services to users. However, such markets may function inefficiently due to collusion of operators which yields a de-facto monopoly. Although it is illegal and detrimental to the users, creation of such cartels arises often in the form of implicit price fixing. In this paper, we consider a general such market where a set of operators sell communication services to a large population of users. We use an evolutionary game model to capture the user dynamics in selecting operators, under limited information about the actual service quality, and we analyze the anticipated interaction of the operators using coalitional game theory. We define a coalition formation game in order to rigorously study the conditions that render monopolistic or oligopolistic markets stable under different notions of coalition stability. We also provide direct and indirect regulation methods, such as setting price upper bounds or allocating different amounts of spectrum, in order to discourage undesirable equilibriums. Our approach provides intuitions about collusion strategies, as well as on directions for identifying and preventing them.

As the ever growing mobile data traffic challenges the economic viability and performance of cellular networks, innovative solutions that harvest idle user-owned network resources are gaining increasing interest. In this work, we propose leasing wireless bandwidth and cache space of residential 802.11 (WiFi) access points (APs) for offloading mobile data. This solution not only reduces cellular network congestion, but, due to caching, improves also the user-perceived network performance without overloading the backhaul links of the APs. To encourage residential users to contribute their bandwidth and cache resources, we design monetary incentive (reimbursement) schemes. The offered reimbursements directly determine the amounts of available bandwidth and cache space in every AP, which in turn affect the caching policy (where to cache each content file) and the routing policy (where to route each mobile data request). In order to reduce operator's total cost for serving mobile data requests and leasing resources, we introduce a framework for the joint optimization of incentive, caching, and routing policies. Using a novel WiFi usage dataset collected from 167 residences, we show that in densely populated areas with relatively costly network capacity upgrades, our proposal can halve operator's total cost, while reimbursing up to 9€ per month each residential user.

As the growth of mobile data traffic places significant strain on cellular networks, plans for exploiting under-utilized network resources become increasingly attractive. In this paper, we propose, design, and evaluate a data offloading architecture, where mobile users are offloaded to mesh networks, which are built and managed by residential users. Such networks are often developed in the context of community networks or, recently, as commercial services. Mobile network operators can lease capacity from these networks and offload traffic to reduce their servicing costs. We introduce an analytical framework that determines the offloading policy, i.e., which mobile users should be offloaded, based on the energy cost induced to the cellular base stations. Accordingly, we design a minimum-cost servicing policy for the mesh networks. Clearly, such architectures are realizable only if the mesh nodes agree with each other to jointly serve the offloaded traffic. To achieve this, we employ the Shapley value rule for dispensing the leasing payment among the mesh nodes. We evaluate this paper by simulating the operation of the LTE-A network, and conducting test bed experiments for the mesh network. The results reveal significant savings for eNBs power consumption and reimbursements for mesh users.

Base station sharing is currently considered one of the most promising solutions for reducing the energy consumption costs of cellular networks. This paper presents a game theoretic framework for the study of such cooperative solutions where different mobile network operators (MNOs) decide to switch off subsets of their base stations during off-peak hours and roam their traffic to the remaining stations. The solution is based on a detailed optimization framework that determines exactly which base stations should remain active and how much traffic each one of them should serve, so as to maximize the aggregate energy savings. Accordingly, using the axiomatic Shapley value rule, it is determined how the benefits from the cooperation, i.e., the cost savings, should be dispersed among the cooperating MNOs. It is proved that this coalitional game with transferrable utilities has a nonempty core, and thus there exists a cooperation solution that incentivizes the participation of all operators. Moreover, using a thorough numerical analysis, it is shown that the benefits achieved with the implementation of the cooperation strategy depend mainly on the power consumption characteristics of the MNOs, which in turn are related to the number, type, and technology of their base stations. Overall, the energy savings are found to be most sensitive to the technology of the used base stations, and more precisely to the no-load base station energy consumption which defines the energy waste in a network.

The landscape toward 5G wireless communication is currently unclear, and, despite the efforts of academia and industry in evolving traditional cellular networks, the enabling technology for 5G is still obscure. This paper puts forward a network paradigm toward next-generation cellular networks, targeting to satisfy the explosive demand for mobile data while minimizing energy expenditures. The paradigm builds on two principles; namely caching and multicast. On one hand, caching policies disperse popular content files at the wireless edge, e.g., pico-cells and femto-cells, hence shortening the distance between content and requester. On other hand, due to the broadcast nature of wireless medium, requests for identical files occurring at nearby times are aggregated and served through a common multicast stream. To better exploit the available cache space, caching policies are optimized based on multicast transmissions. We show that the multicast-aware caching problem is NP-hard and develop solutions with performance guarantees using randomized-rounding techniques. Trace-driven numerical results show that in the presence of massive demand for delay tolerant content, combining caching and multicast can indeed reduce energy costs. The gains over existing caching schemes are 19% when users tolerate delay of three minutes, increasing further with the steepness of content access pattern.

Heterogeneous cellular networks that overlay cache-endowed small-cell networks with macro-cell networks have emerged as a promising solution towards ultra-low latency, extra-high throughput, and sub-multiple energy consumption compared to the conventional cellular paradigm. A technique to further improve the energy efficiency of such multi-tier networks is to apply activation mechanisms, that dynamically power on/off a subset of the small-cells and macro-cells. In this work, we show that the activation policy should be jointly derived with the caching policy, that places popular content files at the base station caches. As a result content is fetched effectively closer to the mobile end-users and can be transported via energy-prudent links, while at the same time as many as possible of the base stations are powered off. We then formulate the energy-minimizing problem, which is NP-hard, and introduce a novel approximation framework for its efficient solution. Numerical results, that are based on system parameters driven from real trace datasets, show that our approach provides an excellent performance that is far better than the schemes that perform caching and base station activation in a disjoint manner.

The recent mobile data explosion has increased the interest for mobile user-provided networks (MUPNs), where users share their Internet access by exploiting the diversity in their needs and resource availability. Although promising, MUPNs raise unique challenges. Namely, the success of such services relies on user participation which in turn can be achieved on the basis of a fair and efficient resource (i.e., Internet access and battery energy) exchange policy. The latter should be devised and imposed in a very fast time scale, based on near real-time feedback from mobile users regarding their needs, resources, and network conditions that are rapidly changing. To address these challenges we design and implement a novel cloud-controlled MUPN system, that employs software defined networking support on mobile terminals, to dynamically apply data forwarding policies with adaptive flow-control. We devise these policies by solving a coalitional game that is played among the users. We prove that the game has a non-empty core and hence the solution, which determines the servicing policy, incentivizes the users to participate. Finally, we evaluate the performance of the service in a prototype, where we investigate its performance limits, quantify the implementation overheads, and justify our architecture design choices.

Exchange of services and resources in, or over, networks is attracting nowadays renewed interest. However, despite the broad applicability and the extensive study of such models, e.g., in the context of P2P networks, many fundamental questions regarding their properties and efficiency remain unanswered. We consider such a service exchange model and analyze the users' interactions under three different approaches. First, we study a centrally designed service allocation policy that yields the fair total service each user should receive based on the service it offers to the others. Accordingly, we consider a competitive market where each user determines selfishly its allocation policy so as to maximize the service it receives in return, and a coalitional game model where users are allowed to coordinate their policies. We prove that there is a unique equilibrium exchange allocation for both game theoretic formulations, which also coincides with the central fair service allocation. Furthermore, we characterize its properties in terms of the coalitions that emerge and the equilibrium allocations, and analyze its dependency on the underlying network graph. That servicing policy is the natural reference point to the various mechanisms that are currently proposed to incentivize user participation and improve the efficiency of such networked service (or, resource) exchange markets.

In this paper we present an optimization framework that formalizes the inherent trade-off between the user perceived quality of wireless video, and the energy consumption cost of the network. The former is formulated in the context of the emerging heterogeneous cellular networks (HCN) based on LTE. We also consider users that employ dynamic adaptive streaming over HTTP (DASH). Our framework quantifies this trade-off carefully, by delving into the details of DASH, the LTE network, and the HCN architecture. The result is a complex problem that is solved in two levels. The master problem is responsible for decisions regarding the the user association and the average power they are allocated. The solution of this problem also entails a decision about the encoding rate of the DASH video segments. The previous decision is used in order to perform resource allocation at a finer level by considering the technical details of LTE that allocates resource blocks and power simultaneously. Numerical results are presented with realistic parameters for the LTE network and the video traffic.

The unprecedented growth of mobile data traffic challenges the performance and economic viability of today's cellular networks and calls for novel network architectures and communication solutions. Mobile data offloading through third-party Wi-Fi or femtocell access points (APs) can significantly alleviate the cellular congestion and enhance user quality of service (QoS), without requiring costly and time-consuming infrastructure investments. This solution has substantial benefits both for the mobile network operators (MNOs) and the mobile users, but comes with unique technical and economic challenges that must be jointly addressed. In this paper, we consider a market where MNOs lease APs that are already deployed by residential users for the offloading purpose. We assume that each MNO can employ multiple APs, and each AP can concurrently serve traffic from multiple MNOs. We design an iterative double-auction mechanism that ensures the efficient operation of the market by maximizing the differences between the MNOs' offloading benefits and APs' offloading costs. The proposed scheme takes into account the particular characteristics of the wireless network, such as the coupling of MNOs' offloading decisions and APs' capacity constraints. Additionally, it does not require full information about the MNOs and APs and creates nonnegative revenue for the market broker.

As the unprecedented growth of mobile data traffic places significant strain on cellular networks, alternative plans for exploiting already existing and under-utilized wireless infrastructure, become quite attractive. In this paper, we study cellular-to-mesh (C2M) data offloading for LTE-A cellular mobile users to WiFi mesh networks, which are built and managed collaboratively by users. Such networks are developed in the context of community networks or, recently, as commercial services among residential users. Mobile network operators can lease these mesh networks to offload their traffic and reduce their servicing cost. In this context, we introduce an analytical framework that determines which mobile users should be offloaded, based on the energy cost incurred to the cellular base stations (eNB) for serving their demands. Accordingly, we design a routing policy that the mesh network can employ so as to serve the offloaded traffic with the minimum possible cost. Moreover, the reimbursement offered by the operator should be dispensed to the different mesh users, according to their contribution and added-value significance. We address this issue by employing the Shapley value profit sharing rule, which ensures the participation of the mesh nodes in this joint task. We evaluate our work by simulating the operation of the LTE-A network, and conducting testbed experimentation for the mesh network. The results reveal significant savings for eNBs power consumption and compensation profits for mesh users.

User-provided connectivity (UPC) is a promising paradigm to achieve a low-cost ubiquitous connectivity. In this paper, we study a network-assisted UPC service model, where a mobile virtual network operator (MVNO) enables its subscribers to operate as mobile WiFi hotspots (hosts) and provide Internet connectivity for others. A unique aspect of this service model is that the MVNO offers some free data quota to hosts as reimbursements (incentives) for connectivity sharing. This reimbursing scheme, together with a usage-based pricing, constitute a revolutionary hybrid data pricing-reimbursing scheme, which has not been considered before. We analyze the different impacts of data price and reimbursement on the host's connectivity sharing decision systematically. Based on this analysis, we further derive the optimal hybrid pricing-reimbursing policy that maximizes the MVNO's revenue. Our numerical result indicates that by using the proposed hybrid pricing policy, the MVNO can increase its revenue by 20% to 135% under an elastic client demand, and by 20% to 550% under an inelastic client demand, comparing to those achieved under a pricing-only policy.

Crowd-sourced mobile Internet access services enable mobile users to connect with each other and share their Internet connections. This is a promising solution for addressing users' increasing needs for ubiquitous connectivity and alleviating network congestion. The success of such services heavily depends on users' willingness to contribute their resources. In this paper, we consider a general model for such services, and design a distributed incentive mechanism for encouraging users' participation. This bargaining based scheme ensures that the contribution of user resources, in terms of Internet access bandwidths and battery energy, and the allocation of service capacity, measured in the delivered mobile data, are Pareto efficient and proportionally fair. The numerical results verify that the service always improves users' performance and that these benefits depend on the diversity of the users' resources.

Video delivery to mobile users is one of the largest challenges that network operators face today. In this work we consider a heterogeneous cellular network with storage capable small-cell base stations, and study this problem for pre-stored video files that can be encoded with two different schemes, namely versions or layers, in various qualities. We introduce a framework for the joint derivation of video caching and routing policies for users with different quality requirements. This allows the operator to optimize a balanced objective of incurred servicing cost, and users experienced delay, according to his priorities. The numerical results indicate that versions and layers may have different impact on the delay and servicing cost, depending on the diversity of users' demand, and that the cost-delay trade off is affected by the network's load.

Cache-endowed small cell networks constitute a timely and effective solution for mobile network operators (MNOs) who strive to serve the massive content demand of the mobile users. However, the deployment of small cell base stations (SBSs) requires significant economic investments, while site acquisition issues render it infeasible in several cases. Yet, one potentially explosive factor for network expansion remains unexploited, namely, an increasing number of residential users install in their premises privately-owned SBSs (femtocell or WiFi access points) in order to serve their own needs. In this work, we envision an MNO offering incentives to the SBS owners to cache and deliver content items requested by the nearby mobile users. We model the interaction between the MNO and the SBS owners as a Stackelberg game, and show that the incentive design problem requires to know the content caching policy, which in turn should be jointly derived with the request routing policy. We then introduce a framework for the joint derivation of incentive, caching and routing policies. Numerical results indicate that our mechanism provides a substantial potential for reducing the MNO's costs, depending on the willingness of the SBS owners to lease their resources and the spatio temporal characteristics of the mobile data demand.

Experimentation in testbeds is gaining increasing ground as a necessary validation step for every theoretical study in communication networks. However, the first-come-first-served policy employed today by most testbeds does not ensure the fair and efficient utilization of their resources, which often lie idle. Ideally, every testbed should be utilized as much as possible and serve the most important requests. In this paper we introduce a novel resource scheduling mechanism for the wireless testbed NITOS. The proposed scheme is based on VCG auctions and includes an allocation and a pricing rule which induce the users to judiciously submit experiment requests. We prove theoretically and demonstrate numerically that this scheme ensures that the testbed resources (nodes and channels) are assigned to the users with the highest needs. Our mechanism can be incorporated in the next generation resource management systems for NITOS and similar testbeds.

The deployment of small cells is expected to gain huge momentum in the near future, as a solution for managing the skyrocketing mobile data demand growth. Local caching of popular files at the small cell base stations has been recently proposed, aiming at reducing the traffic incurred when transferring the requested content from the core network to the users. In this paper, we propose and analyze a novel caching approach that can achieve significantly lower traffic compared to the traditional caching schemes. Our cache design policy carefully takes into account the fact that an operator can serve the requests for the same file that happen at nearby times via a single multicast transmission. The latter incurs less traffic as the requested file is transmitted to the users only once, rather than with many unicast transmissions. Systematic experiments demonstrate the effectiveness of our approach, as compared to the existing caching schemes.

The unprecedented growth of mobile data traffic challenges the performance and economic viability of today's cellular networks and calls for novel network architectures and communication solutions. Data offloading through third-party WiFi or femtocell access points (APs) can effectively alleviate the cellular network congestion in low operational and capital expenditure. This solution requires the cooperation and agreement of mobile cellular network operators (MNOs) and AP owners (APOs). In this paper, we model and analyze the interaction among one MNO and multiple APOs (for the amount of MNO's offloading data and the respective APOs' compensations) by using thew Nash bargaining theory. Specifically, we introduce a one-to-many bargaining game among the MNO and APOs and analyze the bargaining solution (game equilibrium) systematically under two different bargaining protocols: 1) sequential bargaining, where the MNO bargains with APOs sequentially, with one APO at a time, in a given order; and 2) concurrent bargaining, where the MNO bargains with all APOs concurrently. We quantify the benefits for APOs when bargaining sequentially and earlier with the MNO, and the losses for APOs when bargaining concurrently with the MNO. We further study the group bargaining scenario where multiple APOs form a group bargaining with the MNO jointly and quantify the benefits for APOs when forming such a group. Interestingly, our analysis indicates that grouping of APOs not only benefits the APOs in the group but may also benefit some APOs not in the group. Our results shed light on the economic aspects and the possible outcomes of the MNO/APOs interactions and can be used as a roadmap for designing policies for this promising data offloading solution.

Small cells constitute a promising solution for managing the mobile data growth that has overwhelmed network operators. Local caching of popular content items at the small cell base stations (SBSs) has been proposed to decrease the costly transmissions from the macrocell base stations without requiring high capacity backhaul links for connecting the SBSs with the core network. However, the caching policy design is a challenging problem especially if one considers realistic parameters such as the bandwidth capacity constraints of the SBSs that can be reached in congested urban areas. We consider such a scenario and formulate the joint routing and caching problem aiming to maximize the fraction of content requests served locally by the deployed SBSs. This is an NP-hard problem and, hence, we cannot obtain an optimal solution. Thus, we present a novel reduction to a variant of the facility location problem, which allows us to exploit the rich literature of it, to establish algorithms with approximation guarantees for our problem. Although the reduction does not ensure tight enough bounds in general, extensive numerical results reveal a near-optimal performance that is even up to 38% better compared to conventional caching schemes using realistic system settings.

Small cells constitute a promising solution for managing the mobile data growth that has overwhelmed network operators. Local caching of popular content items at the small cell base stations has been proposed in order to decrease the capacity-and hence the cost- of the backhaul links that connect these base stations with the core network. However, deriving the optimal caching policy remains a challenging open problem especially if one considers realistic parameters such as the bandwidth limitation of the base stations. The latter constraint is particularly important for cases when users requests are massive. We consider such a scenario and formulate the joint caching and routing problem aiming to maximize the fraction of content requests served by the deployed small cell base stations. This is an NP-hard problem and hence we cannot obtain an exact optimal solution. Thus, we present a novel approximation framework based on a reduction to a well known variant of the facility location problem. This allows us to exploit the rich literature in facility location problems, in order to establish bounded approximation algorithms for our problem.

Mobile data offloading is a promising approach to alleviate network congestion and enhance quality of service (QoS) in mobile cellular networks. In this paper, we investigate the economics of mobile data offloading through third-party WiFi or femtocell access points (APs). Specifically, we consider a market-based data offloading solution, where macrocellular base stations (BSs) pay APs for offloading traffic. The key questions arising in such a marketplace are following: (i) how much traffic should each AP offload for each BS? and (ii) what is the corresponding payment of each BS to each AP? We answer these questions by using the non-cooperative game theory. In particular, we define a multi-leader multi-follower data offloading game (DOFF), where BSs (leaders) propose market prices, and accordingly APs (followers) determine the traffic volumes they are willing to offload. We characterize the subgame perfect equilibrium (SPE) of this game, and further compare the SPE with two other classic market outcomes: (i) the market balance (MB) in a perfect competition market (i.e., without price participation), and (ii) the monopoly outcome (MO) in a monopoly market (i.e., without price competition). Our results analytically show that (i) the price participation (of BSs) will drive market prices down, compared to those under the MB outcome, and (ii) the price competition (among BSs) will drive market prices up, compared to those under the MO outcome.

Recent technological advances have rendered storage a cheap and at large scale available resource. Yet, there exist only few examples in networking that consider storage for enhancing data transfer capabilities. In this paper we study networks with time varying link capacity and analyze the impact of node storage on their capability to convey data from source to destination. We show that storage capacity is quite beneficial in terms of the amount of data that can be pushed from the source to the destination within a given time horizon. Equivalently, storage can be used to reduce incurred delay for the delivery of a certain amount of data. For linear networks, we show that this performance improvement depends on the relative patterns of link capacity variations. We extend our study to general networks and we use a novel method that iteratively updates the minimum cut of the time expanded graph, in a constructive manner, in the sense that during the process, the storage capacity allocation in the network is shown. Next, we incorporate routing in our methodology and derive a joint storage capacity management and routing policy to maximize the amount of data transferred to the destination. This policy stems from the solution of the maximum flow problem defined for the dynamic network over a certain time period, by using the ε-relaxation solution method. The later is amenable to distributed implementation, which is a very desirable property for the large scale modern networks which operate without central control.

Auction mechanisms are used for allocating a resource among multiple agents with the objective to maximize social welfare. What makes auctions attractive is that they are agnostic to utility functions of agents. Auctions involve a bidding method by agents-buyers, which is then mapped by a central controller to an allocation and a payment for each agent. In autonomic networks comprising self-interested nodes with different needs and utility functions, each entity possesses some resource and can engage in transactions with others to achieve its needs. In fact, efficient network operation relies on node synergy and multi-lateral resource trading. Nodes face the dilemma of devoting their limited resource to their own benefit versus acting altruistically and anticipating to be aided in the future. Wireless ad-hoc networks, peer-to-peer networks and disruption-tolerant networks are instances of autonomic networks where the challenges above arise and the traded resource is energy, bandwidth and storage space respectively. Clearly, the decentralized complex node interactions and the double node role as resource provider and consumer amidst resource constraints cannot be addressed by single-sided auctions and even more by mechanisms with a central controller. We introduce a double-sided auction market framework to address the challenges above. Each node announces one bid for buying and one for selling the resource. We prove that there exist bidding and charging strategies that maximize social welfare and we explicitly compute them. We generalize our result to a generic network objective. Nodes are induced to follow these strategies, otherwise they are isolated by the network. Furthermore, we propose a decentralized realization of the double-sided auction with lightweight network feedback. Finally, we introduce a pricing method which does not need a charging infrastructure. Simulation results verify the desirable properties of our approach.

Peer-to-peer networks are voluntary resource sharing systems among rational agents that are resource providers and consumers. While altruistic resource sharing is necessary for efficient operation, this can only be imposed by incentive mechanisms, otherwise peers tend to behave selfishly. Selfishness in general terms means only consuming resources in order to absorb maximal utility from them and not providing resources to other peers because this would require effort and would not give any utility. In peer-to- peer networks, this behavior, known as free riding, amounts to only downloading content from others and leads to system performance degradation. In this work, we consider a reputation-based mechanism for providing incentives to peers for resource provisioning besides resource consuming. We consider networks where the access technology does not separate upstream and downstream traffic, and these flow through the same capacity-limited access link. Peers do not know other peers' strategies and their intentions to conform to the protocol and share their resources. A separate utility maximization problem is solved by each peer, where the peer allocates a portion of its link bandwidth to its own downloads, acting as client, and it also allocates the remaining bandwidth for serving requests made to it by other peers. The optimization is carried out under a constraint on the level of dissatisfaction the peer intends to cause by not fulfilling others' requests. This parameter is private information for each peer. The reputation of a peer as a server is updated based on the amount of allocated bandwidth compared to the requested one. Reputation acts towards gradually revealing hidden intentions of peers and accordingly guiding the resource allocation by rewarding or penalizing peers in subsequent bandwidth allocations. Our results confirm that the reputation mechanism discourages selfish behavior and drives the system to a state where each peer obtains utility in accordance to its hidden intentions in dissatisfying others.