Recently, with big data and high computing power, deep learning models have achieved high accuracy in prediction problems. However, the challenging issues of utilizing deep learning into the content's popularity prediction remains open. The first issue is how to pick the best-suited neural network architecture among the numerous types of deep learning architectures (e.g., Feed-forward Neural Networks, Recurrent Neural Networks, etc.). The second issue is how to optimize the hyperparameters (e.g., number of hidden layers, neurons, etc.) of the chosen neural network. Therefore, we propose the reinforcement (Q-Learning) meta-learning based deep learning model deployment scheme to construct the best-suited model for predicting content's popularity autonomously. Also, we added the feedback mechanism to update the Q-Table whenever the base station calibrates the model to find out more appropriate prediction model. The experiment results show that the proposed scheme outperforms existing algorithms in many key performance indicators, especially in content hit probability and access delay.

Device to device (D2D) communication is considered as an effective technology for enhancing the spectral efficiency and network throughput of existing cellular networks. However, enabling it in an underlay fashion poses a significant challenge pertaining to interference management. In this paper, mode selection and resource allocation for an underlay D2D network is studied while simultaneously providing interference management. The problem is formulated as a combinatorial optimization problem whose objective is to maximize the utility of all D2D pairs. To solve this problem, a learning framework is proposed based on a problem-specific Markov chain. From the local balance equation of the designed Markov chain, the transition probabilities are derived for distributed implementation. Then, a novel two phase algorithm is developed to perform mode selection and resource allocation in the respective phases. This algorithm is then shown to converge to a near optimal solution. Moreover, to reduce the computation in the learning framework, two resource allocation algorithms based on matching theory are proposed to output a specific and deterministic solution. The first algorithm employs the one-to-one matching game approach whereas in the second algorithm, the one-to many matching game with externalities and dynamic quota is employed. Simulation results show that the proposed framework converges to a near optimal solution under all scenarios with probability one. Moreover, our results show that the proposed matching game with externalities achieves a performance gain of up to 35 percent in terms of the average utility compared to a classical matching scheme with no externalities.

The use of heterogeneous small cell-based networks to offload the traffic of existing cellular systems has recently attracted significant attention. One main challenge is solving the joint problems of interference mitigation, user association, and resource allocation. These problems are formulated as an optimization which is then analyzed using two different approaches: Markov approximation and log-linear learning. However, finding the optimal solutions of both approaches requires complete information of the whole network which is not scalable with the network size. Thus, an approach based on a Markov approximation with a novel Markov chain design and transition probabilities is proposed. This approach enables the Markov chain to converge to the bounded near optimal distribution without complete information. In the game-theoretic approach, the payoff-based log-linear learning is used, and it converges in probability to a mixed-strategy ε-Nash equilibrium. Based on the principles of these two approaches, a highly randomized self-organizing algorithm is proposed to reduce the gap between optimal and converged distributions. Simulation results show that all of the proposed algorithms effectively offload more than 90 percent of the traffic from the macrocell base station to small cell base stations. Moreover, the results also show that the algorithms converge quickly irrespective of the number of possible configurations.

The use of heterogeneous small cell-based networks to offload the traffic of existing cellular systems has recently attracted significant attention. One main challenge is solving the joint problems of user association, resource allocation, and interference mitigation. The goal of this paper is to design a self-organizing algorithm that can solve these problems, simultaneously. To this end, this joint resource allocation problem is formulated as an optimization problem which is then solved using log-sum-exp approximation. This solution is then shown to require complete information of the whole network which is not scalable with the network size. To address this scalability issue, a novel Markov chain approach is proposed and its transition probabilities are shown to eventually converge to the near optimal solution without complete information. Furthermore, the gap between the optimal and converged solutions is shown to be bounded. Simulation results show that our proposed algorithm effectively offloads the traffic from macro-cell base station to small-cell base stations. Moreover, the results also show that this algorithm converges very quickly independent of the number of possible configurations.

Even though demand response of data centers has attracted many studies, there are very limited attempts on an important segment: colocation datacenters. Unlike large-scale (Google-type) datacenters, the colocation operator lacks control over its tenant servers, which entails a special interest in a design of incentive mechanisms, such that the operator can coordinate tenants to reduce the power usage for demand response. However, most previous studies ignore the role of the demand response provider (DRP), who uses pricing signals as a guide for customer response and as a compensation for their cutting electricity usage. To address this oversight, we propose an incentive mechanism Reward-to-Reduce for colocation's economic demand response, which shows an interaction between the DRP compensation to the colocation operator, and the colocation operator reward to tenants. Observing that this interaction contains strategic behaviors, we first formulate a two-stage Stackelberg game, where we show a unique competitive equilibrium of the operator strategy in the second stage, and a nonconvex problem of finding the optimal DRP compensation price in the first stage. We next analyze the second-stage equilibrium using an exact analysis and design an algorithm that can efficiently search the first-stage optimal DRP price with a reduced search space. Since the exact analysis can be impractical due to required tenants' private information, we also propose an approximate approach with limited tenant information. Extensive case studies show that the approximate approach can have the same performance as the exact analysis in a wide array of case studies and the optimal DRP price can be determined effectively, with which the corresponding DRP individual cost is compared with the social cost.

The installation of low power small cells under macro cells using the same spectrum is a promising approach to enhance the spectral efficiency and data-rate for the end users. These installations are becoming very dense in order to support the users' requirements (especially 5G networks) which make resource allocation using the same spectrum a very challenging problem. In this study, we address the downlink resource allocation problem for underlay small cell tier. We formulate the optimization problem for resource (channel) allocation in small cells while keeping the total interference to macro tier under an acceptable level. The objective of resource allocation is to maximize the throughput of small cells under the cross tier interference constraint. We employ matching theory to find a stable match for the resource allocation problem. We simulate our proposition to validate the stability of the network and the convergence of the resource allocation algorithm in terms of rate in a dense heterogeneous network. The matching results in an optimal solution which outperforms the existing sub-optimal resource allocation solutions.

In this paper, we study the relay-aided cooperative modeling that supports the load-sharing in uplink two-tier cellular networks. In our model, users in heavily loaded macrocell are shifted to lightly loaded smallcells with the assistance of relay users to mitigate Signal to Interference plus Noise Ratio (SINR) degradation problem in conventional direct handover. In order to promote relaying data of users which are selfish and rational, a trading exchange model based on Stackelberg game is proposed to optimize strategies of users. Relay users have pricing-based strategies on theirs power unit while shifted heavily loaded macrocell users have strategies to buy power levels of relay users. Optimal strategies are investigated using the backward induction analysis. Specifically, problems of NP-hard combinatorial optimization in relay user selections in the game are solved with a distributed algorithm based on matching theory. We intensively evaluate our proposed model by simulating it in Matlab which shows the efficiency of our proposal.

The increase in the number of connected smart mobile devices has fueled the exponential growth in mobile data. The next-generation networks must meet the demand for higher capacity. Heterogeneous cognitive networks with multiple base station tiers are a promising approach to achieving the higher data rate target. The user association problem is a major issue in the heterogeneous cognitive networks because of the disparity in transmission powers of the base stations involved. Our objective is to achieve the optimal user association under interference constraints. We formulate the problem into an optimization problem and employ matching theory to propose an algorithm to obtain the optimal user association. The proposed matching algorithm for the optimal user association plays the role of SDN application. We then perform simulations and compare our proposed algorithm with existing ones. The simulations results depict that our proposed algorithm outperforms others.

Heterogeneous cellular networks offload the mobile data traffic to small cell base stations to reduce the workload on the macro base stations. Our objective is to maximize the sum rate of the down-links for the whole network under outage QoS constraint. To achieve the objective, we have to jointly solve the user association problem and resource allocation problem. We formulate the two problems into a joint optimization problem and convert it into an equivalent game theoretic formulation. We employ payoff based log linear learning and propose an algorithm that converges to one of the existing Nash equilibrium. We then provide extensive simulation results to verify the performance of our proposed algorithm.

Content Centric Networking (CCN) is one of the most promising network architectures of future Internet. In CCN, Content Router (CR) floods request in the network to find the content. This flooding may degrade the network performance by generating too much traffic. Also, in legacy CCN architecture, each CR caches all contents that pass through it. Thus, same contents are replicated in all CRs along the request path, which incurs faster cache replacement and degrades cache utilization consequently. Furthermore, caching and forwarding decisions are made by the CR only on the basis of its local knowledge, which may not be optimum decisions. In this paper we propose a hybrid caching, cache replacement and requests forwarding approaches to overcome the above drawbacks. In our proposal, there is a virtual controller inside the data center with high performance computational capacity. The controller makes all the forwarding and caching decisions and passes it to the physical CRs via virtualized core router. Physical CRs are grouped inside regions on the basis of their geographical location, in order to enhance users' QoE. We have intensively simulated the proposed mechanism in a chunk level simulator and the performance is compared with existing schemes. The simulation results show that the proposed mechanism outperforms the existing state-of-the-art schemes.

Vehicular Ad hoc NETworks (VANETs) should provide the reliable safety message broadcasts and the efficient non-safety message transmissions to vehicles. The IEEE 1609.4 MAC, which supports multi-channel operations in VANETs, is not reliable enough for the safety message broadcast and not efficient in the Service CHannel (SCH) resources utilization. In this paper, we propose a MAC protocol which supports a Reliable Multi-hop Safety message Broadcast (RMSB-MAC) in VANETs. Each Multi-hop Forwarder (MF) collects the safety messages from the neighbor vehicle nodes, and then the MF uses its reserved time slot to broadcast them to all vehicle nodes in its transmission range as well as to forward them to the next MF. Moreover, by allowing vehicle nodes to exchange non-safety messages during the Control CHannel Interval (CCHI), the RMSB-MAC utilizes the SCH resources more efficiently.