Next generation Radio Access Networks (RANs) consider virtualized architectures in which base station functions are distributed in different logical nodes, connected through fronthaul (FH) links. To reduce the FH deployment costs and the required FH capacity, operators may install a single FH link shared among multiple cells and exploit key enabling techniques, such as modulation compression, to reduce FH data. In shared FH capacity scenarios, it is essential to provide efficient methods to control and optimize the FH resources’ utilization with limited impact on the air interface performance. In this paper, a multi-cell multi-user scenario with a shared FH link across multiple cells is considered. We focus on optimizing the resource allocation and modulation compression of each user, in a centralized and dynamic manner, aiming to maximize the air interface performance subject to a shared FH capacity constraint. The problem is formulated as a convex optimization problem, which allows deriving the optimal resource allocation and modulation compression per user. Then, we evaluate the proposed FH-aware scheduling methods against baseline holistic strategies over an end-to-end dynamic 5G NR system-level simulator based on ns-3. Under a tight available FH capacity, results show gains that vary from 16% to 567% in different percentile statistics of the user-perceived throughput.

Motivated by the heterogeneous nature of devices participating in large-scale federated learning (FL) optimization, we focus on an asynchronous server-less FL solution empowered by blockchain technology. In contrast to mostly adopted FL approaches, which assume synchronous operation, we advocate an asynchronous method whereby model aggregation is done as clients submit their local updates. The asynchronous setting fits well with the federated optimization idea in practical large-scale settings with heterogeneous clients. Thus, it potentially leads to higher efficiency in terms of communication overhead and idle periods. To evaluate the learning completion delay of BC-enabled FL, namely FLchain, we provide an analytical model based on batch service queue theory. Furthermore, we provide simulation results to assess the performance of both synchronous and asynchronous mechanisms. Important aspects involved in the BC-enabled FL optimization, such as the network size, link capacity, or user requirements, are put together and analyzed. As our results show, the synchronous setting leads to higher prediction accuracy than the asynchronous case. Nevertheless, asynchronous federated optimization provides much lower latency in many cases, thus becoming an appealing solution for FL when dealing with large datasets, tough timing constraints (e.g., near-real-time applications), or highly varying training data.

Next generation Radio Access Networks (RANs) consider virtualized architectures in which base station functions are distributed in different logical nodes, connected through fronthaul (FH) links. To reduce the FH deployment costs and the required FH capacity, operators may install a single FH link shared among multiple cells and exploit key enabling techniques, such as modulation compression, to decrease the data rates over the FH. In shared FH capacity scenarios, it is essential to provide efficient methods to control and optimize the FH resources' utilization with limited impact on the air interface performance. In this paper, we propose and analyze different fronthaul-aware scheduling strategies, leveraging on modulation compression, for multi-cell multi-user scenarios with a shared FH link across multiple cells. We consider holistic approaches based on packet dropping at the PHY layer and postponing of MAC scheduling decisions, combined with the reduction of the modulation order per cell. Additionally, we propose optimization methods for resource allocation and dynamic modulation compression, in which the modulation order and resource block assignment is dynamically optimized per user and slot. We finally evaluate the proposed FH-aware scheduling methods over an end-to-end dynamic 5G NR system-level simulator based on ns-3.

Blockchain (BC) technology can revolutionize the future of communications by enabling decentralized and open sharing networks. In this paper, we propose the application of BC to facilitate Mobile Network Operators (MNOs) and other players such as Verticals or Over-The-Top (OTT) service providers to exchange Radio Access Network (RAN) resources (e.g., infras-tructure, spectrum) in a secure, flexible and autonomous manner. In particular, we propose a BC-enabled reverse auction mecha-nism for RAN sharing and dynamic users' service provision in Beyond 5G networks, and we analyze its potential advantages with respect to current service provisioning and RAN sharing schemes. Moreover, we study the delay and overheads incurred by the BC in the whole process, when running over both wireless and wired interfaces.

Next generation Radio Access Networks (RANs) consider virtualized architectures in which base station functions are distributed in different logical nodes that are connected through fronthaul links. To reduce the required fronthaul capacity, modulation compression is considered as a key enabler. Modulation compression achieves fronthaul capacity reduction at the cost of reducing the maximum modulation order that can be used over the air interface, thus creating a cell-fronthaul trade-off. The trade-off is further accentuated and needs to be optimized appropriately when multiple cells share the same fronthaul link. In this paper, a multi-cell scenario with a shared fronthaul link across multiple cells is considered, and we focus on optimizing the modulation compression of each cell. We propose semi-static optimization procedures that aim at maximizing the air interface performance subject to a shared fronthaul capacity constraint, by taking into account the average traffic load and system configuration of every cell. The problem is formulated as a convex optimization problem, which allows deriving the optimal maximum modulation order that is permitted per cell, under two different optimization criteria. Then, we use a dynamic multi-cell 5G NR system-level simulator based on ns-3 to evaluate the proposed optimized solutions.

Blockchain (BC) technology can revolutionize future networks by providing a distributed, secure, and unalterable way to boost collaboration among operators, users, and other stakeholders. Its implementations have traditionally been supported by wired communications, with performance indicators like the high latency introduced by the BC being one of the key technology drawbacks. However, when applied to wireless communications, the performance of BC remains unknown, especially if running over contention-based networks. In this paper, we evaluate the latency performance of BC technology when the supporting communication platform is wireless, specifically we focus on IEEE 802.11ax, for the use case of users’ radio resource provisioning. For that purpose, we propose a discrete-time Markov model to capture the expected delay incurred by the BC. Unlike other models in the literature, we consider the effect that timers and forks have on the transaction confirmation latency.

The automatic classification of applications and services is an invaluable feature for new generation mobile networks. Here, we propose and validate algorithms to perform this task, atruntime, from theraw physical control channelof anoperative mobile network, without having to decode and/or decrypt the transmitted flows. Towards this, we decode Downlink Control Information (DCI) messages carried within the LTE Physical Downlink Control CHannel (PDCCH). DCI messages are sent by the radio cell in clear text and, in this article, are utilized to classify the applications and services executed at the connected mobile terminals. Two datasets are collected through a large measurement campaign: one labeled, used to train the classification algorithms, and one unlabeled, collected from four radio cells in the metropolitan area of Barcelona, in Spain. Among other approaches, our Convolutional Neural Network (CNN) classifier provides the highest classification accuracy of 98%. The CNN classifier is then augmented with the capability of rejecting sessions whose patterns do not conform to those learned during the training phase, and is subsequently utilized to attain a fine grained decomposition of the traffic for the four monitored radio cells, in anonlineandunsupervisedfashion.

A physical layer (PHY) abstraction model estimates the PHY performance in system-level simulators to speed up the simulations. This paper presents a PHY abstraction model for 5G New Radio (NR) and its integration into an open-source ns-3 based NR system-level simulator. The model capitalizes on the exponential effective signal-to-interference-plus-noise ratio (SINR) mapping (EESM) and considers the latest NR specification. To generate it, we used an NR-compliant link-level simulator to calibrate the EESM method as well as to obtain SINR-block error rate (BLER) lookup tables for various NR configurations. We also illustrate the usability of the developed model through end-to-end simulations in ns-3, under different NR settings of modulation and coding schemes, hybrid automatic repeat request combining methods, and link adaptation approaches.

In this paper, we discuss a handover management scheme for Next Generation Self-Organized Networks. We propose to extract experience from full protocol stack data, to make smart handover decisions in a multi-cell scenario, where users move and are challenged by deep zones of an outage. Traditional handover schemes have the drawback of taking into account only the signal strength from the serving, and the target cell, before the handover. However, we believe that the expected Quality of Experience (QoE) resulting from the decision of target cell to handover to, should be the driving principle of the handover decision. In particular, we propose two models based on multi-layer many-to-one LSTM architecture, and a multi-layer LSTM AutoEncoder (AE) in conjunction with a MultiLayer Perceptron (MLP) neural network. We show that using experience extracted from data, we can improve the number of users finalizing the download by 18 %, and we can reduce the time to download, with respect to a standard event-based handover benchmark scheme. Moreover, for the sake of generalization, we test the LSTM Autoencoder in a different scenario, where it maintains its performance improvements with a slight degradation, compared to the original scenario.

One of the main design targets of New Radio (NR) is to support multiple applications, including low-latency data transmissions. To achieve that, multiple features have been introduced, among which dynamic scheduling timings (denoted by K0, K1, K2) is the one which determines the delay between the different paired control and data transmissions. For example, K2 is the delay (in unit of slots) between an uplink grant reception and the corresponding uplink data transmission. The End-to-End (E2E) latency would be highly impacted by these scheduling timings. Based on the common observation, lower scheduling timings would lead to low E2E latency. However, in this paper, especially for uplink traffic, we show that this is not always true, and there are cases in which increasing the scheduling timings provides better E2E delay performance. This is due to the interplay between traffic patterns, gNB-UE process, and the NR numerology. To show such impact of scheduling timings on E2E latency with different uplink traffic patterns and NR numerologies, we implement an E2E ns-3 based NR simulator.

Detecting urban anomalies is of upmost importance for public order management, since they can pose serious risks to public safety if not timely handled. However, monitoring large metropolitan areas requires complex systems that can potentially lead to elevated costs. In this paper, we discuss the opportunity of exploiting the mobile network as a supplementary sensing platform for detecting urban anomalies. To favour the reliable and low latency anomaly recognition, we rely on a Multi-access Edge Computing (MEC) architecture, which enables a deep and detailed mobile traffic characterization almost in real-time and allows for a performance-responsive service, that is crucial in our problem. We focus on urban anomaly detection, by monitoring known events that gather a high concentration of people. The mobile network information is collected from LTE Physical Downlink Control Channel (PDCCH), which contains the radio scheduling information and has the benefit of being unencrypted and fine-grained, since the messages are exchanged every LTE subframe of 1 ms. To this purpose, we design an anomaly detection system based on Long Short-Term Memory (LSTM) Neural Networks, to deal with sequential and recurrent inputs. We demonstrate that a stacked LSTM architecture is able to identify traffic anomalies provoked by a rapid growth in the number of users, when a crowded event takes place nearby the monitored area. The numerical results show that the proposed algorithm reaches an F-score = 1 and overcomes the performance of other state-of -the-art benchmarks.

The 5G New Radio (NR) access technology defines multiple numerologies to support a wide range of carrier frequencies, deployment scenarios, and variety of use cases. In this paper, we consider a resource allocation problem to efficiently support multiple numerologies simultaneously. We assume frequency division multiplexing (FDM) of numerologies in a time division duplex (TDD) system with a self-contained slot format. We focus on optimizing the numerology subband (SB) configuration, as well as the duplexing ratio between downlink (DL) and uplink (UL) directions within each SB. The optimization problem minimizes the weighted sum of the normalized load (NL) for each SB in each direction. We prove that our optimization problem is convex and, furthermore, we derive the optimal closed-form expressions for the numerology SB configuration and the DL-UL duplexing ratio per SB. The effectiveness of the proposed resource allocation is validated through an end-to-end ns-3 based simulator, which shows how the optimization of the NLs is translated into an improved throughput and delay performance.

Predictive analysis on mobile network traffic is becoming of fundamental importance for the next generation cellular network. Proactively knowing the user demands, allows the system for an optimal resource allocation. In this paper, we study the mobile traffic of an LTE base station and we design a system for the traffic prediction using Recurrent Neural Networks. The mobile traffic information is gathered from the Physical Downlink Control CHannel (PDCCH) of the LTE using the passive tool presented in [1]. Using this tool we are able to collect all the control information at 1 ms resolution from the base station. This information comprises the resource blocks, the transport block size and the modulation scheme assigned to each user connected to the eNodeB. The design of the prediction system includes long short term memory units. With respect to a Multilayer Perceptron Network, or other artificial neurons structures, recurrent networks are advantageous for problems with sequential data (e.g. language modeling) [2]. In our case, we state the problem as a supervised multivariate prediction of the mobile traffic, where the objective is to minimize the prediction error given the information extracted from the PDCCH. We evaluate the one-step prediction and the long-term prediction errors of the proposed methodology, considering different numbers for the duration of the observed values, which determines the memory length of the LSTM network and how much information must be stored for a precise traffic prediction.

Listen-Before-Talk (LBT) has been adopted as the spectrum sharing technique that guarantees a fair LTE/Wi-Fi coexistence in the unlicensed spectrum at the 5 GHz band. Differently, at mmWave bands, where beamforming is a must to overcome propagation limits, LBT scope becomes limited because the interference layout changes due to the directionality of transmissions. In this regard, this paper proposes a Listen-Before-Receive (LBR) technique for shared spectrum access and analyzes its potentials to promote a fair coexistence of multiple Radio Access Technologies (RATs) in unlicensed mmWave bands, as, e.g., 5G New Radio (NR) access technology and Wireless Gigabit (WiGig) devices using IEEE 802.11ad/ay standard. Since the less likely but still harmful interference situations with directional transmissions can no longer be detected easily at the transmitter, we believe that the receiver has useful information to be used. The main idea of LBR is that we provide to the receiver a say when it comes to allowing/preventing the access to the channel. In this line, we propose potential implementations of LBR, in conjunction with LBT and the self-contained slot, for NR-based access to unlicensed mmWave bands.

License Assisted Access (LAA) is the technology introduced by the Third Generation Partnership Project (3GPP) that enables the deployment of LTE networks in the unlicensed 5 GHz spectrum. To ensure a fair coexistence of LAA in the unlicensed spectrum with other technologies, e.g., with Wi-Fi, 3GPP has standardized the use of Listen Before Talk (LBT) as the default channel-access scheme for LAA. However, the performance of Wi-Fi when coexisting with LAA mainly relies on how the LBT parameters are configured by the LAA. In this paper, we focus on the Contention Window (CW) size parameter of LBT in LAA. We propose a Neural Network (NN) based scheme that adapts the CW size based on the predicted number of Negative Acknowledgments (NACKs) for all the subframes in a Transmit Opportunity (TXOP) of LAA. In particular, our proposed scheme learns from the past experience how many NACKs per subframe of a TXOP were received under certain channel conditions. The performance evaluation shows that our proposed scheme, when compared to the state-of-the-art approaches, provides the best trade-off between the fairness to Wi-Fi and the LAA performance in terms of both throughput and latency.

The analysis of real mobile traffic traces is helpful to understand usage patterns of cellular networks. In particular, mobile data may be used for network optimization and management in terms of radio resources, network planning, energy saving, for instance. However, real network data from the operators is often difficult to be accessed, due to legal and privacy issues. In this paper, we overcome the lack of network information using a LTE sniffer capable of decoding the unencrypted LTE control channel and we present a temporal and spatial analysis of the recorded traces. Moreover, we present a methodology to derive a stochastic characterization for the daily variation of the LTE traffic. The proposed model is based on a discrete-time Markov chain and is compared with the real traces. Results show that, with a limited number of states, our model presents a high level of accuracy in terms of first and second order statistics.

As Mobile Network Operators (MNOs) are shifting towards Cloud- Radio Access Network (C-RAN), they have to upgrade their infrastructure to not only support higher processing capacities but also to be more resilient. We consider the problem where a MNO is faced with the choice of selecting virtualized Baseband Units (BBUs) from various cloud service providers, that are each characterized with distinct failure probabilities and prices. We propose to solve the BBU selection problem, formulated as an Integer Linear Program (ILP) subject to BBU capacity and virtualization cost using the Branch- and- Price algorithm. We present several schemes depicting which optimization goal the MNO can foster the most: BBU processing power minimization, resiliency, traffic handling or all. Simulation results demonstrate the good performance of our algorithm to solve the BBU selection problem for all schemes, while also emphasizing the advantages of a particular one that can realize more than 10% in virtualization cost savings.

Planning of current and future mobile networks is becoming increasingly complex due to the heterogeneity of deployments, which feature not only macrocells, but also an underlying layer of small cells whose deployment is not fully under the control of the operator. In this paper, we focus on selecting the most appropriate Quality of Service (QoS) prediction techniques for assisting network operators in planning future dense deployments. We propose to use machine learning as a tool to extract the relevant information from the huge amount of data generated in current 4G and future 5G networks during normal operation, which is then used to appropriately plan networks. In particular, we focus on radio measurements to develop correlative statistical models with the purpose of improving QoS-based network planning. In this direction, we combine multiple learners by building ensemble methods and use them to do regression in a reduced space rather than in the original one. We then compare the QoS prediction accuracy of various approaches that take as input the 3GPP Minimization of Drive Tests (MDT) measurements collected throughout a heterogeneous network and analyse their trade-offs. We also explain how the collected data is processed and used to predict QoS expressed in terms of Physical Resource Block (PRB)/ Megabit (MB) transmitted. This metric was selected because of the interest it may have for operators in planning, since it relates lower layer resources with their impact in terms of QoS up in the protocol stack, hence closer to the end-user.

This paper presents a machine learning based handover management scheme for LTE to improve the Quality of Experience (QoE) of the user in the presence of obstacles. We show that, in this scenario, a state-of-the-art handover algorithm is unable to select the appropriate target cell for handover, since it always selects the target cell with the strongest signal without taking into account the perceived QoE of the user after the handover. In contrast, our scheme learns from past experience how the QoE of the user is affected when the handover was done to a certain eNB. Our performance evaluation shows that the proposed scheme substantially improves the number of completed downloads and the average download time compared to state-of-the-art. Furthermore, its performance is close to an optimal approach in the coverage region affected by an obstacle.

In the coming years, planning future mobile networks will be infinitely more complex than nowadays. Future networks are expected to present multiple Network Management (NM) challenges to operators, such as managing network complexity in terms of densification of scenarios, heterogeneous nodes, applications, Radio Access Technologies (RAT), among others. In this context, the exploitation of past information gathered by the network is highly relevant when planning future deployments. In this paper we present a network planning tool based on Machine Learning (ML). In particular, we propose an approach which allows to predict Quality of Service (QoS) offered to end-users, based on data collected by the Minimization of Drive Tests (MDT) function. As a QoS indicator, we focus on Physical Resource Block (PRB) per Megabit (Mb) in an arbitrary point of the network. Minimizing this metric allows serving users with the same QoS by consuming less resources, and therefore, being more cost-effective. The proposed network planning tool considers a Genetic Algorithm (GA), which tries to reach the operator targets. The network parameters we desire to optimise are set as the input to the algorithm. Then, we predict the QoS of the network by means of ML techniques. By integrating these techniques in a network planning tool, operators would be able to find the most appropriate deployment layout, by minimizing the resources (i.e., the cost) they need to deploy to offer a given QoS in a newly planned deployment.

In a self-organized Long Term Evolution (LTE) network, different Self-Organizing Network (SON) functions or different instances of the same SON function can execute parallel actions, which interact or collide among each other. When the effect of the interaction negatively affects the performances of the system, this is referred to in 3rd Generation Partnership Project (3GPP) as a SON conflict, which needs to be handled by means of a self-coordination framework. We focus on the self-coordination of different actions taken by two SON functions in a distributed SON (D-SON) architecture, which implements the SON functions at the edges of the network. We propose a multi-agent framework where each Enhanced Node Base station (eNB), is an autonomous agent modeled by means of a Markov Decision Process (MDP). We subdivide this global Markov Decision problem onto simpler subMDPs modeling the different SON functions. Each sub-problem is defined as an MDP and solved independently, and their individual policies are combined to obtain a global policy. This combined policy can execute several actions per state in parallel, but can introduce policy conflicts, which model the mentioned SON conflicts. Each subMDP is solved by means of a Reinforcement Learning (RL) approach. We focus on the SON conflict generated by the concurrent execution of Coverage and Capacity Optimization (CCO) and Inter-Cell Interference Coordination (ICIC) SON functions, which may require to update the same parameter, i.e. the transmission power level. Coordination among different actions is achieved by means of a coordination game where the players are the subMDPs and the actions and rewards are those provided by the independent RL solutions. Performance evaluation is carried out in a ns3 release 10 compliant LTE system simulator and it shows that our self-coordination approach provides satisfying solutions in terms of system performances for both the conflicting SON functions.

This paper presents a novel framework for joint self-optimization of backhaul as well as coverage links spectral efficiency in relay enhanced heterogeneous networks. Considering a realistic heterogeneous network deployment, where some cells contain Relay Station (RS), while others do not, we develop an analytical framework for self-optimisation of macrocell Base Station (BS) antenna tilts. Our framework exploits a unique system level perspective to enable dynamic maximization of system-wide spectral efficiency of the BS-RS backhaul links as well as that of the BS-user coverage links. A distributed and practical self-organising solution is obtained by decomposing the large scale system-wide optimization problem into local small scale optimization problems, by mimicking the operational principles of self-organisation in biological systems. The local problems are non-convex but have very small scale and can be solved via appropriate numerical methods, such as sequential quadratic programming. The performance of developed solution is evaluated through extensive system level simulations for LTE-A type networks and compared against conventional tilting benchmarks. Numerical results show that up to 50% gain in average spectral efficiency is achievable through the proposed solution depending on users geographical distributions.

In this paper we present an automatic and self-organized Reinforcement Learning (RL) based approach for Cell Outage Compensation (COC). We propose that a COC module is implemented in a distributed manner in the Enhanced Node Base station (eNB)s in the scenario and intervenes when a fault is detected and so the associated outage. The eNBs surrounding the outage zone automatically and continually adjust their downlink transmission power levels and find the optimal antenna tilt value, in order to fill the coverage and capacity gap. With the objective of controlling the intercell interference generated at the borders of the extended cells, a modified Fractional Frequency Reuse (FFR) scheme is proposed for scheduling. Among the RL methods, we select a Temporal Difference (TD) learning approach, the Actor Critic (AC), for its capability of continuously interacting with the complex wireless cellular scenario and learning from experience. Results, validated on a Release 10 Long Term Evolution (LTE) system level simulator, demonstrate that our approach outperforms state of the art resource allocation schemes in terms of number of users recovered from outage. Index Terms-Self-Organizing Network (SON), Self Healing, Reinforcement Learning, LTE/LTE-Advanced, COC.

This paper elaborates on self-organized and distributed interference management for femtocells that share the available radio resources with macrocells. A multi-agent learning approach is examined, based on distributed Q-learning, where femtocell base stations control their transmit power, such that the femtocell capacity is maximized, while the aggregated downlink interference generated at macro users' receivers is maintained within acceptable limits. The distributed Q-learning algorithm is carried out at the femto nodes, in the way that the interference is controlled at each resource block. The contribution of this work is to integrate multi-user scheduling in the operation of the macrocell network, so that instantaneous changes, with 1 ms granularity, are encountered in the perception that the femtocell agents get of the environment under observation. We demonstrate that, by relying on 3GPP Long Term Evolution (LTE) compliant signalling from the macro network on the intended macrocell scheduling policies, the proposed learning approach allows each femto node to react on these instantaneous changes in the environment, such that the femto-to-macro interference is appropriately controlled.

This paper proposes a distributed solution for resource allocation in femtocell systems in order to control the downlink femto-to-macro aggregated interference. We propose a solution based on intelligent and self-organized femtocells implementing a decentralized Fuzzy Q-Learning (FQL), which with respect to other realtime multiagent Reinforcement Learning (RL) techniques allows to generalize the state space and to generate continuous actions, besides significantly speeding up the learning process. In particular, we propose a novel scheme able to maintain interference at the macrocell users below a threshold and at the same maximize the femtocells capacity, which was not considered by previous works of the same authors. We evaluate the FQL paradigm in the context of a 3rd Generation Partnership Project (3GPP) compliant Orthogonal Frequency Division Multiple Access (OFDMA) femtocell network.

This paper deals with the problem of aggregated interference generated by multiple cognitive radios (CR) at the receivers of primary (licensed) users. In particular, we consider a secondary CR system based on the IEEE 802.22 standard for wireless regional area networks (WRAN), and we model it as a multi-agent system where the multiple agents are the different secondary base stations in charge of controlling the secondary cells. We propose a form of real-time multi-agent reinforcement learning, known as decentralized Q-learning, to manage the aggregated interference generated by multiple WRAN cells. We consider both situations of complete and partial information about the environment. By directly interacting with the surrounding environment in a distributed fashion, the multi-agent system is able to learn, in the first case, an optimal policy to solve the problem and, in the second case, a reasonably good suboptimal policy. Simulation results reveal that the proposed approach is able to fulfill the primary users interference constraints, without introducing signaling overhead in the system.

We address the coexistence problem between macrocell and femtocell systems by controlling the aggregated interference generated by multiple femtocell base stations at the macrocell receivers in a distributed fashion. We propose a solution based on intelligent and self-organized femtocells implementing a realtime multi-agent reinforcement learning technique, known as decentralized Q- learning. We compare this cognitive approach to a non-cognitive algorithm and to the well known iterative water- filling, showing the general superiority of our scheme in terms of (non-jeopardized) macrocell capacity. Furthermore, in distributed settings of such femtocell networks, the learning may be complex and slow due to mutually impacting decision making processes, which results in a non-stationary environment. We propose a timely solution -referred to as docition- to improve the learning process based on the concept of teaching and expert knowledge sharing in wireless environments. We demonstrate that such an approach improves the femtocells' learning ability and accuracy. We evaluate the docitive paradigm in the context of a 3GPP compliant OFDMA (Orthogonal Frequency Division Multiple Access) femtocell network modeled as a multi-agent system. We propose different docitive algorithms and we show their superiority to the well known paradigm of independent learning in terms of speed of convergence and precision.

This paper proposes a decentralized interference control scheme for an OFDMA (Orthogonal Frequency Division Multiple Access) cellular scenario where multiple macrocells and femtocells are deployed. In particular, we model the distributed resource allocation problem by means of a potential game, which is demonstrated to converge to a unique and pure Nash equilibrum. In this game, the femto base stations are the players, their actions are the resource blocks and the corresponding power levels to be allocated for downlink transmission, and the utility is designed to guarantee coexistence with the macro system, guaranteeing a fair tradeoff between femto and macro systems' performances. To do that, the utility function does not only consider the capacity of the femtocells, but also the different sources of inter-system interference: macrocell to femtocell, femtocell to femtocell, and femtocell to macrocell. The game is solved by applying a better response dynamic, which first selects the best resource blocks allocation policy for the femtousers, and then selects the most appropriate power levels for each resource block, through the solution of the corresponding convex optimization problem. Simulation results show that the introduction of femtocells in the scenario increases the total system capacity, and that the considered utility function provides an improvement in macro system performance of up to 40%, while reducing the femto capacity only up to 7%, with respect to the well known iterative waterfilling game.

This paper proposes a self-organized power allocation technique to solve the interference problem caused by a femtocell network operating in the same channel as an orthogonal frequency division multiple access cellular network. We model the femto network as a multi-agent system where the different femto base stations are the agents in charge of managing the radio resources to be allocated to their femtousers. We propose a form of real-time multi-agent reinforcement learning, known as decentralized Q-learning, to manage the interference generated to macro-users. By directly interacting with the surrounding environment in a distributed fashion, the multi-agent system is able to learn an optimal policy to solve the interference problem. Simulation results show that the introduction of the femto network increases the system capacity without decreasing the capacity of the macro network.

In this paper we propose network coding to perform multi-hop multicast wireless transmissions. More specifically, we address the problem of reducing redundancy in the completely distributed operation of network coding, so as to increase network throughput. In order to do so, we propose a distributed system where network nodes autonomously make decisions with respect to the packets to encode and forward, with the goal of maximizing the data detection probability and minimizing the overhead of transmitted packets. Our setup consists of a Slotted Aloha network where nodes inject packets at a rate which depends on how useful these packets are to other nodes. We model the selection of transmission probabilities by means of a game theoretic approach, in which nodes reach the desired transmission probabilities as an equilibrium solution to the game. Simulation results show that the proposed scheme outperforms a Slotted Aloha system with optimal uniform retransmission probability.

This paper presents an approach to detect misbehaving nodes in cognitive and cooperative networks. In particular, we propose a cooperative scheme, where cognitive radios are granted access in licensed bands, as long as they compensate primary (licensed) users for the additional interference generated in their bands through cooperation. However, this cooperative transmission can be vulnerable to malicious attacks, which are not easy to detect, since they may affect the error probability experienced by the primary users, but not their signal to noise and interference ratio. We propose a misbehavior detection approach based on measuring and comparing the entropy of data derived by anomalous and normal behaviors. Simulation results show that the probability of detection strongly depends on the transmission power and spectrum handoff rate of the outliers.

This paper presents a Bayesian potential game to model distributed joint power and channel allocation for cognitive radios with incomplete information. We propose a cooperative approach where secondary users (SUs) devote part of their transmission power in licensed channels to relaying primary users' (PUs) messages. In addition, we consider incomplete information in the decision making process, so as to avoid the need for a common control channel (CCC), where users share information. This hypothesis improves the robustness and feasibility of the cognitive radio network supported by the proposed approach. Simulation results show that cooperation benefits both PUs and SUs and that the lack of complete information in the decision process slightly reduces performances in terms of signal to interference and noise ratio (SINR) and outage probability.

This paper studies the benefits that cooperation brings to a cognitive radio network. The proposal in this paper considers that secondary unlicensed users are allowed to opportunistically use the radio spectrum allocated to the primary licensed users, as long as they agree on facilitating the primary user communications by cooperating with them. We refer to this approach as overlay paradigm for cognitive radio and we compare this to the underlay paradigm, according to which cooperation techniques among primary and secondary users are not exploited. To model these schemes we make use of theory of exact potential games. We analyze the convergence properties of the proposed games and we evaluate the outputs in terms of quality of service perceived by both primary and secondary users, outage probability and interference temperature, showing that the overlay paradigm for cognitive radio is a promising framework.

Joint radio resource management (JRRM) is the envisaged process aimed at optimizing the radio resource usage of wireless systems to satisfy the requirements of both the network operators and the users in the context of future generation wireless networks. In particular, this paper proposes a two-layered JRRM framework to improve the efficiency of multiradio and multioperator cellular scenarios. On the one hand, the intraoperator JRRM relies on fuzzy neural mechanisms with economic-driven reinforcement learning techniques to exploit radio resources within a single-operator domain. Microeconomic concepts are included in the proposed approach so that user profile differentiation can be considered when making a JRRM decision. On the other hand, interoperator JRRM enables subscribers to obtain service through other operators, if the home operator network is blocked. Simulation results in a number of different scenarios show that interoperator agreements established in a cooperative scenario benefit both the operators and users, which enables efficient load management and increased operator revenue.

In this paper we present a framework to better exploit spectrum resources of a TDMA/FDMA based primary system through a secondary usage of spectrum. The secondary users of the spectrum are the nodes of an overlay cognitive ad-hoc network, which opportunistically transmits in data channels left unused by the primary system. The focus of the paper is on spectrum decision to properly select the best available data channel where to opportunistically transmit without interfering to the primary network. Spectrum decisions schemes are proposed and compared in terms of improved spectrum usage and interference caused on the primary system.

This paper proposes a two-layer joint radio resource management (JRRM) framework to improve the efficiency in multi-radio and multi-operator cellular scenarios. On the one hand, the intra-operator JRRM relies on fuzzy-neural mechanisms with economic-driven reinforcement learning techniques to exploit radio resources within a single operator domain. On the other hand, inter-operator JRRM allows subscribers to get service through other operators in case the home operator network is blocked. Simulation results in a number of different scenarios show that inter-operator agreements established in a cooperative scenario bring benefits to both operators and users, enabling an efficient load management and increasing the operators' revenue.

In this paper it will be shown how the fuzzy neural methodology can be used to develop an innovative mechanism to perform joint radio resource management (JRRM) in the context of heterogeneous radio access networks (RANs). In particular, an algorithm able to ensure certain quality of service (QoS) constraints in a multi-cell scenario deployment with three different radio access technologies (RATs), namely WLAN (wireless local area network), UMTS (universal mobile telecommunications system) and GERAN (GSM EDGE radio access network), is discussed. In addition, particular attention is paid to the design of an approach capable of managing handoff calls in a heterogeneous network. It is based on predicting future JRRM decisions. A RLS (recursive least square) predictor has been selected to provide reliable and accurate estimations of handoff calls and handoff call droppings. Performance improvements in terms of new connection blocking and handoff call dropping probabilities are presented.

This paper proposes a new joint radio resource management (JRRM) strategy which, as a difference from previous works in the literature, introduces economic and user differentiation concepts. This economic-driven algorithm is based on guaranteeing a certain user acceptance of a given service, combining both the economic and technical dimensions, while at the same time increasing the operator revenue. Two user profiles with different needs are considered, namely consumer and business users, and it is shown that the needs of both profiles can be met by means of the proposed framework. Three implementations of the proposed JRRM are compared. Two of them are based on a user-centric approach, whereas the third implementation falls into the joint user-centric and network-centric category aiming at maximizing the operator revenue while maintaining the user satisfaction at a target contracted value. In this case, particular attention is paid to business users and to strategies based on resource reservation to improve their utility as well as the operator revenue.

The vision of the wireless communications beyond 3G is characterized by flexibility. Future communication networks should be able to flexibly allocate resources and radio access technologies (RATs) to maintain high quality of communication and efficient use of radio resources. This should be the task of the joint radio resource management (JRRM) algorithms. Furthermore, this flexibility should also cover the pricing mechanism which should be able to react to the instantaneous users' needs and resource availability. This will be tackled by a real time spectrum auction system. This paper draws an architecture which embodies the two mechanisms: JRRM and spectrum auction, in order to create highly efficient wireless systems.

This paper presents a joint radio resource management algorithm to operate in a heterogeneous network scenario including cellular and wireless local area network radio access technologies. It makes use of a methodology based on fuzzy-neural systems in order to carry out a coordinated management of the radio resources among the different access networks. In order to reveal the potentials of the proposed algorithm, it is compared with other strategies in a multicellular and multi-RAT scenario.

This paper presents a novel JRRM strategy based on reinforcement learning mechanisms that control a fuzzy-neural algorithm to ensure certain QoS constraints. Three RATs (radio access technologies), namely UMTS, GERAN and WLAN are considered as common available technologies to select. The fuzzy logic allows for a very simple handling of the joint radio resource manager simply by activating a set of rules. The membership functions considered by these rules are adaptive so that a desired performance in terms of the probability of user satisfaction can be guaranteed by means of the reinforcement learning algorithm. Some illustrative simulation results to evaluate the behaviour of the proposed JRRM technique are presented.

This paper presents a comprehensive framework to develop joint RRM (radio resource management) strategies taking full advantage of the reconfigurable equipment capabilities and the diversity offered by available RATs (radio access technologies) in a multi-radio environment. The envisaged JRRM treatment calls for establishing links with all the entities involved, first at functional level, identifying realistic scenarios in terms of deployment, technologies and services, and managing the emerging complexity with proper algorithms. Then, a fuzzy-neural methodology framework able to cope with the complexities and uncertainties these new scenarios rise is presented. In particular both technical and economical aspects are considered when selecting a particular RAT. Finally some significant examples of the algorithm behaviour are shown.