To efficiently serve heterogeneous demands in terms of data rate, reliability, latency and mobili... more To efficiently serve heterogeneous demands in terms of data rate, reliability, latency and mobility, network operators must optimize the utilization of their infrastructure resources. In this context, we propose a framework to orchestrate resources for 5G networks by leveraging Machine Learning (ML) techniques. We start by classifying the demands for resources into groups in order to adequately serve them by dedicated logical virtual networks or Network Slices (NSs). To optimally implement these heterogeneous NSs that share the same infrastructure, we develop a new dynamic slicing approach of Physical Resource Blocks (PRBs). On first hand, we propose a predictive approach to achieve optimal slicing decisions of the PRBs from a limited resource pool. On second hand, we design an admission controller and a slice scheduler and formalize them as Knapsack problems. Finally, we design an adaptive resource manager by leveraging Deep Reinforcement Learning (DRL). Using our 5G experimental p...
In this demonstration, we present an autonomous anomaly detector for Cloud Radio Access Network (... more In this demonstration, we present an autonomous anomaly detector for Cloud Radio Access Network (C-RAN) performance metrics through a prototype using OpenAirInterface implemented as Docker containers. First, we show network slicing configuration capabilities including slice lifecycle management in addition to devices re-homing through our developed northbound interface allowing to manage a software-defined radio access network controller. Next, using conventional off-the-shelf smartphones, we run experiments to present the real-time data display in addition to the instant detection of anomalies in time-series CRAN data. We demonstrate a reaction to such detection through the auto-trigger of planned responses.
The enhanced Interference Mitigation and Traffic Adaptation (eIMTA) mechanism is a key enabler fo... more The enhanced Interference Mitigation and Traffic Adaptation (eIMTA) mechanism is a key enabler for 5G networks and beyond. Knowing that a User Equipment (UE) cannot transmit more than the maximum power allowed by its power class. The level of available Transmission Power (TP) in each UE is an essential input for the Uplink (UL) scheduler of the next Generation Node-B (gNB). Scheduling higher data rate than what is supported by the available TP is a waste of resources. In the Downlink (DL), the power level is known by the gNB that manages the power amplifier and the DL-scheduler. Conversely, in the UL, the available power is estimated by the UE and sent to the gNB as a key input for eIMTA, known as Power Head-Room (PHR). In this context, we propose in this paper, a Deep Neural Network (DNN) based model to predict the PHR and reduce dependency on reported measures. We evaluate the effectiveness of our proposal in a 5G experimental prototype, based on Open Air Interface (OAI). Obtained...
The increased power demand and the renewable energy integration problems have led to the evolutio... more The increased power demand and the renewable energy integration problems have led to the evolution of the traditional electric power grid toward smart grid. In order to permit the interaction among computational and physical elements, the smart grid supports bidirectional information flows between the energy user and the utility grid by integrating Information and Communication Technologies (ICTs). Thus, bidirectional flows between smart grid entities allow energy users not only to consume energy, but also to generate energy and to share it with the utility grid or with other energy consumers. Some researchers have paid attention to the energy management in the smart grids in order to provide an efficient way to maximize the savings of energy bills. However, these researches present some common drawbacks such as: the lack of integration of storage system and the high energy losses. Therefore, this chapter discusses a novel agent-based approach for energy management and control by balancing electric power supply, and minimizing energy bill, while considering residential consumers preferences and comfort level. Simulation results show that our proposal minimizes the energy costs for each energy demand and reduces conventional energy utilization.
Abstract. The combination between DiffServ (Differentiated Services) and Multi-Protocol Label Swi... more Abstract. The combination between DiffServ (Differentiated Services) and Multi-Protocol Label Switching (MPLS) presents a very attractive strategy to backbone network service providers. It provides scalable QoS and traffic engineering capabilities. However, the management of such a network is not a simple function and could not be done manually. In fact, it would be much more economic and effective to automatically manage networks. In this paper, we discuss the essential characteristics needed to build an autonomic network. We also propose a novel architecture based on Multi-Agent Systems (MAS) in order to automatically manage an MPLS-DiffServ TE domain. Simulation results are provided to illustrate the efficiency of our proposition.
Abstract. Gathering information in an energy-efficient and scalable manner from a wireless sensor... more Abstract. Gathering information in an energy-efficient and scalable manner from a wireless sensor network is always a basic need. In this work, we use the multi-agent approach in order to build an InformationImportance Based Communication for large scale wireless sensor network data processing. The principal goal of our proposition is to tackle the problem of network density and scalability in an energy efficient manner. Simulation results are provided to illustrate the efficiency of our proposition.
Abstract. The Multi-Protocol Label Switching (MPLS) is an Internet Engineering Task Force (IETF) ... more Abstract. The Multi-Protocol Label Switching (MPLS) is an Internet Engineering Task Force (IETF) framework. It is a versatile solution to address the problems faced by present-day networks like speed, scalability and traffic engineering. However, the Quality of Service (QoS) management of MPLS is made by static methods. In this paper, we propose a solution based on MultiAgent Systems (MAS) to manage the QoS into MPLS by adequate adaptive methods.
ICC 2020 - 2020 IEEE International Conference on Communications (ICC), 2020
Cloud computing is being embraced more and more by telecommunication operators for on-demand acce... more Cloud computing is being embraced more and more by telecommunication operators for on-demand access to computing resources. Knowing that 5G Core reference architecture is envisioned to be cloud-native and service-oriented, we propose, in this paper, offloading to the cloud, some of 5G delay-tolerant Network Functions and in particular the Network Data Analytics Function (NWDAF). The dynamic selection of cloud resources to serve off-loaded 5G-NWDAF, while incurring minimum cost and maximizing utilization of served next generation Node-Bs (gNBs) requires agility and automation. This paper introduces a framework to automate the selection process that satisfies resource demands while meeting two objectives, namely, cost minimization and utilization maximization. We first formulate the mapping of gNBs to 5G-NWDAF problem as an Integer Linear Program (ILP). Then, we propose an algorithm to solve it based on branch-cut-and-price technique combining all of branch-and-price, branch-and-cut a...
2019 IEEE Global Communications Conference (GLOBECOM), 2019
5G will serve heterogeneous demands in terms of data-rate, reliability, latency, and efficiency. ... more 5G will serve heterogeneous demands in terms of data-rate, reliability, latency, and efficiency. Mobile operators shall be able to serve all of these requirements using shared network infrastructure's resources. To this end, we propose in this paper a framework for resource orchestration for 5G network slices implementing four Quality of Service pillars. Starting from traffic classification, demands are marked so that they are best served by dedicated logical virtual networks called Network Slices (NSs). To optimally serve multiple NSs over the same physical network, we then implement a new dynamic slicing approach of network resources exploiting Machine Learning (ML). Indeed, as demands change dynamically, a mere recursive optimization leading to progressive convergence towards an optimum slice is not sufficient. Consequently, we need an initial well-informed slicing decision of physical resources from a total available resource pool. Moreover, we formalize both admission contr...
2018 Global Information Infrastructure and Networking Symposium (GIIS), 2018
Cloud-Radio Access Network (C-RAN) is an attractive solution to Mobile Network Operators. Firstly... more Cloud-Radio Access Network (C-RAN) is an attractive solution to Mobile Network Operators. Firstly, C-RAN leverages the effect of pooling multiple Baseband Units (BBUs) to offer centralized processing resources while hosting them on cloud. This results in multiple benefits ranging from statistical multiplexing gains, to energy efficiency. Secondly, C-RAN allows deploying Remote Radio Heads (RRHs) in proximity of end-users allowing exploiting Inter-Cell Interference Cancellation (ICIC) to maximize throughput by coordinating multiple RRHs. In this context, we propose, in this paper, a new throughput-aware RRHs clustering method for C-RAN that maximizes the throughput for end-users, while meeting multiple constrained resources on BBUs. Our approach consists of two stages: First, individual throughput value and requirements of each RRH are calculated taking into account the Signal-to-Interference-plus-Noise Ratio (SINR) values and the distance between RRHs and users. Then, they are inclu...
La taille des reseaux est de plus en plus importante et leur configuration et leur pilotage sont ... more La taille des reseaux est de plus en plus importante et leur configuration et leur pilotage sont devenus complexes. De ce fait, la communaute reseau devient consciente de la necessite de permettre aux reseaux de se configurer et de se piloter de maniere autonome. Cette necessite nous a conduit a proposer une architecture adaptative basee sur les Systemes Multi- Agents (SMA) dans le but d'assurer une gestion autonome et decentralisee du reseau MPLS-DiffServ TE. MPLS-DiffServ TE permet de fournir une differenciation de service tout en optimisant l'utilisation des ressources du reseau mais dont la gestion est devenue tres complexe. Nous avons alors defini les caracteristiques que nous pensons necessaires pour creer un reseau autonome et avons montre qu'elles sont toutes fournies par les SMA. Nous avons egalement propose une strategie de gestion de LSP, basee sur une approche fondee sur le trafic et dependant des conditions du reseau. Le but de cette strategie est de diminue...
The electric grid is radically evolving into the smart grid, which is characterized by improved e... more The electric grid is radically evolving into the smart grid, which is characterized by improved energy efficiency of available resources. The smart grid permits interactions among its computational and physical elements thanks to the integration of Information and Communication Technologies (ICTs). ICTs provide energy management algorithms and allow renewable energy integration and energy price minimization. Given the importance of renewable energy, many researchers developed energy management (EM) algorithms to minimize renewable energy intermittency. EM plays an important role in the control of users' energy consumption and enables increased consumer participation in the market. These algorithms provide consumers with information about their energy consumption patterns and help them adopt energy-efficient behaviour. In this paper, we present a review of the state of the energy management algorithms. We define a set of requirements for EM algorithms and evaluate them qualitativ...
In this demonstration, we present an autonomous anomaly detector for Cloud Radio Access Network (... more In this demonstration, we present an autonomous anomaly detector for Cloud Radio Access Network (C-RAN) performance metrics through a prototype using OpenAirInterface implemented as Docker containers. First, we show network slicing configuration capabilities including slice lifecycle management in addition to devices re-homing through our developed northbound interface allowing to manage a software-defined radio access network controller. Next, using conventional off-the-shelf smartphones, we run experiments to present the real-time data display in addition to the instant detection of anomalies in time-series CRAN data. We demonstrate a reaction to such detection through the auto-trigger of planned responses.
The main goal of wireless sensor networks (WSN) is to gather information from the regions of inte... more The main goal of wireless sensor networks (WSN) is to gather information from the regions of interest through a large number of micro sensor nodes. This gathering is traditionally done by using a client/server communication approach. However, this communication architecture consumes a lot of power and does not take into consideration the information properties. In this paper, we propose a data gathering scheme for WSNs, based on agents cooperation to deal with the importance of the information. This agent cooperation aims to reduce an important amount of the information communicated over the network by eliminating the unimportant information and the inter-sensor-nodes redundancy. This cooperation is empowered by an agent strategy taking into account several parameters related to the node and to the instance of communication for an optimized power management. Successive simulations proved, in large scale WSNs and different densities, the ability of the proposed gathering scheme to re...
The enhanced Interference Mitigation and Traffic Adaptation (eIMTA) mechanism is a key enabler fo... more The enhanced Interference Mitigation and Traffic Adaptation (eIMTA) mechanism is a key enabler for 5G networks and beyond. Knowing that a User Equipment (UE) cannot transmit more than the maximum power allowed by its power class. The level of available Transmission Power (TP) in each UE is an essential input for the Uplink (UL) scheduler of the next Generation Node-B (gNB). Scheduling higher data rate than what is supported by the available TP is a waste of resources. In the Downlink (DL), the power level is known by the gNB that manages the power amplifier and the DL-scheduler. Conversely, in the UL, the available power is estimated by the UE and sent to the gNB as a key input for eIMTA, known as Power Head-Room (PHR). In this context, we propose in this paper, a Deep Neural Network (DNN) based model to predict the PHR and reduce dependency on reported measures. We evaluate the effectiveness of our proposal in a 5G experimental prototype, based on Open Air Interface (OAI). Obtained...
To efficiently serve heterogeneous demands in terms of data rate, reliability, latency and mobili... more To efficiently serve heterogeneous demands in terms of data rate, reliability, latency and mobility, network operators must optimize the utilization of their infrastructure resources. In this context, we propose a framework to orchestrate resources for 5G networks by leveraging Machine Learning (ML) techniques. We start by classifying the demands for resources into groups in order to adequately serve them by dedicated logical virtual networks or Network Slices (NSs). To optimally implement these heterogeneous NSs that share the same infrastructure, we develop a new dynamic slicing approach of Physical Resource Blocks (PRBs). On first hand, we propose a predictive approach to achieve optimal slicing decisions of the PRBs from a limited resource pool. On second hand, we design an admission controller and a slice scheduler and formalize them as Knapsack problems. Finally, we design an adaptive resource manager by leveraging Deep Reinforcement Learning (DRL). Using our 5G experimental p...
In this demonstration, we present an autonomous anomaly detector for Cloud Radio Access Network (... more In this demonstration, we present an autonomous anomaly detector for Cloud Radio Access Network (C-RAN) performance metrics through a prototype using OpenAirInterface implemented as Docker containers. First, we show network slicing configuration capabilities including slice lifecycle management in addition to devices re-homing through our developed northbound interface allowing to manage a software-defined radio access network controller. Next, using conventional off-the-shelf smartphones, we run experiments to present the real-time data display in addition to the instant detection of anomalies in time-series CRAN data. We demonstrate a reaction to such detection through the auto-trigger of planned responses.
The enhanced Interference Mitigation and Traffic Adaptation (eIMTA) mechanism is a key enabler fo... more The enhanced Interference Mitigation and Traffic Adaptation (eIMTA) mechanism is a key enabler for 5G networks and beyond. Knowing that a User Equipment (UE) cannot transmit more than the maximum power allowed by its power class. The level of available Transmission Power (TP) in each UE is an essential input for the Uplink (UL) scheduler of the next Generation Node-B (gNB). Scheduling higher data rate than what is supported by the available TP is a waste of resources. In the Downlink (DL), the power level is known by the gNB that manages the power amplifier and the DL-scheduler. Conversely, in the UL, the available power is estimated by the UE and sent to the gNB as a key input for eIMTA, known as Power Head-Room (PHR). In this context, we propose in this paper, a Deep Neural Network (DNN) based model to predict the PHR and reduce dependency on reported measures. We evaluate the effectiveness of our proposal in a 5G experimental prototype, based on Open Air Interface (OAI). Obtained...
The increased power demand and the renewable energy integration problems have led to the evolutio... more The increased power demand and the renewable energy integration problems have led to the evolution of the traditional electric power grid toward smart grid. In order to permit the interaction among computational and physical elements, the smart grid supports bidirectional information flows between the energy user and the utility grid by integrating Information and Communication Technologies (ICTs). Thus, bidirectional flows between smart grid entities allow energy users not only to consume energy, but also to generate energy and to share it with the utility grid or with other energy consumers. Some researchers have paid attention to the energy management in the smart grids in order to provide an efficient way to maximize the savings of energy bills. However, these researches present some common drawbacks such as: the lack of integration of storage system and the high energy losses. Therefore, this chapter discusses a novel agent-based approach for energy management and control by balancing electric power supply, and minimizing energy bill, while considering residential consumers preferences and comfort level. Simulation results show that our proposal minimizes the energy costs for each energy demand and reduces conventional energy utilization.
Abstract. The combination between DiffServ (Differentiated Services) and Multi-Protocol Label Swi... more Abstract. The combination between DiffServ (Differentiated Services) and Multi-Protocol Label Switching (MPLS) presents a very attractive strategy to backbone network service providers. It provides scalable QoS and traffic engineering capabilities. However, the management of such a network is not a simple function and could not be done manually. In fact, it would be much more economic and effective to automatically manage networks. In this paper, we discuss the essential characteristics needed to build an autonomic network. We also propose a novel architecture based on Multi-Agent Systems (MAS) in order to automatically manage an MPLS-DiffServ TE domain. Simulation results are provided to illustrate the efficiency of our proposition.
Abstract. Gathering information in an energy-efficient and scalable manner from a wireless sensor... more Abstract. Gathering information in an energy-efficient and scalable manner from a wireless sensor network is always a basic need. In this work, we use the multi-agent approach in order to build an InformationImportance Based Communication for large scale wireless sensor network data processing. The principal goal of our proposition is to tackle the problem of network density and scalability in an energy efficient manner. Simulation results are provided to illustrate the efficiency of our proposition.
Abstract. The Multi-Protocol Label Switching (MPLS) is an Internet Engineering Task Force (IETF) ... more Abstract. The Multi-Protocol Label Switching (MPLS) is an Internet Engineering Task Force (IETF) framework. It is a versatile solution to address the problems faced by present-day networks like speed, scalability and traffic engineering. However, the Quality of Service (QoS) management of MPLS is made by static methods. In this paper, we propose a solution based on MultiAgent Systems (MAS) to manage the QoS into MPLS by adequate adaptive methods.
ICC 2020 - 2020 IEEE International Conference on Communications (ICC), 2020
Cloud computing is being embraced more and more by telecommunication operators for on-demand acce... more Cloud computing is being embraced more and more by telecommunication operators for on-demand access to computing resources. Knowing that 5G Core reference architecture is envisioned to be cloud-native and service-oriented, we propose, in this paper, offloading to the cloud, some of 5G delay-tolerant Network Functions and in particular the Network Data Analytics Function (NWDAF). The dynamic selection of cloud resources to serve off-loaded 5G-NWDAF, while incurring minimum cost and maximizing utilization of served next generation Node-Bs (gNBs) requires agility and automation. This paper introduces a framework to automate the selection process that satisfies resource demands while meeting two objectives, namely, cost minimization and utilization maximization. We first formulate the mapping of gNBs to 5G-NWDAF problem as an Integer Linear Program (ILP). Then, we propose an algorithm to solve it based on branch-cut-and-price technique combining all of branch-and-price, branch-and-cut a...
2019 IEEE Global Communications Conference (GLOBECOM), 2019
5G will serve heterogeneous demands in terms of data-rate, reliability, latency, and efficiency. ... more 5G will serve heterogeneous demands in terms of data-rate, reliability, latency, and efficiency. Mobile operators shall be able to serve all of these requirements using shared network infrastructure's resources. To this end, we propose in this paper a framework for resource orchestration for 5G network slices implementing four Quality of Service pillars. Starting from traffic classification, demands are marked so that they are best served by dedicated logical virtual networks called Network Slices (NSs). To optimally serve multiple NSs over the same physical network, we then implement a new dynamic slicing approach of network resources exploiting Machine Learning (ML). Indeed, as demands change dynamically, a mere recursive optimization leading to progressive convergence towards an optimum slice is not sufficient. Consequently, we need an initial well-informed slicing decision of physical resources from a total available resource pool. Moreover, we formalize both admission contr...
2018 Global Information Infrastructure and Networking Symposium (GIIS), 2018
Cloud-Radio Access Network (C-RAN) is an attractive solution to Mobile Network Operators. Firstly... more Cloud-Radio Access Network (C-RAN) is an attractive solution to Mobile Network Operators. Firstly, C-RAN leverages the effect of pooling multiple Baseband Units (BBUs) to offer centralized processing resources while hosting them on cloud. This results in multiple benefits ranging from statistical multiplexing gains, to energy efficiency. Secondly, C-RAN allows deploying Remote Radio Heads (RRHs) in proximity of end-users allowing exploiting Inter-Cell Interference Cancellation (ICIC) to maximize throughput by coordinating multiple RRHs. In this context, we propose, in this paper, a new throughput-aware RRHs clustering method for C-RAN that maximizes the throughput for end-users, while meeting multiple constrained resources on BBUs. Our approach consists of two stages: First, individual throughput value and requirements of each RRH are calculated taking into account the Signal-to-Interference-plus-Noise Ratio (SINR) values and the distance between RRHs and users. Then, they are inclu...
La taille des reseaux est de plus en plus importante et leur configuration et leur pilotage sont ... more La taille des reseaux est de plus en plus importante et leur configuration et leur pilotage sont devenus complexes. De ce fait, la communaute reseau devient consciente de la necessite de permettre aux reseaux de se configurer et de se piloter de maniere autonome. Cette necessite nous a conduit a proposer une architecture adaptative basee sur les Systemes Multi- Agents (SMA) dans le but d'assurer une gestion autonome et decentralisee du reseau MPLS-DiffServ TE. MPLS-DiffServ TE permet de fournir une differenciation de service tout en optimisant l'utilisation des ressources du reseau mais dont la gestion est devenue tres complexe. Nous avons alors defini les caracteristiques que nous pensons necessaires pour creer un reseau autonome et avons montre qu'elles sont toutes fournies par les SMA. Nous avons egalement propose une strategie de gestion de LSP, basee sur une approche fondee sur le trafic et dependant des conditions du reseau. Le but de cette strategie est de diminue...
The electric grid is radically evolving into the smart grid, which is characterized by improved e... more The electric grid is radically evolving into the smart grid, which is characterized by improved energy efficiency of available resources. The smart grid permits interactions among its computational and physical elements thanks to the integration of Information and Communication Technologies (ICTs). ICTs provide energy management algorithms and allow renewable energy integration and energy price minimization. Given the importance of renewable energy, many researchers developed energy management (EM) algorithms to minimize renewable energy intermittency. EM plays an important role in the control of users' energy consumption and enables increased consumer participation in the market. These algorithms provide consumers with information about their energy consumption patterns and help them adopt energy-efficient behaviour. In this paper, we present a review of the state of the energy management algorithms. We define a set of requirements for EM algorithms and evaluate them qualitativ...
In this demonstration, we present an autonomous anomaly detector for Cloud Radio Access Network (... more In this demonstration, we present an autonomous anomaly detector for Cloud Radio Access Network (C-RAN) performance metrics through a prototype using OpenAirInterface implemented as Docker containers. First, we show network slicing configuration capabilities including slice lifecycle management in addition to devices re-homing through our developed northbound interface allowing to manage a software-defined radio access network controller. Next, using conventional off-the-shelf smartphones, we run experiments to present the real-time data display in addition to the instant detection of anomalies in time-series CRAN data. We demonstrate a reaction to such detection through the auto-trigger of planned responses.
The main goal of wireless sensor networks (WSN) is to gather information from the regions of inte... more The main goal of wireless sensor networks (WSN) is to gather information from the regions of interest through a large number of micro sensor nodes. This gathering is traditionally done by using a client/server communication approach. However, this communication architecture consumes a lot of power and does not take into consideration the information properties. In this paper, we propose a data gathering scheme for WSNs, based on agents cooperation to deal with the importance of the information. This agent cooperation aims to reduce an important amount of the information communicated over the network by eliminating the unimportant information and the inter-sensor-nodes redundancy. This cooperation is empowered by an agent strategy taking into account several parameters related to the node and to the instance of communication for an optimized power management. Successive simulations proved, in large scale WSNs and different densities, the ability of the proposed gathering scheme to re...
The enhanced Interference Mitigation and Traffic Adaptation (eIMTA) mechanism is a key enabler fo... more The enhanced Interference Mitigation and Traffic Adaptation (eIMTA) mechanism is a key enabler for 5G networks and beyond. Knowing that a User Equipment (UE) cannot transmit more than the maximum power allowed by its power class. The level of available Transmission Power (TP) in each UE is an essential input for the Uplink (UL) scheduler of the next Generation Node-B (gNB). Scheduling higher data rate than what is supported by the available TP is a waste of resources. In the Downlink (DL), the power level is known by the gNB that manages the power amplifier and the DL-scheduler. Conversely, in the UL, the available power is estimated by the UE and sent to the gNB as a key input for eIMTA, known as Power Head-Room (PHR). In this context, we propose in this paper, a Deep Neural Network (DNN) based model to predict the PHR and reduce dependency on reported measures. We evaluate the effectiveness of our proposal in a 5G experimental prototype, based on Open Air Interface (OAI). Obtained...
Uploads
Papers by Rana Rahim