Ramy Farha

A generic architecture for autonomic service and network management

As the Internet evolves into an all-IP communication infrastructure, a key issue to consider is that of creating and managing IP-based services with efficient resource utilization in a scalable, flexible, and automatic way. In this paper, we present the Autonomic Service Architecture (ASA), a uniform framework for automated management of both Internet services and their underlying network resources. ASA ensures the delivery of services according to specific service level agreements (SLAs) between customers and service providers. As an illustrative example, ASA is applied to the management of DiffServ/MPLS networks, where we propose an autonomic bandwidth sharing scheme. With the proposed scheme, the bandwidth allocated for each SLA can be automatically adjusted according to the measured traffic load and under policy control for efficient resource utilization, while SLA compliance over the network is always guaranteed.

Multipath delay estimations using matrix pencil

This paper presents a technique for the recovery of time delays associated with components of a signal in a multipath communication channel. Matrix pencil is used to recover these delays from the channel frequency response. This algorithm has some key advantages over traditional super-resolution techniques such as MUSIC. Most importantly, matrix pencil only requires a single channel estimate and can estimate the delays associated with coherent multiple components.

Virtual network approach to scalable IP service deployment and efficient resource management

As the Internet evolves into a global all-service communication infrastructure, a key consideration is providing quality of service guarantees over IP with efficient resource utilization in a scalable, flexible, and automatic way. In this article we present a virtual network (VN) based architecture for scalable IP service deployment and efficient network resource management. Particularly considering a DiffServ/MPLS III transport network supporting multiple VNs, we propose a dynamic approach for efficient bandwidth sharing among VNs. The bandwidth sharing is service-level-agreement-based; the spare capacity in underloaded VNs is adaptively and efficiently utilized, and SLA compliance for all the VNs involved is always guaranteed.

Peer-to-Peer Mobility Management for all-IP Networks

With the increasing number of wireless devices, the importance of mobility management in future mobile networks is growing. Traditional mobility management approaches are based on client/server paradigms, and suffer from their well-known shortcomings (single point of failure, congestion, bottlenecks). With the success of P2P for file sharing applications, we believe that its benefits can be brought into new mobility management schemes to improve their scalability, robustness, availability, and performance. To the best of our knowledge, this paper is a first attempt to examine the potential of P2P concepts for mobility management. We perform experiments to quantify the performance of the proposed scheme, and compare it to traditional approaches such as Mobile IP.

Blueprint for an Autonomic Service Architecture

Service providers offer an array of services and applications that are media-rich, personalized, and context-aware. In this environment, new services and applications are introduced, provisioned, operated, maintained, and retired at a pace that tracks changing customer requirements and demands. In order to be cost-effective, these services and applications need to be delivered over an all-IP infrastructure, which is autonomically managed to ensure their delivery at a satisfactory level for subscribed customers. In this paper, we propose a blueprint for a generic autonomic service architecture (ASA) to address these challenges

Market-Based Resource Management for Cognitive Radios Using Machine Learning

With the growth in wireless network technologies and the emergence of cognitive radios, the need arises for mechanisms to effectively manage the resources involved in such environments. In this paper, we propose a market-based resource management approach for cognitive radios using machine learning, consisting of a negotiation phase where nodes are allocated resources in order to meet their requested bit rate, and a learning phase where nodes adjust their pricing of the resources in order to steer the cognitive radio environment towards the greater network good. We are interested in improving the utilization of the resources through price adjustments as compared to the case where the prices are kept fixed. We perform extensive simulations to study the performance of the proposed resource management mechanism in the cognitive radio environment.

Peer-to-Peer Vertical Mobility Management

With the increasing number of wireless devices, the importance of mobility management in future mobile networks is growing. In addition, the number of access technologies available to those wireless devices is more diversified, leading to further heterogeneity and the need for convergence of mobility management solutions. Furthermore, traditional mobility management solutions, which are based on client/server paradigms, suffer from the shortcoming of such centralized solutions (single point of failure, congestion, bottlenecks). With the initial success of peer-to-peer for file sharing applications, and based on our previous attempt to exploit its benefits by proposing novel horizontal mobility management schemes, we examine in this paper the potential of peer-to-peer for vertical mobility management. We also perform extensive simulations to quantify the performance of our proposed peer-to-peer vertical mobility management scheme.

Mobility analysis for all-IP networks

This paper discusses and compares different approaches to handle mobility in all-IP networks, at both the application and network layers. First, three key parameters of mobility schemes are quantified: handoff delay, registration/binding cost, and packet delivery cost. Next, these schemes are compared based on criteria such as the need for encapsulation, changes in endsystems and/or infrastructure, scalability, and reliability. Finally, a mobility model is used to perform extensive simulations in order to compare the different all-IP mobility approaches.

Towards an autonomic service architecture

Traditional telecommunications service providers are undergoing a transition to a shared infrastructure in which multiple services will be offered to customers. These services will be introduced, modified, and retired at a pace that tracks changing requirements and demands. In order to be cost-effective, these services will need to be delivered over a shared infrastructure that is managed to support delivery requirements at a given point in time. In this paper, we present an Autonomic Service Architecture (ASA) for the automated management of networking and computing resources. ASA ensures the delivery of services according to specific agreements between customers and service providers.

A novel peer-to-peer naming infrastructure for next generation networks

One of the major challenges in next generation networks is naming, which allows the different entities on the network to identify, find, and address each other. In this paper, we propose a novel Peer-to-Peer naming infrastructure, which takes into account the expected changes in the next generation networks due to the trends shaping network evolution. The success of the Peer-to-Peer paradigm for applications such as file sharing and instant messaging has lead to research on other areas where such a paradigm could be useful, to provide the scalability, robustness, and flexibility that characterize Peer-to-Peer applications.