Despite the massive increase in network capacities and devices capabilities, intra-domain TE problems have remained a challenge due to high traffic demand in dynamic network settings. In this paper, we fundamentally adopt and construct the use of Deep Reinforcement Learning (DRL) algorithm to improve end-to-end delay and maximum link utilisation (MLU) in a wireless All-IP access network. In an experimental pseudo setting, we propose a routing method DRL-MPR (Multi-Plane Routing) within a Dueling Deep Q-Network (DDQN) framework as a controller in which by simultaneously allocating multiple robust IP sessions, each session will take the near optimal path/plane. We first revisit Multi-Plane Routing Protocol (MPR) to configure a set of routing paths/Routing Planes (RP) from a physical topology as pre-condition ahead of the traffic injection. We compare our decision making scheme results with some traditional routing methodologies such as OSPF and standard Multi-Plane Routing (MPR). Our simulation results show that DRL obtains lower maximum link utilisation (MLU) and end-to-end delay. For example, in the 19 nodes and 41 links topology, a small traffic load results in 60% maximum link utilisation for DRL-MPR, whereas those of OSPF and MPR are 100% and 94% respectively. In this simulation environment, the traffic demand in the IP access network is gradually increasing to the point of saturated network link utilisation.

Wireless access networks are ever-changing their setup, topologies and provisions of coverage. Forthcoming deployment trends include densification of cells and convergence of the ‘native IP’ with cellular access networks with more than best-effort expectations. Greater meshing in networks due to practical deployment requirements is inevitable and needed. This paper proposes a routing protocol to address the “randomness” of topology interconnections, routing paths and sizes in emerging IP access networks. Termed as Minimum Set Cover (MSC) approach, it is a generalization of the NP and NP complete mathematical problem. The offline component of the MSC multiplies intra-domain routing installations, called Routing Planes (RPs) modelled as graphs, to suitably cover the whole routing topology prior to the traffic injections in the online component. We introduce novel offline optimization features using a dynamic cost function that plans availability of capacities and correlation of routing paths via the chosen set of RPs. Our simulations verify effective path diversity using MSC with its modest protocol overhead and a heuristic used for RPs selection. For randomly constructed access network topologies with 7, 18 and 33 routers and various meshing levels, our results show convincing suitability and key performances gains compared with rival routing solutions.

The expected variable nature of future access network structures calls for network providers to adapt accordingly in order to facilitate loss free networks supported with the developing IP infrastructure. To this end, a routing optimization mechanism applicable to randomly shaped all-IP access networks must be developed. In this paper, Multi-Plane Routing (MPR) that consolidates various aspects in all-IP infrastructure as a whole is redesigned and reformulated to provide a comprehensive solution in consideration of the randomness of future access networks. We prove the offline Routing Plane (RP) construction problem as being the generalization of the Minimum Set Cover (MSC) problem which is NP and also NP-complete. To this end, we propose a novel MSC-based paths-diverse offline TE algorithm which is comprehensively applicable to random wireless access network structures. Our simulation results demonstrate the constructed RPs for complex random networks of different sizes and sparseness (i.e. meshing). Hence, the comprehensive applicability of our novel approach is verified. Moreover, we propose a new optimization framework along with a dynamic cost function (considering capacity and correlation of paths) that formally describe our novel offline TE mechanism for future randomly shaped wireless IP access networks.

With the expected surge in the global IP traffic, service providers would need to adapt accordingly to operate disruption and loss free networks supported with the developing IP infrastructure. With the disposal of the hierarchical network structure, radio access networks are moving towards a flat-IP architecture and novel topological set-ups in the backhaul. Hence, a routing paradigm that employs suitable Traffic Engineering (TE) techniques aligned with the developing nature of future access networks must be applied. It becomes imminent that the routing considerations for IP access networks converge with the ones found in conventional intra-domain routing. In this paper, Multi-Plane Routing (MPR) that consolidates various aspects in all-IP infrastructure is extensively studied in access network structures. We propose a MPR-based TE approach considering two different scenarios to reflect the evolution in the architectural design of access network structures under a realistic traffic scenario with a varying range of internal/external traffic. Moreover, a new optimization framework for the offline and online TE mechanisms of MPR have been formulated. Accordingly, a practical performance evaluation testing the validity of the aforementioned scenarios is presented. Our simulation results demonstrate extensive analysis in terms of several performance criteria in networks. It is convincingly shown that for ranges of topologies, MPR's utilization of whole topology in building path diversity in networks, allows for significant improvement of networks capacity, performance, and support for meshing.

There has been a rapid rise in the IP traffic throughout the Internet which takes advantage of the already established widespread IP infrastructure. Different suggestions are being explored to facilitate the next-generation access networks via IP mechanisms, with a growing trend towards a flat-IP structure and novel topological set-ups in the backhaul. Aligned with this evolution, there are increasingly more user applications flooding the Internet that calls for a consistent routing strategy to minimize loss in data transmission. In this paper, Multi-Plane Routing (MPR), which incorporates various aspects in all-IP infrastructure will be studied under the new access network structure. MPR is based on Multi-Topology Open Shortest Path First (MT-OSPF) principle and divides the physical network topology into several logical Routing Planes (RPs). The offline Traffic Engineering (TE) strategy for MPR has been optimized using a heuristic hop-constraint solution that suits the “flattened” network realized through the incorporation of direct communication between Aggregation Routers. With our approach, despite of a higher number of Ingress —Egress pairs for traffic in the access network, the number of RPs has been kept to the desirable level whilst the reliability indicator and the path diversity index ratio have increased up to 47% and 33% respectively. Our proposed MPR-based offline approach has also shown improvement compared with the Multi-Protocol Label Switching (MPLS) offline approach.

With the increasing interest in continuous media streaming applications such as audio and video, quality of service (QoS) is becoming more and more important. Providing QoS guarantees in packet access networks gives rise to several challenging issues. There are several aspects of QoS to be considered. For example, to support video communication high throughput is required and therefore, high bandwidth guarantees will have to be made. Audio communication, in contrast, does not usually require high bandwidth. End-to-end delay and delay variations are other factors that must be taking into account for time-critical traffic such as VoIP. In this paper, we propose a new mechanism using Multi-Plane Routing (MPR) principle based on multi-topology OSPF (MT-OSPF). Our approach, Q-MPR, leverages multiple alternate routing planes through a network for enabling maximum path diversity, and then maps traffic flows onto specific planes, using a multi-constrained cost function, respecting QoS requirements and improving quality of experience seen from end users. A maximum gain of 55%, 91% and 82% in throughput, packet loss and one-way packet delay can be achieved using our approach.

It becomes more and more recognized that using IP as the underlying foundation for the next-generation of wireless networks makes strong economic sense. This is evident in installation and operation, and technical aspects as it takes advantage of the ubiquitous installed IP infrastructure. Because of that, a resilient, efficient and easy implementable routing paradigm is needed. Multi-Plane Routing (MPR) is a routing technique, using the recent multi-topology approach, that leverages multiple alternative paths (from each source-destination pair) through a network. In this paper, we introduce MPR for wireless access networks and show that, with an optimal link weight assignment, it can reduce the maximum link utilization by up to 40% and end-to-end delay by up to 90%. MPR also increases the overall network performance by maximizing path diversity across the routing planes and splitting traffic using a branching factor. Finally, our experiments show that our nascent approach can achieve high performance in next-generation all-IP wireless access networks with a moderate number of routing planes.

This paper presents the functionalities involved in enabling the use of compartments in wireless access networks for achieving capacity and coverage optimization. Compartments are defined as dynamic medium-term federations of collaborating elements for intelligent organizational purposes. The work presented in the paper uses compartments as a self-management apparatus for emerging wireless access environments and proposes a compartment formation algorithm for achieving self-optimization of the coverage of access points. Facilitation of the compartment apparatus is founded in the synergy of self-management tools and the application of cognition in operations of Future Internet networks. The paper presents the details of cognition managers in access points for localized management by forming compartments and proceeds to the analysis as well as providing results in coverage optimizations of wireless networks.

Many existing communication protocols fail to perform efficiently when groups of users move in unison and at high velocity. A classic example is the Mobile IP protocol, which, as it stands, leads to storms of binding updates at every IP handover, but is resolved by the NEMO basic support protocol. However, a problem of similar nature is also existent in Quality of Service protocols, which again tend to produce signalling storms at handover, leading to, somewhat paradoxically, surges of congestion within the access network. While QoS aggregation improves protocol scalability by maintaining a single QoS state for potentially the entire moving network rather than for each individual node, the act of having to constantly maintain this state when sessions are frequently created and terminated undermines its benefit. We therefore propose a dynamic cost-driven QoS aggregation policy, which aims at minimising network operator costs for each aggregation cycle. Simulations of the cost-driven policy show that costs are reduced by up to 6.5% relative to other QoS aggregation policies, with expected user waiting times prior to session establishment increasing by a only a fraction of a second.

In this paper we provide an overview of the ASCOT project and present intermediate results on investigations concerning address management. ASCOT considers managed vehicular networks, i.e. moving networks attached to public transport vehicles that provide network connectivity to its passengers. We briefly describe how ASCOT positions itself among other projects and activities that focus on moving network aspects. We consider addressing: the pros and cons of the various options for the addressing of end user terminals - including a brief discussion of the relevant end user requirements and the impact on mobility management.

Amongst the activities of the European Commission IST research project-MIND is extensive consideration of the multi-homing issues in IP access networks. Analysis of the various multi-homing scenarios reveals diverse aspects of multi-homing relevant to both the operations of the access network as well as the hosts connected to it. The originality of the work presented in this paper arises from the uniqueness of the multi-homing setup in IP access networks that support terminal mobility such as the one developed in the BRAIN project and the ad hoc and self-organising network extensions developed in MIND. This paper presents a part of our work on multi-homing by showing the required architectural support from the involved network elements and utilisation of the existing general mechanisms for the support of IP connectivity.

TCP is a reliable connection-oriented transport protocol that performs well in traditional networks. However, in networks with wireless and other lossy links the protocol suffers from losses and delays due to frequent handoffs like in wireless networks running mobile IP. Many micromobility protocols have been proposed to reduce handoff latency and the load on the network when a mobile host moves among small wireless cells. In this paper, the performance of several TCP and UDP schemes over one of these protocols, MMP (multicast for mobility protocol), is compared and improvements to TCP in a multiple handoff network are proposed. The simulation results are presented using the Network Simulator-ns2 and different metrics for comparison. Our results show that the number of UDP packets lost is reduced when using MMP instead of mobile IP in a micromobility environment and that TCP provides better throughput performance for a traditional scheme like Tahoe TCP when it is adapted to the conditions of the network.

In this paper we suggest a classification scheme for IP mobility protocols and propose a framework for comparing them. We then use the framework to provide an initial comparison of previous proposals for supporting micro-mobility. The authors are part of project BRAIN (Broadband Radio Access for IP based Networks)-a European collaborative project under the IST (Information Societies Technology) programme. One aim of the project is to propose an open architecture for wireless broadband Internet access, concentrating on issues in the access network.

This paper contains a design description for an IP-mobility protocol called multicast for mobility protocol (MMP). The protocol is intended to provide fast and efficient location management and routing for mobile hosts in internet networks. The general configuration of the protocol which is divided into two parts: global and local, or macro and micro mobility, makes it an optimal and attractive solution for the emerging internet-based mobility-enabled stub networks, referred to as IP access networks. This version of MMP contains various enhancements to the initial version such as the introduction of a new control message MMP instruct and protocol mechanisms for handling new protocol issues such as: paging, support for mobile sources, advance registrations, implementation adaptations. However, the initial main concepts of mobile IP-controlled macro mobility and micro mobility enabled by a sparse mode multicast protocol called core based trees (CBT) are still the main characteristics of the protocol.