Content uploaded by Ralf Tönjes
Author content
All content in this area was uploaded by Ralf Tönjes
Content may be subject to copyright.
A preview of the PDF is not available
GSA: an architecture for optimising gateway selection in dynamic routing groups
Abstract and Figures
The proliferation of the increasing number of devices and technologies has brought about the extension of the concept of roaming users to roaming networks. This has inspired many researchers to investigate optimal mechanisms to enable communications for such roaming networks. Within the European 6th Framework Ambient Networks Project, we identify such roaming networks as dynamic routing groups, which consist of a number of different types of nodes with different capabilities. External communications from nodes within the routing group can be done via selected Gateway nodes. In this paper we present the gateway selection architecture (GSA) which provides support for Gateway identification, management and selection for nodes within a RG. We describe the benefits of this architecture and compare it to other known approaches
Figures - uploaded by Ralf Tönjes
Author content
All figure content in this area was uploaded by Ralf Tönjes
Content may be subject to copyright.
... In previous work [7], the Gateway Selection Architecture (GSA) was introduced to provide support for gateway identification, management, and selection within a routing group. Of course, one might expect that by grouping nodes and delegating mobility management to the cluster head and the gateways certain performance optimizations are possible as discussed in [7]. ...
... In previous work [7], the Gateway Selection Architecture (GSA) was introduced to provide support for gateway identification, management, and selection within a routing group. Of course, one might expect that by grouping nodes and delegating mobility management to the cluster head and the gateways certain performance optimizations are possible as discussed in [7]. Later, the performance of the GSA was evaluated by simulations in [8]. ...
... Of course, one might expect that by grouping nodes and delegating mobility management to the cluster head and the gateways certain performance optimizations are possible as discussed in [7]. The motivation for including the GWSs aims at simplifying the signalling overhead regarding gateways and their capabilities. ...
Many anticipate a future wireless world filled by a multitude of user devices and wireless technologies. Effective management of this kind of heterogeneous, mobile, and rapidly changing ad hoc networks will be a challenging task. We present and evaluate the Ambient Networks Gateway Selection Architecture (GSA), which provides support for gateway discovery, management, and selection for mobile nodes within dynamic routing groups. A routing group (RG) is a cluster of nodes in physical proximity, aware of the group membership, with a common goal of optimizing mobility management and routing functionality in the group. A gateway is a mobile node that provides packet relaying and connectivity services to other nodes in the RG. GSA can be also used outside the Ambient Networks architecture, and we present how it can be used with two existing mobility management protocols, namely Mobile IP and Host Identity Protocol, especially in the case of moving networks. Our simulation studies show the benefits gained from group formation when compared to same functionalities implemented in every individual node. We also compare the GSA hybrid signaling strategy with proactive and reactive approaches; the simulation results show that the hybrid approach scales better when the routing group size grows. Keywords—Ambient Networks, gateway selection, Host Iden- tity Protocol (HIP), mobile computing, Mobile IP (MIP), mobility management, moving networks, routing group
... In previous work [3], the Gateway Selection Architecture (GSA) was introduced to provide support for gateway identification , management, and selection within a routing group. Emphasis was given to node clustering and gateway selection. ...
... This allows even more possibilities for optimizations than a cluster. Of course, one might expect that by grouping nodes and delegating mobility management to the cluster head and the gateways certain performance optimizations are possible as discussed in [3]. The second contribution, presented in §V-VI is an attempt to quantify using simulation the benefits from employing GSA in scenarios where several nodes move in a mass transit vehicle. ...
We study the performance of the gateway selection architecture (GSA), developed within the framework of the EU ambient networks integrated project, which provides support for gateway discovery, management, and selection for mobile nodes within dynamic routing groups. A routing group (RG) is a cluster of nodes in physical proximity, aware of the group membership, especially in the context of the Ambient Networks architecture, with a common goal of optimizing mobility management and routing functionality in the group. We present two examples on how GSA can be introduced using existing protocols, namely MIP and HIP. Our simulation studies show the benefits gained from group formation when compared to same functionalities implemented in every individual node, and compare the GSA optimized signaling strategy with other competitive approaches.
... The Gateway Selection Architecture (GSA) [2] was developed based on the assumption that some RG nodes are not capable or willing to do GW selection for themselves. GSA introduces special Gateway Selector (GWS) nodes that make the GW decision on behalf of other nodes. ...
This chapter addresses one of the challenges in cooperative networking, namely, mobility support in a heterogeneous ambient network environment. We motivate the need for efficient mechanisms for handling the large amount of network and channel state information required in assisting fast handovers and network and service adaptation strategies. Managing a variety of network and protocol events and triggering information is a challenging task even in a homogeneous networking environment when different mobility schemes (node, network, session) and application adaptation are considered and, not unexpectedly, the heterogeneity of access networks increases further the amount of such information. We present triggering management mechanisms which efficiently handle triggering information at node and network level, dealing with a greater variety of events originating from any component of the node's protocol stack as well as mobility management entities within the network. We then discuss the benefits of arranging mobile nodes into specific mobile routing groups, and how such approaches can benefit from the availability of the triggering management mechanisms in an ambient network environment.
Forthcoming wireless communication systems, well represented by the term beyond 3G, are likely to impose some new requirements that go beyond the traditional view on today's networking paradigm. In particular, mobility procedures will no longer be restricted to the change of the point of attachment to the network. The work presented in this paper aims at proving, following a fully experimental approach, the feasibility of some architectural components of a mobility control space, which has been designed so the context of the ambient networks project. Especially, in this study we focused on and successfully realized two concepts, a facility for triggering mobility events and support for moving networks
TUB), Germany, hollos@tkn.tu-berlin.de Abstract— Mobile nodes are generally treated individually and in an uncoordinated fashion, even when nodes are, in fact, moving as a group. However, the ability to recognize that the nodes are travelling together enables a number of important optimizations to routing and mobility management mechanisms, in particular, aggregation and delegation of mobility management and use of ad-hoc routing protocols. The key challenges to providing these optimizations are recognizing the presence of a group of nodes, assessing the stability of the group, and deciding what optimizations are appropriate. Developing solutions to these challenges has been a key focus of the mobility work within the Ambient Networks project and this paper discusses these challenges and presents the solutions developed.
When a node in a mobile ad hoc network wants to send data packets to the Internet, and therefore outside of its local ad hoc network, it has to obtain information about the available Internet gateways: i.e. which one to use and how to get there. To accomplish this, nodes can utilize either a unsolicited gateway discovery mechanism or relay on unsolicited gateway advertisement packets sent by gateways. Obviously, the effectiveness of periodic and unsolicited advertisement depends entirely on the traffic and mobility patterns. The most important factors involved in sending unsolicited gateway advertisements are the time interval between sending two consecutive advertisements and the TTL value of the advertisement packet, i.e., the gateway should carefully decide when to send advertisements, and the advertisement flooding area should be limited only to nodes that need to update their gateway information and their paths to the gateway. In this paper, two kinds of advertising schemes are proposed, which are based on the observation of traffic and mobility patterns, and are designed to avoid generating unnecessary packets in the MANET (Mobile Ad hoc Network), in addition to giving mobile nodes more opportunity to use the shortest path to the Internet.
Mobile ad hoc networks are autonomous, infrastructureless networks that support multihop communication through IP routing. This paper examines the use of mobile IP in order to provide global Internet connectivity to ad hoc networks that use an on-demand routing protocol. We present a hybrid scheme that uses techniques such as TTL scoping of agent advertisements, eavesdropping and caching agent and advertisements to combine the advantages of proactive and reactive approaches to providing connectivity. We present simulation results to show that our approach achieves excellent connectivity while keeping overhead costs low.
This article presents a multilayer mobility management scheme for all-IP networks where local mobility movements (micro-mobility) are handled separately from global movements (macro-mobility). Furthermore, a hybrid scheme is proposed to handle macro-mobility (mobile IP for non-real-time services and SIP for real-time services). The internetworking between micromobility and macro-mobility is implemented at an entity called the enhanced mobility gateway. Both qualitative and quantitative results have demonstrated that the performance of the proposed mobility management is better than existing schemes. Furthermore, a context transfer solution for AAA is proposed to enhance the multilayer mobility management scheme by avoiding the additional delay introduced by AAA security procedures.
Beyond-3G (B3G) systems have been envisaged as an evolution and convergence of mobile/wireless communication systems and IP technologies to offer a multitude of services over a variety of access technologies. To fulfill the vision, it is necessary to understand the requirements with respect to the support of heterogeneity in network accesses, communication services, mobility, user devices, and so on. Besides, it is equally important to promote the necessary research in networking technology by providing a guiding framework of research areas and technical issues with priority. The new architectures and technologies have to address the fundamental assumptions and requirements that govern the design. All these issues are being tackled by the cooperative network group (CoNet) of WWRF; the group is working on a series of white papers outlining B3G visions and roadmap, architectural principles, research challenges, and candidate approaches. This article outlines the CoNet concept, architectural principles, and guidelines for research into cooperative networks assuming that the B3G systems are built over generic IP networking technologies. The article also presents the key research challenges, research framework, and major network components and technologies. The key points are that the system should be layered on demand, encourage reuse if independent modularized functional blocks, support multiple services and service creation, ensure consistent end-to-end connectivity across different access technologies, and cooperate in terms of network control, operations, and maintenance. The architecture includes the endpoints of communications as part of the communications system, and provides a secure and trusted environment in which network functions are performed; the network should self-organize dynamically. Additionally, the article presents the IST WSI architecture proposed to CoNet as reference model along with some approaches to the outlined research challenges. Finally, this study selects three important network components and technologies (i.e., mobility management, multiple access, and moving networks) in order to provide with answers and possible solutions the research challenges presented in earlier sections.
The growing deployment rate of wireless LANs indicates that wireless networking is rapidly becoming a prevalent form of communication. As users become more accustomed to the use of mobile devices, they increasingly want the additional benefit of roaming. The Mobile IP protocol has been developed as a solution for allowing users to roam outside of their home networks while still retaining network connectivity. The problem with this solution, however, is that the deployment of foreign agents is expensive because their coverage areas are limited due to fading and interference. To reduce the number of foreign agents needed while still maintaining the same coverage, ad hoc network functionality can cooperate with Mobile IP such that multihop routes between mobile nodes and foreign agents can be utilized. In this work, we present a method for enabling the cooperation of Mobile IP and the Ad hoc On-Demand Distance Vector (AODV) routing protocol, such that mobile nodes that are not within direct transmission range of a foreign agent can still obtain Internet connectivity. In addition, we describe how duplicate address detection can be used in these networks to obtain a unique co-located care-of address when a foreign agent is not available.
When an ad hoc network is connected to the Internet, it is important for the mobile nodes to detect the available gateways providing access to the Internet. Therefore, a gateway discovery mechanism is required. The two main approaches for discovering Internet gateways are the reactive and the proactive one. This paper compares the performance of these approaches in various scenarios by means of simulation. We show that the proactive approach performs better in the simulated scenario.
Inter-vehicle communication systems play an important role in future road communication scenarios like FleetNet. For such communication scenarios, ad hoc networks offer a promising approach due to their characteristics such as low latency and cost efficiency. While multi-hop ad hoc network communication among vehicles provides for many interesting and important applications (e.g., traveling safety, smoothed traffic flow), users also will be interested in accessing Internet services. This access can be achieved via roadside installed Internet gateways. However, the Internet integration of inter-vehicle communication systems entails several difficulties, such as mobility support, communication efficiency, or the discovery and handover of connections from one gateway to the next. In this paper, we propose a communication architecture that addresses these issues. Our communication architecture combines a proxy-based approach with Mobile IP. Additionally, the architecture integrates the passengers' mobile devices used within the vehicles, i.e. they are able to access the Internet using the inter-vehicle communication system.