David Remondo

Ad hoc networking in future wireless communications

This paper contains an overview of the discussions on future research directions within the subject of ad hoc networking, held at the Wireless World Research Forum meetings during 2001. Ad hoc networking is an emerging research field where ad hoc networks are no longer viewed as stand-alone groups of wireless terminals. On the contrary, ad hoc networks are expected to become fundamental in the future development of infrastructure networks and they will be the basis for enabling ubiquitous communications. Ad hoc networking involves new research issues at all layers.

Adaptive importance sampling for performance evaluation and parameter optimization of communication systems

We present new adaptive importance sampling techniques based on stochastic Newton recursions. Their applicability to the performance evaluation of communication systems is studied. Besides bit-error rate (BER) estimation, the techniques are used for system parameter optimization. Two system models that are analytically tractable are employed to demonstrate the validity of the techniques. As an application to situations that are analytically intractable and numerically intensive, the influence of crosstalk in a wavelength-division multiplexing (WDM) crossconnect is assessed. In order to consider a realistic system model, optimal setting of thresholds in the detector is carried out while estimating error rate performances. Resulting BER estimates indicate that the tolerable crosstalk levels are significantly higher than predicted in the literature. This finding has a strong impact on the design of WDM networks. Power penalties induced by the addition of channels can also be accurately predicted in short run-times.

Energy-efficient wireless mesh infrastructures

The Internet comprises access segments with wired and wireless technologies. In the future, we can expect wireless mesh infrastructures (WMIs) to proliferate in this context. Due to the relatively low energy efficiency of wireless transmission, as compared to wired transmission, energy consumption of WMIs can represent a significant part of the energy consumption of the Internet as a whole. We explore different approaches to reduce energy consumption in WMIs, taking into account the heterogeneity of the technologies and the interaction with wired networks. Finally, we present an example scenario where the application of these methods is discussed.

A simple routing scheme for improving ad hoc network survivability

This article presents a new version of the dynamic source routing (DSR) protocol that favours the selection of a route containing nodes with high battery levels, in order to maximize the lifetime of the ad hoc mobile network as a whole. This proposal implies little modifications to the original algorithm. After a quantitative analysis, we see simulation results that indicate that the proposed scheme, SEADSR (simple energy aware DSR), outperforms the standard DSR in terms of network survivability and network capacity.

Integration of Optical and Wireless Technologies in the Metro-Access: QoS Support and Mobility Aspects

Future metropolitan and access networks are expected to comprise heterogeneous optical and broadband wireless technologies. The growing demand of users for transparent, ubiquitous access to diverse communication services poses several challenges. We envision a future metro-access architecture that comprises Optical Burst Switching networks that feed Ethernet Passive Optical Networks (PON) or upcoming Wavelength-Division Multiplexing PON, which in turn feed IEEE 802.16 and IEEE 802.11 nodes. Nodes in the wireless realm may communicate in a multihop fashion, forming mesh networks. To maintain cost and resource efficiency, we propose the introduction of Quality of Service (QoS) proxies at the border between different link technologies. These entities handle QoS requirements and aid to the support of mobility. The architecture requires no modification of the Medium Access Control mechanisms of the different technologies.

An interaction model between ad-hoc networks and fixed IP networks for QoS support

We propose a new protocol, named DS-SWAN (Differentiated Services-Stateless Wireless Ad Hoc Networks), to support end-to-end QoS (Quality of Service) in ad hoc networks connected to fixed DiffServ domains. DS-SWAN warns nodes in the ad hoc network when congestion is excessive for the correct functioning of real-time applications. These nodes react by slowing down best-effort traffic. Simulation results indicate that DS-SWAN significantly improves end-to-end delays for real-time flows without starvation of background traffic. In this work we introduce different conditions to force that a wireless node throttles its best-effort traffic and study their behaviour and differences.

A cooperation model between ad hoc networks and fixed networks for service differentiation

We propose a new protocol, named DS-SWAN (differentiated services-stateless wireless ad hoc networks), to support end-to-end QoS (quality of service) in ad hoc networks connected to fixed DiffServ domains. DS-SWAN warns nodes in the ad hoc network when congestion is excessive for the correct functioning of real-time applications. These nodes react by slowing down best-effort traffic. Simulation results indicate that DS-SWAN significantly improves end-to-end delays for real-time flows without starvation of background traffic.

QoS support between ad hoc networks and fixed IP networks

We propose a new protocol, named DS-SWAN (Differentiated Services-Stateless Wireless Ad Hoc Networks), to support end-to-end Quality of Service (QoS) in ad hoc networks connected to a fixed DiffServ domain. When congestion is excessive for the correct functioning of real-time applications, DS-SWAN determines the source of the problem, warns the nodes in the ad hoc network if it is the case and the nodes react by slowing down best-effort traffic. Furthermore, we present an associated routing protocol for the ad hoc network in this context, named SD-AODV (Service Differentiation-Ad Hoc On-Demand Distance Vector), where new route requests are suppressed at certain nodes to maintain the desired QoS requirements for real-time flows. A simulation study is presented for the case when traffic is sent from the ad hoc network towards the fixed network as well as in the opposite direction. The scalability of DS-SWAN with respect to the network size, number of real-time traffic sources, node mobility and best-effort traffic load is analyzed. Simulation results indicate that DS-SWAN and SD-AODV significantly improve end-to-end delays and jitter for real-time flows without starvation of background traffic.

Analysis of VBR VoIP traffic for ad hoc connectivity with a fixed IP network

We study how to provide end-to-end QoS (quality of service) between nodes in a mobile ad hoc network and a fixed IP network that supports differentiated services (DiffServ). The ad hoc network incorporates a stateless wireless ad hoc networks (SWAN) scheme to support service differentiation. SWAN uses local rate control for best-effort traffic and sender-based admission control for real-time traffic. Best-effort traffic and real-time flows that have not yet been admitted as real-time sessions are delayed by a best-effort traffic shaper. In this paper, we propose FA-SWAN (fast admission-SWAN), a modified version of the SWAN QoS scheme. In contrast to SWAN, FA-SWAN enables the real-time flows to bypass the traffic shaper during the admission control process. Only after a real-time session is rejected as such, will the traffic be considered as best-effort and will have to be rate controlled. The results indicate that the modifications improve the performance of the voice application in comparison with the original SWAN.

Quality of Service Support in Wireless Ad Hoc Networks Connected to Fixed DiffServ Domains

This paper analyzes the provision of end-to-end Quality of Service between nodes in a mobile ad hoc network and a fixed IP network that supports Differentiated Services. The ad hoc network incorporates the Stateless Wireless Ad Hoc Networks (SWAN) model to perform admission control for real-time traffic flows. We propose a new protocol, named DS-SWAN (Differentiated Services-SWAN), where end-to-end delays and loss rates of real-time traffic are monitored continuously at the destination nodes in the fixed network and at the edge routers respectively. In this way, nodes in the ad hoc network are warned when congestion is excessive for the correct functioning of a real-time application (specifically, Variable Bit Rate Voice-over-IP), so that the nodes restrain best-effort traffic in order to favour real-time flows. The results indicate that DS-SWAN significantly improves end-to-end delays without starvation of background traffic, adapting itself to changing traffic and network conditions in a relatively small ad hoc network. Besides, we compare different notification procedures in DS-SWAN aimed to improve scalability.