Machine Learning (ML)-based network models provide fast and accurate predictions for complex network behaviors but require substantial training data. Collecting such data from real networks is often costly and limited, especially for critical scenarios like failures. As a result, researchers commonly rely on simulated data, which reduces accuracy when models are deployed in real environments. We propose a hybrid approach leveraging transfer learning to combine simulated and real-world data. Using RouteNet-Fermi, we show that fine-tuning a pre-trained model with a small real dataset significantly improves performance. Our experiments with OMNeT++ and a custom testbed reduce the Mean Absolute Percentage Error (MAPE) in packet delay prediction by up to 88%. With just 10 real scenarios, MAPE drops by 37%, and with 50 scenarios, by 48%.

Multi-agent LLM systems are an emerging networked workload whose rapid deployment raises questions about the traffic patterns they generate. Compared to conventional applications, these systems generate requests internally: a single user task can induce a structured sequence of model calls whose timing is governed by coordination logic rather than by user arrival rate. It is not clear whether classical traffic models, designed for human-driven workloads, apply to this setting.

Network performance modeling is a field that predates early computer networks and the beginning of the Internet. It aims to predict the traffic performance of packet flows in a given network. Its applications range from network planning and troubleshooting to feeding information to network controllers for configuration optimization. Traditional network performance modeling has relied heavily on Discrete Event Simulation (DES) and analytical methods grounded in mathematical theories such as Queuing Theory and Network Calculus. However, as of late, we have observed a paradigm shift, with attempts to obtain efficient Parallel DES, the surge of Machine Learning models, and their integration with other methodologies in hybrid approaches. This has resulted in a great variety of modeling approaches, each with its strengths and often tailored to specific scenarios or requirements. In this paper, we comprehensively survey the relevant network performance modeling approaches for wired networks over the last decades. With this understanding, we also define a taxonomy of approaches, summarizing our understanding of the SotA and how both technology and the concerns of the research community evolve over time. Finally, we also consider how these models are evaluated, how their different nature results in different evaluation requirements and goals, and how this may complicate their comparison.

Network simulation is pivotal in network modeling, assisting with tasks ranging from capacity planning to performance estimation. Traditional approaches such as Discrete Event Simulation (DES) face limitations in terms of computational cost and accuracy. This paper introduces RouteNet-Gauss, a novel integration of a testbed network with a Machine Learning (ML) model to address these challenges. By using the testbed as a hardware accelerator, RouteNet-Gauss generates training datasets rapidly and simulates network scenarios with high fidelity to real-world conditions. Experimental results show that RouteNet-Gauss significantly reduces prediction errors by up to 95% and achieves a 488x speedup in inference time compared to state-of-the-art DES-based methods. RouteNet-Gauss’s modular architecture is dynamically constructed based on the specific characteristics of the network scenario, such as topology and routing. This enables it to understand and generalize to different network configurations beyond those seen during training, including networks up to 10x larger. Additionally, it supports Temporal Aggregated Performance Estimation (TAPE), providing configurable temporal granularity and maintaining high accuracy in flow performance metrics. This approach shows promise in improving both simulation efficiency and accuracy, offering a valuable tool for network operators.

We present TSGFM, a Time Series Graph Foundation Model for zero-shot network monitoring, leveraging spatiotemporal Graph Neural Networks (GNNs) to extract transferable representations across diverse multivariate time series (MTS) domains. Pretrained on heterogeneous time series datasets, TSGFM enables generalization without task-specific fine-tuning, addressing core challenges in dynamic network environments. TSGFM is benchmarked across five real-world MTS datasets and seven zero-shot forecasting scenarios, outperforming five state-of-the-art baselines in six out of seven tasks. Most notably, in zero-shot network monitoring analysis, TSGFM surpasses all competing models by at least 18%, even without any prior exposure to network monitoring data. We further compare TSGFM against leading Time Series Foundation Models (TSFMs), including TimeGPT and TimesFM. TSGFM achieves performance on par with TimeGPT, occasionally surpassing it, and consistently outperforms TimesFM, while using significantly less pretraining data and relying on a much simpler architecture. A detailed analysis of TSGFM’s learned spatial attention patterns reveals domain-specific connectivity structures. In particular, lower attention weights in network monitoring tasks suggest that dense spatial graphs may be unnecessary, opening opportunities for efficient spatial pruning without sacrificing accuracy. This challenges prevailing assumptions favoring fully connected spatiotemporal GNNs. To foster transparency and reproducibility, we release the complete implementation of TSGFM as open source, as well as the tested datasets.

Congestion Control (CC) plays a fundamental role in optimizing traffic in Datacenter Networks (DCNs). Currently, DCNs implement two main CC protocols: DCTCP and DCQCN. Both protocols are based on Explicit Congestion Notification (ECN), where switches mark packets when they detect congestion. Nowadays, network experts carefully set ECN parameters to optimize the average network performance. However, today’s DCNs experience rapid and abrupt changes that severely affect the network state (e.g., dynamic workloads, incasts), which leads to under-utilization and sub-optimal performance. In this paper we present GraphCC , a framework for in-network CC optimization. GraphCC relies on Multi-agent Reinforcement Learning (MARL) and Graph Neural Networks (GNN), and is compatible with widely deployed ECN-based CC protocols. The proposed solution deploys distributed agents on switches that communicate with their neighbors to cooperate and optimize the global ECN configuration. In our evaluation, we test GraphCC with three real-world traffic workloads, focusing on its capability to accommodate scenarios unseen during training (e.g., traffic changes, failures). We compare GraphCC with a state-of-the-art MARL solution for ECN tuning, and observe that our method outperforms the state-of-the-art baseline in all evaluation scenarios, with improvements up to 20% in average Flow Completion Time, similar mean throughput (within 1%), and significant reductions in buffer occupancy (38.0–85.7%).

The Border Gateway Protocol (BGP) is central to the global connectivity of the Internet, enabling fast and efficient dissemination of routing information. Hence, detecting any anomaly concerning BGP announcements is of critical importance to ensure the continuous operation of Internet services. Typically, BGP anomaly detection algorithms have relied on features of the BGP messages, such as the average length of the AS_PATH attribute, the volume of messages, or the type of message (announcement or withdrawal). Even though these algorithms provide good performance, they do not take into account the Internet topology, that is, the graph of Autonomous Systems (AS) created by the BGP announcements. In addition, some of the existing algorithms can detect only specific types of anomalies, while others require retraining them to support new scenarios. In this paper we propose detecting BGP anomalies by leveraging the raw BGP topology graph, instead of manually curated features of the BGP messages. We implement a Machine Learning algorithm to process the entire BGP topology and evaluate it with real-world data from 4 well-known incidents. We compare our proposal against two state-of-the-art solutions and a classical method that use BGP features and features of the BGP topology, not the topology itself. Our results show that our solution obtains remarkable performance identifying the incidents. Finally, we test our model with regular data (non-anomalous) to prove that it can be used in a production scenario, with samples processed on the fly and guaranteeing a low false alarm rate.

In the present-day, distributed applications are commonly spread across multiple datacenters, reaching out to edge and fog computing locations. The transition away from single datacenter hosting is driven by capacity constraints in datacenters and the adoption of hybrid deployment strategies, combining on-premise and public cloud facilities. However, the performance of such applications is often limited by extended Flow Completion Times (FCT) for short flows due to queuing behind bursts of packets from concurrent long flows. To address this challenge, we propose a solution to prioritize short flows over long flows in the Software-Defined Wide-Area Network (SD-WAN) interconnecting the distributed computing platforms. Our solution utilizes eBPF to segregate short and long flows, transmitting them over separate tunnels with the same properties. By effectively mitigating queuing delays, we consistently achieve a 1.5 times reduction in FCT for short flows, resulting in improved application response times. The proposed solution works with encrypted traffic and is application-agnostic, making it deployable in diverse distributed environments without modifying the applications themselves. Our testbed evaluation demonstrates the effectiveness of our approach in accelerating the run-time of distributed applications, providing valuable insights for optimizing multi-datacenter and edge deployments.

Wireless communications at the chip scale emerge as a interesting complement to traditional wire-based approaches thanks to their low latency, inherent broadcast nature, and capacity to bypass pin constraints. However, as current trends push towards massive and bandwidth-hungry processor architectures, there is a need for wireless chip-scale networks that exploit and share as many channels as possible. In this context, this work addresses the issue of channel sharing by exploring the design space of multi-channel Medium Access Control (MAC) protocols for chip-scale networks. Distinct channel assignment strategies for both random access and token passing are presented and evaluated under realistic traffic patterns. It is shown that, even with the improvements enabled by the multiple channels, both protocols maintain their intrinsic advantages and disadvantages.

Programmable metasurfaces constitute an emerging paradigm, envisaged to become a key enabling technology for Reconfigurable Intelligent Surfaces (RIS) due to their powerful control over electromagnetic waves. The HyperSurface (HSF) paradigm takes one step further by embedding a network of customized integrated circuit (IC) controllers within the device with the aim of adding intelligence, connectivity, and autonomy. However, little is known about the traffic that the network needs to support as the target electromagnetic function or boundary conditions change. In this paper, the framework of a methodology is introduced to characterize the workload of programmable metasurfaces which is then used to analyze the beam steering HSFs. The workload characterization leads to many useful insights into traffic behavior, including the spatio-temporal load incurred and the HSF limitations in terms of fine-grained tracking of moving targets. It is observed that the traffic is inherently bursty with an uneven spatial distribution of load and that finer resolution comes at the cost of an increased but less bursty load. An indoor mobility model indicates reasonable signaling load on the deployed surfaces. Finally, a statistical analysis on the traffic patterns is performed, showing that the incoming traffic can be well represented by an ON-OFF model.

Network models are an essential block of modern networks. For example, they are widely used in network planning and optimization. However, as networks increase in scale and complexity, some models present limitations, such as the assumption of Markovian traffic in queuing theory models, or the high computational cost of network simulators. Recent advances in machine learning, such as Graph Neural Networks (GNN), are enabling a new generation of network models that are data-driven and can learn complex non-linear behaviors. In this paper, we present RouteNet-Fermi, a custom GNN model that shares the same goals as Queuing Theory, while being considerably more accurate in the presence of realistic traffic models. The proposed model predicts accurately the delay, jitter, and packet loss of a network. We have tested RouteNet-Fermi in networks of increasing size (up to 300 nodes), including samples with mixed traffic profiles — e.g., with complex non-Markovian models — and arbitrary routing and queue scheduling configurations. Our experimental results show that RouteNet-Fermi achieves similar accuracy as computationally-expensive packet-level simulators and scales accurately to larger networks. Our model produces delay estimates with a mean relative error of 6.24% when applied to a test dataset of 1,000 samples, including network topologies one order of magnitude larger than those seen during training. Finally, we have also evaluated RouteNet-Fermi with measurements from a physical testbed and packet traces from a real-life network.

Emerging applications such as the metaverse, telesurgery or cloud computing require increasingly complex operational demands on networks (e.g., ultra-reliable low latency). Likewise, the ever-faster traffic dynamics will demand network control mechanisms that can operate at short timescales (e.g., sub-minute). In this context, Traffic Engineering (TE) is a key component to efficiently control network traffic according to some performance goals (e.g., minimize network congestion).This paper presents Routing By Backprop (RBB), a novel TE method based on Graph Neural Networks (GNN) and differentiable programming. Thanks to its internal GNN model, RBB builds an end-to-end differentiable function of the target TE problem (MinMaxLoad). This enables fast TE optimization via gradient descent. In our evaluation, we show the potential of RBB to optimize OSPF-based routing (≈25% of improvement with respect to default OSPF configurations). Moreover, we test the potential of RBB as an initializer of computationally-intensive TE solvers. The experimental results show promising prospects for accelerating this type of solvers and achieving efficient online TE optimization.

Network modeling is an essential tool for network planning and management. It allows network administrators to explore the performance of new protocols, mechanisms, or optimal configurations without the need for testing them in real production networks. Recently, Graph Neural Networks (GNNs) have emerged as a practical solution to produce network models that can learn and extract complex patterns from real data without making any assumptions. However, state-of-the-art GNN-based network models only work with traffic matrices, this is a very coarse and simplified representation of network traffic. Although this assumption has shown to work well in certain use-cases, it is a limiting factor because, in practice, networks operate with flows. In this paper, we present FlowDT a new DL-based solution designed to model computer networks at the fine-grained flow level. In our evaluation, we show how FlowDT can accurately predict relevant per-flow performance metrics with an error of 3.5%, FlowDT’s performance is also benchmarked against vanilla DL models as well as with Queuing Theory.

Network modeling is a fundamental tool in network research, design, and operation. Arguably the most popular method for modeling is Queuing Theory (QT). Its main limitation is that it imposes strong assumptions on the packet arrival process, which typically do not hold in real networks. In the field of Deep Learning, Graph Neural Networks (GNN) have emerged as a new technique to build data-driven models that can learn complex and non-linear behavior. In this paper, we present RouteNet-Erlang, a pioneering GNN architecture designed to model computer networks. RouteNet-Erlang supports complex traffic models, multi-queue scheduling policies, routing policies and can provide accurate estimates in networks not seen in the training phase. We benchmark RouteNet-Erlang against a state-of-the-art QT model, and our results show that it outperforms QT in all the network scenarios.

The recent growth of emergent network applications (e.g., satellite networks, vehicular networks) is increasing the complexity of managing modern communication networks. As a result, the community proposed the Digital Twin Networks (DTN) as a key enabler of efficient network management. Network operators can leverage the DTN to perform different optimization tasks (e.g., Traffic Engineering, Network Planning).Deep Reinforcement Learning (DRL) showed a high performance when applied to solve network optimization problems. In the context of DTN, DRL can be leveraged to solve optimization problems without directly impacting the real-world network behavior. However, DRL scales poorly with the problem size and complexity. In this paper, we explore the use of Evolutionary Strategies (ES) to train DRL agents for solving a routing optimization problem. The experimental results show that ES achieved a training time speed-up of 128 and 6 for the NSFNET and GEANT2 topologies respectively.

Wide Area Networks (WAN) are a key infrastructure in today’s society. During the last years, WANs have seen a considerable increase in network’s traffic and network applications, imposing new requirements on existing network technologies (e.g., low latency and high throughput). Consequently, Internet Service Providers (ISP) are under pressure to ensure the customer’s Quality of Service and fulfill Service Level Agreements. Network operators leverage Traffic Engineering (TE) techniques to efficiently manage the network’s resources. However, WAN’s traffic can drastically change during time and the connectivity can be affected due to external factors (e.g., link failures). Therefore, TE solutions must be able to adapt to dynamic scenarios in real-time. In this paper we propose Enero, an efficient real-time TE solution based on a two-stage optimization process. In the first one, Enero leverages Deep Reinforcement Learning (DRL) to optimize the routing configuration by generating a long-term TE strategy. To enable efficient operation over dynamic network scenarios (e.g., when link failures occur), we integrated a Graph Neural Network into the DRL agent. In the second stage, Enero uses a Local Search algorithm to improve DRL’s solution without adding computational overhead to the optimization process. The experimental results indicate that Enero is able to operate in real-world dynamic network topologies in 4.5 s on average for topologies up to 100 links.

Network modeling is a critical component of Quality of Service (QoS) optimization. Current networks implement Service Level Agreements (SLA) by careful configuration of both routing and queue scheduling policies. However, existing modeling techniques are not able to produce accurate estimates of relevant SLA metrics, such as delay or jitter, in networks with complex QoS-aware queueing policies (e.g., strict priority, Weighted Fair Queueing, Deficit Round Robin). Recently, Graph Neural Networks (GNNs) have become a powerful tool to model networks since they are specifically designed to work with graph-structured data. In this paper, we propose a GNN-based network model able to understand the complex relationship between (i) the queueing policy (scheduling algorithm and queue sizes), (ii) the network topology, (iii) the routing configuration, and (iv) the input traffic matrix. We call our model TwinNet, a Digital Twin that can accurately estimate relevant SLA metrics for network optimization. TwinNet can generalize to its input parameters, operating successfully in topologies, routing, and queueing configurations never seen during training. We evaluate TwinNet over a wide variety of scenarios with synthetic traffic and validate it with real traffic traces. Our results show that TwinNet can provide accurate estimates of end-to-end path delays in 106 unseen real-world topologies, under different queuing configurations with a Mean Absolute Percentage Error (MAPE) of 3.8%, as well as a MAPE of 6.3% error when evaluated with a real testbed. We also showcase the potential of the proposed model for SLA-driven network optimization and what-if analysis.

The digital transformation is pushing the existing network technologies towards new horizons, enabling new applications (e.g., vehicular networks). As a result, the networking community has seen a noticeable increase in the requirements of emerging network applications. One main open challenge is the need to accommodate control systems to highly dynamic network scenarios. Nowadays, existing network optimization technologies do not meet the needed requirements to effectively operate in real time. Some of them are based on hand-crafted heuristics with limited performance and adaptability, while some technologies use optimizers which are often too time-consuming. Recent advances in Deep Reinforcement Learning (DRL) have shown a dramatic improvement in decision-making and automated control problems. Consequently, DRL represents a promising technique to efficiently solve a variety of relevant network optimization problems, such as online routing. In this paper, we explore the use of state-of-the-art DRL technologies for real-time routing optimization and outline some relevant open challenges to achieve production-ready DRL-based solutions.

WireGuard is a VPN protocol that has gained significant interest recently. Its main advantages are: (i) simple configuration (via pre-shared SSH-like public keys), (ii) mobility support, (iii) reduced codebase to ease auditing, and (iv) Linux kernel implementation that yields high performance. However, WireGuard (intentionally) lacks a control plane. This means that each peer in a WireGuard network has to be manually configured with the other peers’ public key and IP addresses, or by other means. In this paper we present an architecture based on a centralized server to automatically distribute this information. In a nutshell, first we manually establish a WireGuard tunnel to the centralized server, and ask all the peers to store their public keys and IP addresses in it. Then, WireGuard peers use this secure channel to retrieve on-demand the information for the peers they want to communicate to. Our design strives to: (i) offer a key distribution scheme simpler than PKI-based ones, (ii) limit the number of public keys sent to the peers, and (iii) reduce tunnel establishment latency by means of an UDP-based protocol. We argue that such automation can help the deployment in enterprise or ISP scenarios. We also describe in detail our implementation and analyze several performance metrics. Finally, we discuss possible improvements regarding several shortcomings we found during implementation.

Traffic Engineering (TE) is a basic building block of the Internet. In this paper, we analyze whether modern Machine Learning (ML) methods are ready to be used for TE optimization. We address this open question through a comparative analysis between the state of the art in ML and the state of the art in TE. To this end, we first present a novel distributed system for TE that leverages the latest advancements in ML. Our system implements a novel architecture that combines Multi-Agent Reinforcement Learning (MARL) and Graph Neural Networks (GNN) to minimize network congestion. In our evaluation, we compare our MARL+GNN system with DEFO, a network optimizer based on Constraint Programming that represents the state of the art in TE. Our experimental results show that the proposed MARL+GNN solution achieves equivalent performance to DEFO in a wide variety of network scenarios including three real-world network topologies. At the same time, we show that MARL+GNN can achieve significant reductions in execution time (from the scale of minutes with DEFO to a few seconds with our solution).

Enterprise networks, over the years, have become more and more complex trying to keep up with new requirements that challenge traditional solutions. Just to mention one out of many possible examples, technologies such as Virtual LANs (VLANs) struggle to address the scalability and operational requirements introduced by Internet of Things (IoT) use cases. To keep up with these challenges we have identified four main requirements that are common across modern enterprise networks: (i) scalable mobility, (ii) endpoint segmentation, (iii) simplified administration, and (iv) resource optimization. To address these challenges we designed SDA (Software Defined Access), a solution for modern enterprise networks that leverages Software-Defined Networking (SDN) and other state of the art techniques. In this paper we present the design, implementation and evaluation of SDA. Specifically, SDA: (i) leverages a combination of an overlay approach with an event-driven protocol (LISP) to dynamically adapt to traffic and mobility patterns while preserving resources, and (ii) enforces policies to groups of endpoints for scalable segmentation with low operational burden. We present our experience with deploying SDA in two real-life scenarios: an enterprise campus, and a large warehouse with mobile robots. Our evaluation shows that SDA, when compared with traditional enterprise networks, can (i) reduce overall data plane forwarding state up to 70% thanks to a reactive protocol using a centralized routing server, and (ii) reduce by an order of magnitude the handover delays in scenarios of massive mobility with respect to other approaches. Finally, we discuss lessons learned while deploying and operating SDA, and possible optimizations regarding the use of an event-driven protocol and group-based segmentation.

In the inter-domain routing infrastructure, a route leak is defined as a violation of the routing policy agreed between two Autonomous Systems (AS). Route leaks have resulted in large-scale outages on the Internet, taking down several services. Although route leaks seem a simple problem, the solution is complex because: (i) ASes consider -partially- routing policy private, (ii) lack of a formal and standard language to express routing policy and (iii) BGP lacks adequate cryptographic-based security. In this paper, we present an experimental analysis of a distributed ledger-based architecture that provides a solution to route leaks. Specifically, the routing policy is unambiguously expressed using a formal language, that is then stored in a blockchain. This decentralized architecture allows private policies and interfaces seamlessly with the current BGP infrastructure, requiring no changes to routers. We build a prototype to evaluate our proposed architecture using Hyperledger, we analyze its performance using a real-world BGP dataset. Our results show that our architecture scales linearly with relevant metrics. Additionally, we validate the architecture preventing an artificially introduced route leak in a realistic 10 AS topology.

Network modeling is a key enabler to achieve efficient network operation in future self-driving Software-Defined Networks. However, we still lack functional network models able to produce accurate predictions of Key Performance Indicators (KPI) such as delay, jitter or loss at limited cost. In this paper we propose RouteNet, a novel network model based on Graph Neural Network (GNN) that is able to understand the complex relationship between topology, routing, and input traffic to produce accurate estimates of the per-source/destination per-packet delay distribution and loss. RouteNet leverages the ability of GNNs to learn and model graph-structured information and as a result, our model is able to generalize over arbitrary topologies, routing schemes and traffic intensity. In our evaluation, we show that RouteNet is able to predict accurately the delay distribution (mean delay and jitter) and loss even in topologies, routing and traffic unseen in the training (worst case MRE = 15.4%). Also, we present several use cases where we leverage the KPI predictions of our GNN model to achieve efficient routing optimization and network planning.

Ubiquitous multicore processors nowadays rely on an integrated packet-switched network for cores to exchange and share data. The performance of these intra-chip networks is a key determinant of the processor speed and, at high core counts, becomes an important bottleneck due to scalability issues. To address this, several works propose the use of mm-wave wireless interconnects for intra-chip communication and demonstrate that, thanks to their low-latency broadcast and system-level flexibility, this new paradigm could break the scalability barriers of current multicore architectures. However, these same works assume 10+ Gb/s speeds and efficiencies close to 1 pJ/bit without a proper understanding of the wireless intra-chip channel. This paper first demonstrates that such assumptions do not hold in the context of commercial chips by evaluating losses and dispersion in them. Then, we leverage the system's monolithic nature to engineer the channel, this is, to optimize its frequency response by carefully choosing the chip package dimensions. Finally, we exploit the static nature of the channel to adapt to it, pushing efficiency-speed limits with simple tweaks at the physical layer. Our methods reduce the path loss and delay spread of a simulated commercial chip by 47 dB and $7.3\times $ , respectively, enabling intra-chip wireless communications over 10 Gb/s and only 3.1 dB away from the dispersion-free case.

The specification and enforcement of network-wide policies in a single administrative domain is common in today's networks and considered as already resolved. However, this is not the case for multi-administrative domains, e.g. among different enterprises. In such situation, new problems arise that challenge classical solutions such as PKIs, which suffer from scalability and granularity concerns. In this paper, we present an extension to Group-Based Policy -a widely used network policy languagefor the aforementioned scenario. To do so, we take advantage of a permissioned blockchain implementation (Hyperledger Fabric) to distribute access control policies in a secure and auditable manner, preserving at the same time the independence of each organization. Network administrators specify polices that are rendered into blockchain transactions. A LISP control plane (RFC 6830) allows routers performing the access control to query the blockchain for authorizations. We have implemented an end-to-end experimental prototype and evaluated it in terms of scalability and network latency.

Recent advances in Deep Reinforcement Learning (DRL) techniques are providing a dramatic improvement in decision-making and automated control problems. As a result, we are witnessing a growing number of research works that are proposing ways of applying DRL techniques to network-related problems such as routing. However, such proposals failed to achieve good results, often under-performing traditional routing techniques. We argue that successfully applying DRL-based techniques to networking requires finding good representations of the network parameters: feature engineering. DRL agents need to represent both the state (e.g., link utilization) and the action space (e.g., changes to the routing policy). In this paper, we show that existing approaches use straightforward representations that lead to poor performance. We propose a novel representation of the state and action that outperforms existing ones and that is flexible enough to be applied to many networking use-cases. We test our representation in two different scenarios: (i) routing in optical transport networks and (ii) QoS-aware routing in IP networks. Our results show that the DRL agent achieves significantly better performance compared to existing state/action representations.

Wireless Network-on-Chip (WNoC) has emerged as a promising alternative to conventional interconnect fabrics at the chip scale. Since WNoCs may imply the close integration of antennas, one of the salient challenges in this scenario is the management of coupling and interferences. This paper, instead of combating coupling, aims to take advantage of close integration to create arrays within a WNoC. The proposed solution is opportunistic as it attempts to exploit the existing infrastructure to build a simple reconfigurable beamforming scheme. Full-wave simulations show that, despite the effects of lossy silicon and nearby antennas, within-package arrays achieve moderate gains and beamwidths below 90°, a figure which is already relevant in the multiprocessor context.

Metasurfaces, the two-dimensional counterpart of metamaterials, have caught great attention thanks to their powerful control over electromagnetic waves. Recent times have seen the emergence of a variety of metasurfaces exhibiting not only countless functionalities, but also a reconfigurable or even programmable response. Reconfigurability, however, entails the integration of tuning and control circuits within the metasurface structure and, as this new paradigm moves forward, new reliability challenges may arise. This paper examines, for the first time, the reliability problem in programmable metamaterials by proposing an error model and a general methodology for error analysis. To derive the error model, the causes and potential impact of faults are identified and discussed qualitatively. The methodology is presented and instantiated for beam steering, which constitutes a relevant example for programmable metasurfaces. Results show that performance degradation depends on the type of error and its spatial distribution and that, in beam steering, error rates over 10% can still be considered acceptable.

Metasurfaces, ultrathin and planar electromagnetic devices with sub-wavelength unit cells, have recently attracted enormous attention for their powerful control over electromagnetic waves, from microwave to visible range. With tunability added to the unit cells, the programmable metasurfaces enable us to benefit from multiple unique functionalities controlled by external stimuli. In this review paper, we will discuss the recent progress in the field of programmable metasurfaces and elaborate on different approaches to realize them, with the tunability from global aspects, to local aspects, and to software-defined metasurfaces.

Tunable metasurfaces are ultra-thin, artificial electromagnetic components that provide engineered and externally adjustable functionalities. The programmable metasurface, the HyperSurFace, concept consists in integrating controllers within the metasurface that interact locally and communicate globally to obtain a given electromagnetic behaviour. Here, we address the design constraints introduced by both functions accommodated by the programmable metasurface, i.e., the desired metasurface operation and the unit cells wireless communication enabling such programmable functionality. The design process for meeting both sets of specifications is thoroughly discussed. Two scenarios for wireless intercell communication are proposed. The first exploits the metasurface layer itself, while the second employs a dedicated communication layer beneath the metasurface backplane. Complexity and performance trade-offs are highlighted.

Wireless Network-on-Chip (WNoC) appears as a promising alternative to conventional interconnect fabrics for chip-scale communications. The WNoC paradigm has been extensively analyzed from the physical, network and architecture perspectives assuming mmWave band operation. However, there has not been a comprehensive study at this band for realistic chip packages and, thus, the characteristics of such wireless channel remain not fully understood. This work addresses this issue by accurately modeling a flip-chip package and investigating the wave propagation inside it. Through parametric studies, a locally optimal configuration for 60 GHz WNoC is obtained, showing that chip-wide attenuation below 32.6 dB could be achieved with standard processes. Finally, the applicability of the methodology is discussed for higher bands and other integrated environments such as a Software-Defined Metamaterial (SDM).

Wireless Network-on-Chip (WNoC) appears as a promising alternative to conventional interconnect fabrics for chip-scale communications. WNoC takes advantage of an overlaid network composed by a set of millimeter-wave antennas to reduce latency and increase throughput in the communication between cores. Similarly, wireless inter-chip communication has been also proposed to improve the information transfer between processors, memory, and accelerators in multi-chip settings. However, the wireless channel remains largely unknown in both scenarios, especially in the presence of realistic chip packages. This work addresses the issue by accurately modeling flip-chip packages and investigating the propagation both its interior and its surroundings. Through parametric studies, package configurations that minimize path loss are obtained and the trade-offs observed when applying such optimizations are discussed. Single-chip and multi-chip architectures are compared in terms of the path loss exponent, confirming that the amount of bulk silicon found in the pathway between transmitter and receiver is the main determinant of losses.

Graphene is enabling a plethora of applications in a wide range of fields due to its unique electrical, mechanical, and optical properties. In the realm of wireless communications, graphene shows great promise for the implementation of miniaturized and tunable antennas in the terahertz band. These unique advantages open the door to new reconfigurable antenna structures which, in turn, enable novel communication protocols at different levels of the stack. This paper explores both aspects by, first, presenting a terahertz Yagi-Uda-like antenna concept that achieves reconfiguration both in frequency and beam direction simultaneously. Then, a programmable antenna controller design is proposed to expose the reconfigurability to the PHY and MAC layers, and several examples of its applicability are given. The performance and cost of the proposed scheme is evaluated through full-wave simulations and comparative analysis, demonstrating reconfigurability at nanosecond granularity with overheads below 0.02 mm2and 0.2 mW.

On-chip communication remains as a key research issue at the gates of the manycore era. In response to this, novel interconnect technologies have opened the door to new Network-on-Chip (NoC) solutions towards greater scalability and architectural flexibility. Particularly, wireless on-chip communication has garnered considerable attention due to its inherent broadcast capabilities, low latency, and system-level simplicity. This work presents ORTHONOC, a wired-wireless architecture that differs from existing proposals in that both network planes are decoupled and driven by traffic steering policies enforced at the network interfaces. With these and other design decisions, ORTHONOC seeks to emphasize the ordered broadcast advantage offered by the wireless technology. The performance and cost of ORTHONOC are first explored using synthetic traffic, showing substantial improvements with respect to other wired-wireless designs with a similar number of antennas. Then, the applicability of ORTHONOC in the multiprocessor scenario is demonstrated through the evaluation of a simple architecture that implements fast synchronization via ordered broadcast transmissions. Simulations reveal significant execution time speedups and communication energy savings for 64-threaded benchmarks, proving that the value of ORTHONOC goes beyond simply improving the performance of the on-chip interconnect.

In shared-memory multiprocessing, fine-grain synchronization is challenging because it requires frequent communication. As technology scaling delivers larger manycore chips, such pattern is expected to remain costly to support. In this paper, we propose to address this challenge by using on-chip wireless communication. Each core has a transceiver and an antenna to communicate with all the other cores. This environment supports very low latency global communication. Our architecture, called WiSync, uses a per-core Broadcast Memory (BM). When a core writes to its BM, all the other 100+ BMs get updated in less than 10 processor cycles. We also use a second wireless channel with cheaper transfers to execute barriers efficiently. WiSync supports multiprogramming, virtual memory, and context switching. Our evaluation with simulations of 128-threaded kernels and 64-threaded applications shows that WiSync speeds-up synchronization substantially. Compared to using advanced conventional synchronization, WiSync attains an average speedup of nearly one order of magnitude for the kernels, and 1.12 for PARSEC and SPLASH-2.

Wireless RF power transmission from dedicated Energy Transmitters (ETs) is emerging as a promising approach to enable battery-less wireless networked sensor systems. However, when data communication and RF energy recharging occur in-band, sharing the RF medium and devoting separate access times for both operations raises architectural and protocol level challenges. This paper proposes a novel method of concurrent transmission of data and energy to solve this problem, allowing ETs to transmit energy and sensors to transmit data in the same band synchronously. Our key idea concerns devising a physical layer modulation scheme that allows the data transmitting node to introduce variations in the envelope of the energy signal at the intended recipient. We implemented a proof-of-concept receiver, modeled and validated through extensive experimentation. We then propose a new physical layer mechanism for guaranteed successful delivery of information in a point-to-point link. Quantitative results demonstrate the feasibility of joint energy-data transfer, along with its associated benefits and tradeoffs.

Simultaneous Wireless Information and Power Transmission (SWIPT) has been proposed as a feasible solution to enable joint power and data transfer for the nodes of a battery-less wireless networked sensor system. Different from existing approaches, where the incident energy is split between decoding and harvesting blocks at the receiver chain, this paper describes the design and implementation of an all-digital receiver circuit. We leverage the internal control signals of the circuit, targeting ultra-low power consumption, low bit-rate applications in SWIPT. A proof-of-concept receiver is modeled, implemented using off-the-shelf hardware, and validated through extensive experiments. Quantitative results demonstrate the benefits of this joint energy-data reception approach through a single receiver chain, offering bit-rates of 400 bps.

Multisource energy harvesters are a promising, robust alternative to power the future Internet of Nano Things (IoNT), since the network elements can maintain their operation regardless of the fact that one of its energy sources might be temporarily unavailable. Interestingly, and less explored, when the energy availability of the energy sources present large temporal variations, combining multiple energy sources reduce the overall sparsity. As a result, the performance of a multiple energy harvester powered device is significantly better compared to a single energy source even if they harvest the same amount of energy. In this context, a framework to model and characterize the area for multiple source energy harvesting (EH) powered systems is proposed. This framework takes advantage of this improvement in performance to provide the optimal amount of energy harvesters, the requirements of each energy harvester, and the required energy buffer capacity, such that the overall area or volume is minimized. On top of these results, self-tunable energy harvesters are explored as a solution and compared to multisource EH platforms. As the results show, by conducting a joint design of the energy harvesters and the energy buffer, the overall area or volume of an EH powered device can be significantly reduced.

Concerns regarding the scalability of the interdomain routing have encouraged researchers to start elaborating a more robust Internet architecture. While consensus on the exact form of the solution is yet to be found, the need for a semantic decoupling of a node's location and identity is generally accepted as a promising way forward. However, this typically requires the use of caches that store temporal bindings between the two namespaces, to avoid hampering router packet forwarding speeds. In this article, we propose a methodology for an analytical analysis of cache performance that relies on the working-set theory. We first identify the conditions that network traffic must comply with for the theory to be applicable and then develop a model that predicts average cache miss rates relying on easily measurable traffic parameters. We validate the result by emulation, using real packet traces collected at the egress points of a campus and an academic network. To prove its versatility, we extend the model to consider cache polluting user traffic and observe that simple, low intensity attacks drastically reduce performance, whereby manufacturers should either overprovision router memory or implement more complex cache eviction policies.

Networks-on-Chip (NoCs) are currently the paradigm of choice to interconnect the cores of a chip multiprocessor. However, conventional NoCs may not suffice to fulfill the on-chip communication requirements of processors with hundreds or thousands of cores. The main reason is that the performance of such networks drops as the number of cores grows, especially in the presence of multicast and broadcast traffic. This not only limits the scalability of current multiprocessor architectures, but also sets a performance wall that prevents the development of architectures that generate moderate-to-high levels of multicast. In this paper, a Wireless Network-on-Chip (WNoC) where all cores share a single broadband channel is presented. Such design is conceived to provide low latency and ordered delivery for multicast/broadcast traffic, in an attempt to complement a wireline NoC that will transport the rest of communication flows. To assess the feasibility of this approach, the network performance of WNoC is analyzed as a function of the system size and the channel capacity, and then compared to that of wireline NoCs with embedded multicast support. Based on this evaluation, preliminary results on the potential performance of the proposed hybrid scheme are provided, together with guidelines for the design of MAC protocols for WNoC.

Wireless RF power transmission promises battery-less, resilient, and perpetual wireless sensor networks. Through the action of controllable Energy Transmitters (ETs) that operate at-a- distance, the sensors can be re-charged by harvesting the radiated RF energy. However, both the charging rate and effective charging range of the ETs are limited, and thus multiple ETs are required to cover large areas. While this action increases the amount of wireless energy injected into the network, there are certain areas where the RF energy combines destructively. To address this problem, we propose a duty-cycled random- phase multiple access (DRAMA). Non-intuitively, our approach relies on deliberately generating random interferences, both destructive and constructive, at the destination nodes. We demonstrate that DRAMA optimizes the power conversion efficiency, and the total amount of energy harvested. Through real-testbed experiments, we prove that our proposed scheme provides significant advantages over the current state of the art in our considered scenario, as it requires up to 70% less input RF power to recharge the energy buffer of the sensor in the same time.

The current Internet architecture was not designed to easily accommodate mobility because IP addresses are used both to identify and locate hosts. The Locator/Identifier Separation Protocol (LISP) decouples them by considering two types of addresses: Endpoint IDentifiers (EIDs) to identify hosts, and Routing LOCators (RLOCs) that identify network attachment points. LISP, with such separation in place, also offers native mobility. In this context, LISP-MN is a particular case of LISP and specifies mobility. Mobility protocols have an inherent issue with privacy since some users may not want to reveal their location or their identity. In this paper, we present an overview of LISP-MN and propose solutions to enable privacy, both in terms of location and identity.

The cost of broadcast has been constraining the design of manycore processors and of the algorithms that run upon them. However, as on-chip RF technologies allow the design of small-footprint and high-bandwidth antennas and transceivers, native low-latency (a few clock cycles) and low-power (a few pJ/bit) broadcast support through wireless communication can be envisaged. In this paper, we analyze the main networking design aspects and challenges of Broadcast-oriented Wireless Network-on-Chip (BoWNoC), which are basically reduced to the development of Medium Access Control (MAC) protocols able to handle hundreds of cores. We evaluate the broadcast performance and scalability of different MAC designs, to then discuss the impact that the proposed paradigm could exert on the performance, scalability and programmability of future manycore architectures, programming models and parallel algorithms.

The scalability of Network-on-Chip (NoC) designs has become a rising concern as we enter the many core era. Multicast support represents a particular yet relevant case within this context and has been the focus of different research efforts, mainly due to the poor performance of NoCs in the presence of this increasingly important type of traffic. However, most of the proposed schemes have been evaluated using synthetic traffic or within a full system, which is either unrealistic or costly. While traffic models would allow to better assess their performance, existing proposals do not distinguish between unicast and multicast flows and often are bound to a given number of cores. In this paper, a trace-based multicast traffic characterization is presented with the aim to provide guidelines for the modeling of multicast communications in many core settings. To this end, the scaling trends of aspects such as the multicast traffic intensity or the spatiotemporal injection distribution are analyzed. The novelty of this work resides both on its scalability-oriented approach and on the use of correlation metrics to evaluate potential prediction opportunities.

Graphene is enabling a plethora of applications in a wide range of fields due to its unique electrical, mechanical, and optical properties. Among them, graphene-based plasmonic miniaturized antennas (or shortly named, graphennas) are garnering growing interest in the field of communications. In light of their reduced size, in the micrometric range, and an expected radiation frequency of a few terahertz, graphennas offer means for the implementation of ultra-short-range wireless communications. Motivated by their high radiation frequency and potentially wideband nature, this paper presents a methodology for the time-domain characterization and evaluation of graphennas. The proposed framework is highly vertical, as it aims to build a bridge between technological aspects, antenna design, and communications. Using this approach, qualitative and quantitative analyses of a particular case of graphenna are carried out as a function of two critical design parameters, namely, chemical potential and carrier mobility. The results are then compared to the performance of equivalent metallic antennas. Finally, the suitability of graphennas for ultra-short-range communications is briefly discussed.

Graphene is a promising material which has been proposed to build graphene plasmonic miniaturized antennas, or graphennas, which show excellent conditions for the propagation of Surface Plasmon Polariton (SPP) waves in the terahertz band. Due to their small size of just a few micrometers, graphennas allow the implementation of wireless communications among nanosystems, leading to a novel paradigm known as Graphene-enabled Wireless Communications (GWC). In this paper, an analytical framework is developed to evaluate how the channel capacity of a GWC system scales as its dimensions shrink. In particular, we study how the unique propagation of SPP waves in graphennas will impact the channel capacity. Next, we further compare these results with respect to the case when metallic antennas are used, in which these plasmonic effects do not appear. In addition, asymptotic expressions for the channel capacity are derived in the limit when the system dimensions tend to zero. In this scenario, necessary conditions to ensure the feasibility of GWC networks are found. Finally, using these conditions, new guidelines are derived to explore the scalability of various parameters, such as transmission range and transmitted power. These results may be helpful for designers of future GWC systems and networks.

Networks-on-chip (NoCs) are emerging as the way to interconnect the processing cores and the memory within a chip multiprocessor. As recent years have seen a significant increase in the number of cores per chip, it is crucial to guarantee the scalability of NoCs in order to avoid communication to become the next performance bottleneck in multicore processors. Among other alternatives, the concept of wireless network-on-chip (WNoC) has been proposed, wherein on-chip antennas would provide native broadcast capabilities leading to enhanced network performance. Since energy consumption and chip area are the two primary constraints, this work is aimed to explore the area and energy implications of scaling a WNoC in terms of: 1) the number of cores within the chip, and 2) the capacity of each link in the network. To this end, an integral design space exploration is performed, covering implementation aspects (area and energy), communication aspects (link capacity), and network-level considerations (number of cores and network architecture). The study is entirely based upon analytical models, which will allow to benchmark the WNoC scalability against a baseline NoC. Eventually, this investigation will provide qualitative and quantitative guidelines for the design of future transceivers for wireless on-chip communication.

Multi-source energy harvesters are gaining interest as a robust alternative to power wireless sensors, since the sensor node can maintain its operation regardless of the fact that one of its energy sources might be temporarily unavailable. Interestingly, and less explored, when the energy availability of the energy sources present large temporal variations, combining multiple energy sources reduce the overall sparsity. As a result, the performance of a multiple energy harvester powered sensor node is significantly better compared to a single energy source which harvests the same amount of energy. In this context, a circuit area optimization framework for multiple source energy harvesting powered systems is proposed. This framework takes advantage of this improvement in performance to provide the optimal amount of energy harvesters, the requirements of each energy harvester and the required energy buffer capacity, such that the overall area or volume is minimized. As the results show, by conducting a joint design of the energy harvesters and the energy buffer, the overall area or volume of a sensor node can be significantly reduced.

Multicast on-chip communications are expected to become an important concern as the number of cores grows and we reach the manycore era. The increasing importance such traffic flows directly contrasts with the diminishing multicast performance of current Network-on-Chip (NoC) designs, and has lead to a surge of research works that seek to improve on-chip multicast support. Within this context, one-to-many traffic models may become useful for the early-stage design and evaluation of these proposals. However, existing models do not distinguish between unicast and multicast flows and often do not consider different multiprocessor sizes. To bridge this gap, a multicast scalability analysis is presented, aiming to provide tools for the modeling of multicast communications for NoC design and evaluation purposes.

Nanonetworks, the interconnection of nanosystems, are envisaged to greatly expand the applications of nanotechnology in the biomedical, environmental and industrial fields. However, it is still not clear how these nanosystems will communicate among them. This work considers a scenario of Diffusion-based Molecular Communication (DMC), a promising paradigm that has been recently proposed to implement nanonetworks. In a DMC network, transmitters encode information by the emission of molecules which diffuse throughout the medium, eventually reaching the receiver locations. In this scenario, a pulse-based modulation scheme is proposed and two techniques for the detection of the molecular pulses, namely, amplitude detection and energy detection, are compared. In order to evaluate the performance of DMC using both detection schemes, the most important communication metrics in each case are identified. Their analytical expressions are obtained and validated by simulation. Finally, the scalability of the obtained performance evaluation metrics in both detection techniques is compared in order to determine their suitability to particular DMC scenarios. Energy detection is found to be more suitable when the transmission distance constitutes a bottleneck in the performance of the network, whereas amplitude detection will allow achieving a higher transmission rate in the cases where the transmission distance is not a limitation. These results provide interesting insights which may serve designers as a guide to implement future DMC networks.

The sustained growth of the global routing table is exerting an economical strain on ISPs by requiring untimely router upgrades. Notably, it has been speculated that the growth rate of router FIBs is surpassing that of its supporting technology and that the deployment of IPv6 is only to make matters worse. In this paper, we propose LISP-MPS, an architecture based on LISP, that isolates the intra-domain routing of an Autonomous System (AS) from its inter-domain routing. The resulting separation implies the decrease of backbone routing table sizes and enables an AS to control the forwarding of traffic inside its network. For a seamless, cost effective, and incremental deployment, LISP-MPS leverages iBGP to implement the LISP mapping system functionality with minimal modification to a small subset of deployed equipment. Finally, an analysis of realistic topologies shows that, despite changing how packets transit a network, the architecture does not lose resilience to failures. Moreover, we show that it can be a viable alternative to BGP/MPLS deployments due to its low implementation cost.

Nanotechnology is enabling the development of devices in a scale ranging from a few to hundreds of nanometers. The nanonetworks that result from interconnecting these devices greatly expand the possible applications, by increasing the complexity and range of operation of the system. Molecular communication is regarded as a promising way to realize this interconnection in a bio-compatible and energy efficient manner, enabling its use in biomedical applications. However, the transmission range of molecular signals is strongly limited due to the large and inherent losses of the diffusion process. In this paper, we propose the employment of Quorum Sensing so as to achieve cooperative amplification of a given signal. By means of Quorum Sensing, we aim to synchronize the course of action of a certain number of emitters, which will transmit the same signal. Under the assumption of a linear channel, such signal will be amplified and thus the transmission range will be consequently extended. Finally, we validate our proposal through simulation.

Diffusion-based molecular communication is a promising bio-inspired paradigm to implement nanonetworks, i.e., the interconnection of nanomachines. The peculiarities of the physical channel in diffusion-based molecular communication require the development of novel models, architectures and protocols for this new scenario, which need to be validated by simulation. With this purpose, we present N3Sim, a simulation framework for diffusion-based molecular communication. N3Sim allows to simulate scenarios where transmitters encode the information by releasing molecules into the medium, thus varying their local concentration. N3Sim models the movement of these molecules according to Brownian dynamics, and it also takes into account their inertia and the interactions among them. Receivers decode the information by sensing the particle concentration in their neighborhood. The benefits of N3Sim are multiple: the validation of channel models for molecular communication and the evaluation of novel modulation schemes are just a few examples.

When providing end-to-end QoS (Quality of Service), the provider of the service states to each network provider the amount of QoS traffic in the form of traffic descriptor. Nonetheless, the profile of the QoS traffic may deform by multiplexing in successive domains invalidating the traffic descriptor. Therefore, studying traffic profile deformation in the domains results crucial in QoS networks.

During the last years, some operators have expressed concerns about the continued growth of the BGP routing tables in the default-free zone. Several proposed solutions for this issue are centered around the idea of separating the network node's identifier from its topological location. Among the existing proposals, the Locator/ID Separation Protocol (LISP) has seen important development and implementation effort. LISP relies on a mapping system to provide bindings between locators and identifiers. The mapping system is a critical protocol component, and its design is still an open issue. In this paper we present a new mapping system: LISP-TREE. It is based on DNS and has a similar hierarchical topology: blocks of identifiers are assigned to the levels of the hierarchy by following the current IP address allocation policies. We also present measurement-driven simulations of mapping systems' performance, assuming a deployment of LISP in the current Internet.

Wireless technologies are rapidly evolving and the users are demanding the possibility of changing its point of attachment to the Internet (i.e. default router) without breaking the IP communications. This can be achieved by using Mobile IP or NEMO, however mobile clients must forward its data packets through its Home Agent (HA) in order to communicate with its peers. This sub-optimal route (lack of route optimization) reduces considerably the communications performance, increases the delay and the infrastructure load. Additionally, since the HA must forward all the mobile clients' data packets, it can become the bottleneck of such networks. In this paper we present the fP2P-HN architecture, a P2P-based solution that allows deploying several HAes throughout the Internet. With this architecture a mobile client can select a closer HA to its topological position in order to reduce the delay of the paths towards its peers. Furthermore it incorporates flexible HAes that, as we will see, reduce the load at these entities. The main challenge of our solution is signaling the location of the HAes in Internet. We provide an analytical model that evaluates the costs and the benefits of the fP2P-HN architecture. The model shows that the signaling grows logarithmically with the number of HAes and that the reduction is, at least, 20% (lower bound).

WLAN devices based on CSMA/CA access schemes have become a fundamental component of network deployments. In such wireless scenarios, traditional networking applications, tools, and protocols, with their built-in measurement techniques, are usually run unchanged. However, their actual interaction with the dynamics of underlying wireless systems is not yet fully understood. A relevant example of such built-in techniques is bandwidth measurement. When considering WLAN environments, various preliminary studies have shown that the application of results obtained in wired setups is not straightforward. Indeed, the contention for medium sharing among multiple users inherent to CSMA/CA access schemes has remarkable consequences on the behavior of and results obtained by bandwidth measurement techniques. In this paper, we focus on evaluating the effect of CSMA/CA-based contention on active bandwidth measurement techniques. As a result, it presents the rate response curve in steady state of a system with both FIFO and CSMA/CA-based contending cross-traffic. We also find out that the distribution of access delay shows a transient regime before reaching a stationary state. The duration of such transient regime is characterized and bounded. We also show how dispersion-based measurements that use a short number of probing packets are biased measurements of the achievable throughput, the origin of this bias lying on the transient detected in the access delay of probing packets. Overall, the results presented in this paper have several consequences that are expected to influence the design of bandwidth measurement tools as well as to better understand the results obtained with them in CSMA/CA links.

Wireless technologies have rapidly evolved and are becoming ubiquitous. An increasing number of users attach to the Internet using these technologies; hence the performance of these wireless access links is a key point when considering the performance of the whole Internet. In this paper we present a measurement-based analysis of the performance of an IEEE 802.16 (WiMAX) client and an UMTS client. The measurements were carried out in a controlled laboratory. The wireless access links were loaded with traffic from a multi-point videoconferencing application and we measured three layer-3 metrics (one-way-delay, IP-delay-variation and packet loss ratio). Additionally we estimate the performance of a WiFi and Ethernet client as a reference. Our results show that Ethernet and WiFi have comparable performances. Both the WiMAX and the UMTS links exhibited an asymmetric behavior, with the uplink showing an inferior performance. We also assessed the causes of the discretization which appears in the jitter distributions of these links.

A foreseeable scenario is where on the Internet mobile IPv6 is deployed and a large percentage of the clients are mobile nodes. These mobile clients will communicate with large servers, which under the mobile IPv6's point of view, will be correspondent nodes. Usually large servers operate in servers farms with a load balancer device. Mobile clients can communicate with these servers through their home agent (a sub-optimal path) or directly by using the built-in mechanisms of mobile IPv6 route optimization. In this paper we detail an important incompatibility between the mobile IPv6's route optimization and several load balancing techniques. This means that mobile clients need to revert to the sub-optimal path when communicating with these server farms. This issue reduces considerably the communications performance increasing the delay and the infrastructure load. Moreover it may be an important drawback when considering mobile IPv6's deployment. In this paper we show which load balancing techniques are incompatible with route optimization and we propose a novel mobile entity that solves this issue for several load balancing techniques.

A foreseeable scenario is where, on the Internet, mobile IPv6 is deployed and a large percentage of the clients are mobile nodes. These mobile clients will communicate with large servers, which under the point of view of mobile IPv6 will be correspondent nodes. In mobile IPv6 if a mobile node wants to communicate with a correspondent node directly (route optimization) it must perform the return routability procedure which includes sending and receiving signaling messages, cryptographic calculations and storing a state. This procedure must be performed for each mobile client connection and for each handover. The return routability procedure would introduce a very significant load on these servers. Moreover the servers must be modified in order to support the route optimization of mobile IPv6. These issues may be a drawback for the deployment of mobile IPv6. In this paper we propose mobility agents, a centralized solution that performs route optimization on behalf the correspondent nodes. Mobility Agents reduce the deployment cost, do not require modifying the servers, they are compatible with NEMO and other mobility protocols. Moreover they allow deploying different extensions of mobile IPv6 such as optimized MIPv6 without modifying the correspondent nodes.