Network Music Performance (NMP), where musicians perform music together over the Internet, requires ultra-low delays. Achieving a sense of presence in NMP requires both aural and visual communication, preferably coupled with immersive telepresence technologies, such as holographic communication. Existing 4G networks offer neither the ultra-low latencies needed for audio and video, nor the ample bandwidth required for video and (especially) holographic communication. In the Telepresence-Enhanced Network Music Performance (TENeMP) project we are investigating and developing solutions for immersive NMP, to be tested in the 5G testbeds of the SPIRIT project. We provide an overview of the experimental testbeds of TENeMP as well as the software tools developed by the project. We then present baseline video and audio latency measurements over 4G and 5G-NSA networks, which demonstrate NMP’s need for the capabilities of 5G-SA and beyond networks.

A key enabler for the digitization of physical devices is digital twining technology. A digital twin is a virtual representation of a physical object (or a collection of physical objects) that allows their integration into cyber systems. Digital twins are usually hosted in cloud environments, which provide high availability and resilience to failures. This integration creates new opportunities and enables new capabilities, but it also raises security concerns. In this paper, we design a digital certificate management solution that allows building trust on digital twins independently of their network location. Our solution allows digital twins to securely receive certificates, which can be used to digitally sign data at the application layer. Our scheme does not depend on the certificate infrastructure used to secure the communication between end-users and the (cloud-hosted) digital twins. Our solution is feasible, realistic and resilient against key breaches, with a marginal communication overhead. Finally, our scheme automates the process of certificate issuance for digital twins, thus enabling very fast key and certificate rotation.

The InterPlanetary File System (IPFS) is a popular distributed storage system, with worldwide presence. Due to the open nature and distributed routing of IPFS, malicious users can monitor the content other users are retrieving, violating their privacy. Unlike previous privacy enhancement schemes that rely on user anonymity, hiding the identity of the requesting user behind a proxy, we hide instead the identity of the content, using only hash functions. This prevents intermediaries from detecting what users are retrieving, without relying on any trusted third parties. Our solution also supports optional content encryption, which is especially useful for caching. We present the design and implementation of our solution for IPFS and evaluate its privacy and security properties, showing that our scheme enhances privacy and prevents a range of DDoS attacks.

The Secure Named Data Sharing (SNDS) architecture is a content brokering service built on top of Named-Data Networking (NDN), leveraging its native support for multicast, multisource and caching. SNDS transparently supports IP-based content providers and consumers via a gateway that implements ETSI’s NGSI-LD data spaces API, translating HTTP requests to and from the appropriate NDN messages. To fully support the query-based NGSI-LD API, we build appropriate protocols over NDN. We present a prototype implementation of the SNDS architecture and a preliminary evaluation of its features.

Named Data Networking (NDN) is a realization of the Information-Centric Networking (ICN) paradigm, where routing is based on content identifiers rather than on network location identifiers. The routing state in NDN can grow exponentially, not only due to the huge number of content identifiers (as opposed to network addresses) but also because it is difficult to detect "fake" routing advertisements. For example, in contrast to IP-based routing, a potentially valid routing entry in NDN can be advertised from multiple network locations, making NDN susceptible to Denial-of-Service attacks at the routing layer. In this paper, we leverage Decentralized Identifiers (DIDs) to build self-verifiable "content advertisements." With our solution, any router can verify that a content advertisement originates from an "authorized" entity, without requiring any trusted third party. We implement our solution and we evaluate it in a scenario where filtering is implemented by the edge routers. We show that our solution reduces fake routing advertisements with minimal computational overhead.

Efficient congestion control is critical to the operation of MPTCP, the Multipath extension of TCP. Congestion control in such an environment primarily aims at enhancing the cumulative TCP throughput over the available paths, while preserving TCP-friendliness by fairly sharing the available bandwidth with single-path TCP flows in each path. While most existing multipath congestion control algorithms fulfill the TCP-friendliness objective in their steady state, their throughput convergence latency is high, rendering them ineffective for short-lived flows. We have proposed Normalized Multipath Congestion Control (NMCC), an MPTCP congestion control algorithm that achieves TCP-friendliness faster, by normalizing the growth of individual sub-flow throughput rather than the throughput itself. As NMCC can become unfriendly when it experiences sparse congestion events, in this paper we introduce the extended NMCC (e-NMCC) protocol that caters for TCP-friendliness upon both throughput growth and throughput reduction epochs. We analytically characterize e-NMCC in terms of TCP-friendliness and responsiveness and compare it with alternative algorithms. Finally, we assess the performance of e-NMCC through experimentation with the htsim simulator and a real Linux implementation. Our results confirm that e-NMCC accelerates throughput convergence, thus ensuring TCP-friendliness regardless of connection duration and underlying network conditions.

As end-user and edge-network devices are becoming ever more powerful, they are producing ever increasing amounts of data. Pulling all this data into the cloud for processing is impossible, not only due to its enormous volume, but also due to the stringent latency requirements of many applications. Instead, we argue that end-user and edge-network devices should collectively form edge computing swarms and complement the cloud with their storage and processing resources. This shift from centralized to edge clouds has the potential to open new horizons for application development, supporting new low-latency services and, ultimately, creating new markets for storage and processing resources. To realize this vision, we propose Named Functions at the Edge (NFE), a platform where functions can i) be identified through a routable name, ii) be requested and moved (as data objects) to process data on demand at edge nodes, iii) pull raw or anonymized data from sensors and devices, iv) securely and privately return their results to the invoker and v) compensate each party for use of their data, storage, communication or computing resources via tracking and accountability mechanisms. We use an emergency evacuation application to motivate the need for NFE and demonstrate its potential.

Low latency is a crucial requirement for demanding conferencing applications, such as Networked Music Performance (NMP), the collaboration of musicians in real time. Modern conferencing systems employ a Selective Forwarding Unit (SFU) to transparently duplicate and forward media streams between participants. Since an SFU does not process the media streams, so as to reduce delay, its latency is mainly determined by the underlying network I/O mechanism that moves packets to/from the network hardware and user space. Such mechanisms are usually based on POSIX sockets, which were not designed for high performance networking. We designed and implemented pktswitch, a minimal, socket-based SFU and measured its performance. We then modified pktswitch to employ netmap, a framework for fast packet I/O, to overcome the performance bottlenecks imposed by the socket-based design. The modified implementation handles packets in user space, with minimal kernel interaction. We describe and contrast the two implementations and then compare their performance in terms of packet processing overhead and delay. Our results show that the netmap-based implementation reduces packet processing overhead by 76% and delay by 89% compared to the socket-based implementation, thus allowing an SFU to host much higher loads (e.g. more users with more media streams and higher bit rates) without introducing delays.

Gathering information efficiently from a, possibly fragmented, sensor network presents a serious problem in disaster recovery applications. Unless a transmission control mechanism exists, a sink can be flooded with information sent by sensors, or sensor transmissions may be lost on their way to the sink. Consequently, there is a necessity for a reliable protocol that automatically and speedily adapts to losses, congestion and network changes due to sink mobility. This paper describes RT-SENMOS, a reliable transport protocol for controlling sensor transmissions based on sink-assigned rates. The sink decides how to share the available bandwidth among the sensors and also determines the reliability to be achieved in each case. Our protocol operates on top of UDP/IP, therefore it can be directly integrated into a disaster recovery application that will set its parameters depending on the situation. Moreover, as it is fully sink-controlled, it enables the use of simple and inexpensive fixed sensors, which offload all protocol intelligence to a more expensive but reusable mobile sink. We present the design of the protocol, comparing it with similar approaches, and evaluate its performance using a real implementation.

Name resolution is at the heart of Information-Centric Networking (ICN), where names are used to both identify information and/or services, and to guide routing and forwarding inside the network. The ICN focus on information, rather than hosts, raises significant concerns regarding the scalability of the required Name Resolution System (NRS), especially when considering global scale, inter-domain deployments. In the route-by-name approach to NRS construction, name resolution and the corresponding state follow the routing infrastructure of the underlying inter-domain network. The scalability of the resulting NRS is therefore strongly related to the topological and routing characteristics of the network. However, past work has largely neglected this aspect. In this paper, we present a detailed investigation and comparison of the scalability properties of two route-by-name inter-domain NRS designs, namely, DONA and CURLING. Based on both real, full-scale inter-domain topology traces and synthetic, scaled-down topologies, our work quantifies a series of important scalability-related performance aspects, including the distribution of name-resolution state across the Internet topology and the associated processing and signaling overheads. We show that by avoiding DONA's exchange of state across peering links, CURLING results in deployment costs proportional to the total number of downstream customers of each Autonomous System. This translates to a 62-fold global state size reduction, at the expense of a 2.78-fold increase in lookup processing load, making CURLING a feasible approach to ICN name resolution.

Routers in the Content-Centric Networking (CCN) architecture maintain state for all pending content requests, so as to be able to later return the corresponding content. By employing stateful forwarding, CCN supports native multicast, enhances security and enables adaptive forwarding, at the cost of excessive forwarding state that raises scalability concerns. We propose a semi-stateless forwarding scheme in which, instead of tracking each request at every on-path router, requests are tracked at every d hops. At intermediate hops, requests gather reverse path information, which is later used to deliver responses between routers using Bloom filter-based stateless forwarding. Our approach effectively reduces forwarding state, while preserving the advantages of CCN forwarding. Evaluation results over realistic ISP topologies show that our approach reduces forwarding state by 54%-70% in unicast delivery, without any bandwidth penalties, while in multicast delivery it reduces forwarding state by 34%-55% at the expense of 6%-13% in bandwidth overhead.

This short paper describes the objectives and initial results of project I-CAN: Information-Centric Future Mobile and Wireless Access Networks. I-CAN seeks to radically advance the integration of cellular and wireless access technologies by developing and evaluating architectures and procedures for future access networks based on Information-Centric Networking (ICN).

In-network packet-level caching is one of the most promising features offered by Information-centric Networking (ICN) architectures. In ICN, routers can use their queueing buffers as temporal storage units, thus allowing on-path caching by exploiting the network's storage resources. Packet-caches can be highly beneficial for content delivery, but they are also known to have three significant weaknesses: (i) packet-granularity produces huge cache indexes, (ii) Zipf-like content popularity penalizes the hit-ratio at core nodes and (iii) any discontinuity in the stored packets disrupts RTT-based congestion control. This paper presents OPC, a novel caching management strategy designed to support wire-speed in-network caching, while dealing with the above problems. OPC works at the object-level, thus reducing indexing requirements, is destined for access routers, thus avoiding the small hit-ratios of caches at core-nodes, and stores contiguous groups of packets, thus easing RTT-based congestion control.

Many proposals for the next generation of the Internet suggest moving from an end-point oriented to an information-centric oriented architecture. Many of these proposals are based on the publish/subscribe paradigm, which lends itself naturally to native multicast support, a key factor for efficient content distribution. However, the design of efficient reliable transport protocols for multicast is a largely open problem, due to the problem of feedback implosion towards the sender as group size grows. In this paper we propose a hierarchical retransmission-based error control scheme for a native publish/subscribe internetwork. We compare our protocol with similar approaches proposed for IP multicast and evaluate its performance against IP multicast with unicast-based error control.

Stateless multicast forwarding with in-packet Bloom filters (iBF) has recently been proposed as a highly scalable way for supporting a large number of multicast groups. However, iBF multicast generates redundant traffic due to false positive forwarding decisions and it also scales poorly with multicast group size. In this paper we investigate scaling iBF multicast to arbitrary multicast group sizes, by partially sacrificing the network's fully stateless operation. We propose a switched-iBF multicast scheme that places multicast forwarding state at a few network nodes, so as to minimize redundant traffic regardless of the group size. We evaluate the scheme through simulations and find that switched-iBF multicast can scale to any group size while keeping redundant traffic below 1%-4% at the (minimal) cost of placing state at no more than 0.5%-2.5% of network nodes. We also compare the state requirements of switched-iBF multicast against other multicast schemes. Our evaluation shows that switched-iBF multicast achieves a tremendous reduction of multicast state in the range of 87%-99.6%. Hence, even though the system is no more fully stateless, it remains far more scalable than other approaches.

Large, reliable and efficient storage systems are becoming increasingly important in enterprise environments. Our research in storage system design is oriented towards the exploitation of commodity hardware for building a high performance, resilient and scalable storage system. We present the design and implementation of DHTbd, a general purpose decentralized storage system where storage nodes support a distributed hash table based interface and clients are implemented as in-kernel device drivers. DHTbd, unlike most storage systems proposed to date, is implemented at the block device level of the I/O stack, a simple yet efficient design. The experimental evaluation of the proposed system demonstrates its very good I/O performance, its ability to scale to large clusters, as well as its robustness, even when massive failures occur.

Information-Centric Networking (ICN) is constantly gaining momentum within the Future Internet research community. In the PURSUIT research project we are developing a clean-slate Pub/Sub Internetworking (PSI or Ψ) approach with integrated seamless mobility support. The novel ICN mechanisms supported in Ψ, along with smartly placed in-network caches, enable the architecture to handle both mobile and fixed devices in a uniform way. This paper presents a blueprint for optimizing mobility support in Ψ without modifications to the architecture or add-on solutions. We demonstrate a micro-mobility scenario that describes the functionality of Ψ's core components in supporting mobility and then sketch our plans for future work and a proper assessment of these designs.

Peer-to-peer (P2P) protocols have been widely accepted by users and operators alike as efficient mechanisms for non real-time content distribution. It is therefore reasonable to extend these protocols to also handle more demanding applications, such as multimedia streaming. Many researchers have proposed modifications to the well known BitTorrent protocol in order to adapt it to the needs of such applications. In this paper we present findings from our experiments with three proposals for the integration of multimedia streaming into BitTorrent: the fixed-size window approach, the high-priority set approach and the stretching window approach. We evaluate these proposals under identical circumstances using our detailed packet-level BitTorrent simulator, showing that while all approaches are reasonably capable of supporting multimedia streaming, their different design choices have a pronounced effect on their streaming performance.

We present our work on a sensor-based smart system automatically trained to recognize the activities of individuals in their home. In this paper we present and analyze a method for recognizing the indoor everyday activities of a monitored individual. This method is based on the data mining technique of association rules and Allen's temporal relations. Our experimental results show that for many (but not all) activities, this method produces a recognition accuracy of nearly 100%, in contrast to other methods based on data mining classifiers. The proposed method is accurate, very flexible and adaptable to a dynamic environment such as the "Smart Home" and we believe that it deserves further attention.

In the past few years numerous P2P file-sharing and content distribution systems have been designed, implemented, and evaluated via simulations, real world measurements, and mathematical analysis. Yet, only few of them have stood the test of time and gained wide user acceptance. BitTorrent is the one that holds the lion's share among them and the reasons behind its success have been studied to a great extent with interesting results. Nevertheless, even though P2P content distribution remains one of the most active research areas, little progress has been made towards the study of the BitTorrent protocol (and its variations), in a fully controllable and realistic simulation environment. In this paper we describe and analyze a full-featured and extensible implementation of BitTorrent for the OMNeT++ simulation platform. Moreover, since we aim at realistic simulations, we present our enhancements on a popular conversion tool for practical Internet topologies, as well as our churn generator that is based on the analysis of real BitTorrent traces. Finally, we set forth the results from the evaluation of our prototype implementation regarding resource demands under different simulation scenarios.

During the past few years, large, reliable and efficient storage systems have become increasingly important in enterprise environments. Additional requirements for these environments include low installation, maintenance and administration costs. In this paper we propose a hash-based storage approach, combined with block-level operating system semantics. The experimental evaluation confirms that the proposed approach is viable and can offer a cost-effective storage solution.

Abstract This article provides an overview of the multimedia broadcast/multicast service (MBMS) for universal mobile telecommunications system (UMTS) networks. We first outline the features of UMTS networks as defined by the 3rd generation partnership project (3GPP) in order to provide a background for the discussion to follow. We then present the overall MBMS architecture, the services that it provides to the users and the differences between the broadcast and multicast options. The implementation details of MBMS are explained in terms of the modifications needed to the network, the new signaling procedures required and the impact of MBMS on the radio part of the network. We then describe how the security architecture of UMTS is used to support content protection and key distribution for MBMS groups. We evaluate the prospects of MBMS by comparing it first with IP multicast and then with DVB‐H, considered by many to be the closest competitor of MBMS. Finally, we conclude with a discussion of the technical and business challenges still faced by MBMS. Copyright © 2006 John Wiley & Sons, Ltd.

A considerable body of evidence indicates that the use of reliable link layer protocols over error prone wireless links dramatically improves the performance of Internet protocols and applications. While traditional link layer protocols set their timeout values assuming that they fully control the underlying link, some wireless networks allow multiple link layer sessions to co-exist over the same link. Since the optimal timeout values for a reliable link layer protocol depend on the available bandwidth, with dynamic link sharing such a protocol should ideally adapt its timeout values accordingly. We have thus designed an adaptive selective repeat protocol that modifies its timeout values based on the policy used by TCP. We compare the performance of Web browsing over selective repeat when using our adaptive timeout scheme with a range of parameters, against a manually tuned fixed timeout version. Our measurements show that these adaptive timeout policies outperform the fixed one, regardless of the level of contention, and that the best adaptive timeout policy in this setting is not the one used by TCP.

Although TCP and UDP application performance over wireless links may be substantially improved via link layer error recovery, different schemes are appropriate for each application class. We present a multi-service link layer architecture that simultaneously enhances the performance of diverse applications by supporting multiple error recovery mechanisms in parallel. We simulated concurrent file transfers and WWW browsing over TCP and continuous media distribution over UDP using our architecture. The results show that each application achieves similar improvements as when it operates alone over its preferred scheme, despite the parallel execution of diverse applications.

We present a comprehensive set of measurements of a 2.4 GHz DSSS wireless LAN and analyze its behavior. We examine issues such as host and interface heterogeneity, bidirectional (TCP) traffic and error modeling, that have not been previously analyzed. We uncover multiple problems with TCP and UDP performance in this system. We investigate the causes of these problems (radio hardware, device drivers, network protocols) and discuss the effectiveness of proposed improvements.

While support for IP multicasting continues to spread enabling new applications, an increasing number of hosts connects to the worldwide Internet via low bandwidth point-to-point links, such as wireline or wireless telephone lines. We discuss existing proposals for local and wide area IP multicasting and their implications for point-to-point links, identify problems with their integration in this environment, and propose alternative special purpose mechanisms to solve these problems. The main problems are overhead due to IGMP leave latency and unnecessary continuous probing of potentially power constrained hosts. Our solution is an alternative to IGMP mechanisms based on join/leave messages for tracking group membership over PtP networks. After presenting the implementation requirements of the proposed and existing mechanisms, we compare them with respect to the performance, interoperability, robustness and implementation complexity, demonstrating that our join/leave protocol is uniformly superior.