Reliable consistent communication over millimeter-wave (mmWave) channels is a challenging problem due to their sensitivity to blocking of Line of Sight connections. MmWave is a key building block in 5G and future generation cellular networks, making solutions to this problem space important. Our aim is to use predictive control to manage and simultaneously use multiple available mmWave paths to achieve reliable consistent communication (i.e., steady transmission rate with low delay) with a multipath proxy. To this end we investigate transient solutions of Markov Modulated Fluid Queue models (MMFQ), apt because the mmWave blocking has been modeled with Markovian models. We propose a combination of models that can be solved using newly proposed matrix analytic techniques in a timely enough manner for use in real-time control. This gives us a prediction of either proxy queue distributions or probabilities of reaching proxy buffer levels over a short time horizon, enabling the proxy to make preemptive path decisions to maintain a desired Quality of Service. A proof of concept simulation study demonstrates the efficacy of our proposed MMFQ-based predictive approach over either static or purely reactive control approaches.

Most Internet hosts today support multiple access technologies and network interfaces. Multipath transport protocols, like MPTCP, are being deployed (e.g., in smartphones), allowing transparent simultaneous use of multiple links. Besides providing increased resilience to link failures, multipath transports may better exploit available (aggregate) capacity across all interfaces. The safest way to ensure fairness is to assume that any subflows of a multipath end-to-end connection may share bottleneck links, but knowledge of non-shared bottlenecks could allow multipath senders to exploit more capacity without being unfair to other flows. The problem of reliably detecting the existence of (non)-shared bottlenecks is not trivial and is compounded by the fact that bottlenecks may change due to traffic dynamics. In this paper we focus on practical methods to reliably group flows that share, possibly dynamic, bottlenecks online and in a passive manner (i.e., without injecting measurement traffic). We introduce a novel dynamic clustering algorithm that we apply to update our previous shared bottleneck flow grouping (SBFG) method standardized by the IETF, based on delay statistics. We also adapt an offline SBFG method based on wavelet filters to enable it for online operation. These SBFG methods are evaluated by a simple testbed, rigorous simulation and real-world Internet experiments in a testbed comprised of multihomed hosts. Our results suggest that there is no clear winner, and selection of the “best” SBFG method will have to consider tradeoffs regarding accuracy, lag, and application requirements.

There is a growing concern that the Internet transport layer has become less adaptive to the requirements of new applications, and that further evolution has become very difficult. This is because a fundamental assumption no longer holds: it can no longer be assumed that the transport layer is only in the scope of end-hosts. The success of TCP and UDP and the ubiquity of middleboxes have led to ossification of both the network infrastructure and the API presented to applications. This has led to the development of workarounds and point solutions that fail to cover many facets of the problem. To address this issue, this paper identifies requirements for a new transport layer and then proposes a conceptual architecture that we argue is both flexible and evolvable. This new architecture requires that applications interface to the transport at a higher abstraction level, where an application can express communication preferences via a new richer API. Protocol machinery can use this information to decide which of the available transport protocols is used. By placing the protocol machinery in the transport layer, the new architecture can allow for new protocols to be deployed and enable evolution of the transport layer.

This work proposes a solution to overcome the inherent degradation of TCP performance in multi- hop wireless networks. Based on a Network Coding entity integrated between the TCP and the IP layer, we propose a mechanism to encapsulate TCP acknowledgments, which are transmitted along with other segments, thus reducing the number of frames to be sent. We assess the performance of the combination of these two techniques by means of an extensive simulation campaign, carried out using the ns-3 framework. The results show that the addition of the encapsulation scheme brings about a performance enhancement (around 15%), compared to a standard Network Coding scheme. On the other hand, the performance deteriorates when transmission errors occur.

Since it was originally proposed by Ahlswede et al. (2000), network coding (NC) has been the focus of many research efforts and proposals. In particular, NC has been put forward as a means to increase the capacity of wireless mesh networks, by exploiting the broadcast nature of wireless links. However, in spite of the fairly large number of papers on the subject of network coding, the impact of NC on TCP performance in wireless mesh networks is still not fully understood. By means of a thorough simulation study, this paper aims thus at providing insights about the interaction between TCP's congestion control mechanisms and the underlying NC done by wireless relays. In particular, we focus on the influence of the main parameters related to the coding buffer used by such relays to encode packets. Our results suggest the feasibility of devising engineering rules for setting such parameters. Another main conclusion, in line with previous preliminary findings by the authors (Hassayoun et al., 2010) is that TCP performance can indeed be improved by the use of NC, but only as long as the wireless channel quality is above a certain level.

One of the primary mechanisms to mitigate idling energy consumption and reduce communication activities in Wireless Sensor Networks (WSNs) is to put nodes to low duty cycle operation as long as possible. Keeping a small time window (the listen period) is a challenging task, and thus requires appropriate tailoring of duty cycle so as not to compromise network performance. In this work, we explore an adaptive duty cycle mechanism for scheduling a node to sleep mode using node density. A forwarding candidate assesses its vicinity and adaptively adjusts its duty cycle. We show via simulation that the adaptive duty cycle mechanism can reduce the energy usage and prolong the network lifetime for the data delivery toward the sink.

The traditional problem of end-hosts efficiently and fairly utilizing end-to-end paths becomes significantly harder when the end-hosts are multihomed. Such is the case, for instance, when an end-host has simultaneous connectivity through several service providers, or when a mobile device is simultaneously connected via both a wireless LAN and a cellular network. A multihoming-aware transport protocol, such as MPTCP or SCTP, that sends data over the multiple resulting end-to-end paths must be fair to other flows in the network while being able to maximize its own throughput. In this paper, we present Dynamic Window Coupling (DWC), a multipath congestion control mechanism that seeks to achieve both these goals. DWC uses loss and delay signals to detect shared bottlenecks, explicitly grouping and sharing congestion control across subflows on paths that have a common bottleneck, while separating congestion control for subflows on paths with distinct bottlenecks. DWC detects shifting bottlenecks in the network and responds by dynamically regrouping subflows. Simulations demonstrate that DWC detects shared bottlenecks under most network topologies and conditions that we considered, regroups subflows correctly as bottlenecks shift, aggregates throughput across distinct bottlenecks, and is fair to other TCP flows at all bottlenecks.

The 6LoWPAN layer was recently introduced as a means to enable the transmission of IPv6 datagrams over Low-power and Lossy Networks (LLNs). That layer provides schemes for fragmenting the network-layer datagrams and reassembling link-layer fragments. However, the loss of one fragment of the original datagram will lead to the loss of the latter, and eventually the retransmission of all the fragments, which is not energy-efficient. Simple Frame Forwarding and Recovery (SFFR), introduced by Thubert and Hui, is a simple end-to-end scheme for coping with MAC-layer losses in multi-hop LLNs. In this paper, we study the performance of SFFR in multi-hop wireless networks, in terms of energy consumption. Numerical results show that SFFR offers an efficient alternative to classical, hop-by-hop error-control schemes like ARQ (Automatic Repeat Request).

This paper presents an experimental study of TCPHC, a novel algorithm for compressing the Transmission Control Protocol (TCP) header to reduce its overhead in IPv6-enabled Low-power and Lossy Networks (6LoWPANs). Results show that TCPHC outperforms TCP both in low-loss and high-loss networks. In fact, TCPHC can reduce the TCP header to 6 bytes in more than 95% of the cases. Moreover, experimental results show that our TCP header compression algorithm reduces the energy consumption by up to 15%.

A number of studies show that most of current probing tools, for available bandwidth, exhibit low accuracy, and slow convergence in Home Networks. Some reasons of this behavior include the high queuing variability of probing flows, which is induced by multiple-access mechanisms, time-variant capacity links of hybrid paths, and the instability of beta probing tools. Considering these constrains of the measurement process and the limitations of probing-tools, this paper proposes using Iperf in a lightweight mode to obtain maximum throughput of non-disruptive TCP flows to estimate the available bandwidth. Our experimental results show that Iperf doubles the accuracy of the estimates compared with IGI/PTR, Pathchirp and Wbest, on representative hybrid paths. We also show that due to the low bandwidth delay products (BDP), of home network paths, we can keep low overhead using TCP flows with convenient probing/polling intervals, which reduces the intrusiveness associated to the default Iperf usage.

In IEEE 802.16 networks, bandwidth allocation is centralized at the base station (BS). For the uplink traffic, data transmission is scheduled after the BS processes bandwidth requests (BW-REQs) sent by mobile stations (MSs). In this paper we focus on the multicast/broadcast polling method, where MSs send their BW-REQs during contention periods. Contention-based polling is deemed suitable for Best-Effort TCP flows, since the latter have less stringent needs, in terms of delay and throughput, than traffic carried by other classes of service. We propose a novel analytical model for Best-Effort TCP performance over 802.16 networks. This model explicitly considers the impact of the uplink request/grant mechanism, including the influence of bandwidth-perception mismatch at the BS. The effect of MAC-layer parameters, such as the MAC queue length, is also studied. The accuracy of the model is verified by means of ns-2 simulations.

Network coding (NC) is a promising technique to improve throughput in wireless mesh networks. However, some previous studies have found that the actual performance improvements offered by NC may be much lower than the gains predicted by theory. This is especially so when the bulk of the traffic carried by the mesh network is composed of TCP flows. By means of both mathematical modeling and ns-2 simulations, we explore some issues due to random packet loss in coded mesh networks. Our results illustrate how the use of network coding may induce synchronization between TCP flows; they also suggest that, under particular conditions of random packet loss, the aggregate throughput may actually be lower when NC is used.

In 802.16 networks, bandwidth request-grant mechanisms allow managing the uplink bandwidth needs of subscriber stations. Subscriber stations send their bandwidth requests to the base station, where most of the bandwidth scheduling is centralized. After processing such requests, the base station schedules the transmission of uplink traffic. In this paper we present a preliminary investigation of some bandwidth allocation issues, related to the management of the perception of subscriber stations' bandwidth needs at the base station. We illustrate how the bandwidth perception varies depending on the policy used to manage bandwidth requests and grants. Focusing on the case of TCP Best Effort traffic, we assess by simulation the potential impact of such policies on the system's aggregate throughput.

Adapting router buffer sizing rules and TCP's congestion control to Gbit/s link speeds has been the subject of many studies and proposals. In a previous work (Hassayoun and Ros, 2008), we found that high-speed versions of TCP may be prone to strong packet-loss synchronization between flows. Given that synchronized losses may lead to oscillatory behavior in router queues, a relatively large amount of buffering may be needed to sustain an adequate TCP performance. In this paper, we are interested in evaluating the potential impact of the Random Early Detection (RED) queue management algorithm, in the context of high-speed TCP versions. In principle, under mild congestion RED should tend to reduce loss synchronization between flows. Hence, we wanted to test in particular whether RED would allow to use moderately-sized buffers, while ensuring good performance. Simulation results are used to assess the effect of RED on goodput, link utilization and fairness among high-speed flows, for a wide range of buffer sizes and several high-speed TCP variants.

The IEEE 802.16 is the standard for broadband wireless access. One promise of this technology (also called WiMAX) is to provide high-speed access to the Internet where the transmission control protocol (TCP) is the core transport protocol. In this paper, we study the impact of network asymmetry in WiMAX on TCP performance. In particular, we investigate the dependence of the aggregate throughput and goodput of TCP on factors such as frame duration, direction of flow, DL:UL ratio, modulation and coding schemes, and offered loads. We find that these factors affect TCP performance by exacerbating the network asymmetry inherent to the MAC layer.

Nowadays Internet Mobile terminals are now all equipped with 802.11 devices. Moreover, a high and increasing number of hot spots are deployed in a variety of places, such as homes, cafes and airports. A typical mobile user uses TCP-based applications to access emails, browse the web or download files from the Internet. Thus the selfish interest for a user to have a better performing version of the Transmission Control Protocol (TCP).

High Speed Downlink Packet Access (HSDPA) is an enhancement to UMTS networks that supports data rates of several Mbps, making it suitable for applications such as video streaming. Nevertheless, the shared downlink radio channel used in HSDPA is a challenging environment for such applications. In this paper, we focus on the issue of subjective video quality, and design a novel estimator of the user-perceived quality that operates in real time. We propose to integrate such estimator in User Equipments (UEs) so that it can provide regular feedback and help the UMTS resource management procedures. Then, we use it to study the impact of a recently proposed HSDPA scheduler directly on the perceived video quality. HSDPA scheduling is one of the salient points of HSDPA and is used to perform resource management (i.e., bandwidth allocation between terminals), taking into account the radio channel conditions of all users.

The VoIP suite of protocols are extended to propose a computer network capable of terrestrially storing in real time the flight data records and cockpit voice records of every commercial flight. The design process took into account the satellite access, the aviation, the black box manufacturing and the VoIP industriespsila recent developments. As a result, a system architecture development framework and a transport-signaling protocol are proposed. The economic feasibility of the proposal is finally addressed.

High speed downlink packet access (HSDPA) is a packet-based data service in UMTS networks that supports data rates of several Mbit/s, making it suitable for applications ranging from file transfer to multimedia streaming. One of the salient points of HSDPA is the use of MAC-layer scheduling to perform resource management (i.e., bandwidth allocation between terminals), taking into account the radio channel conditions of all users. Usually, additional factors like fairness between users, cell throughput or quality-of-service (QoS) parameters are also considered in the scheduling mechanism. In this paper we consider different QoS classes or user groups and study resource allocation among those user groups using HSDPA MAC-layer scheduling. We propose a normalized rate guarantee (NRG) scheduler, which is an extension of a previous QoS scheduler by Hosein (2002). When allocating the available bandwidth, the NRG scheduler takes into account the fact that QoS requirements (in terms of minimum bandwidth) may vary from one flow to the other. It tries to apportion loss rates in a fairer way during congestion. Another goal of NRG is to avoid the deterioration of QoS when best-effort load is increased.

TCP-friendly rate control (TFRC) is an end-to-end congestion control mechanism, whose goal is to provide rate control for unicast flows in IP networks. The main feature of TFRC is its ability to smoothly adapt the sending rate of a flow to network conditions, while competing for bandwidth with TCP flows in a relatively fair manner. TFRC was designed to offer a more stable sending rate than TCP on wired, best-effort networks, making it suitable for applications like multimedia streaming. This paper presents a simulation study of TFRC over UMTS networks with high-speed downlink packet access (HSDPA). We evaluate the performance of TFRC and compare it with that of TCP, under different scenarios of MAC-layer scheduling, radio conditions and background traffic. Our results show that, in this context, the rate stability of TFRC is not necessarily better than that of TCP, and that the performance of TFRC may be greatly improved by the use of an appropriate QoS scheduling policy for HSDPA.

High speed downlink packet access (HSDPA) is a packet-based data service in UMTS networks that supports data rates of several Mbit/s, making it suitable for data applications ranging from file transfer to multimedia streaming. In spite of the fairly high data rates that HSDPA offers, the shared downlink radio channel used in HSDPA is a challenging environment for delay- and loss-sensitive applications. It is relatively easy to partition resources over wired links with known link capacity. Nevertheless, such partitioning is complex for wireless links with variable capacity; due to the dynamic nature of the link, it is difficult to effectively allocate the resources (i.e., time slots) between users who may be paying different prices for different services. In this paper we study the general problem of resource partitioning for shared wireless links like HSDPA. We consider different QoS classes or user groups, and the use of MAC-layer scheduling for resource partitioning between groups and between individual users. We propose a variant of an existing HSDPA scheduler called required activity detection (RAD) scheduler. The proposal is studied using a detailed UMTS/HSDPA simulator.

Many networked applications use end-to-end congestion control protocols to avoid the occurrence of congestion collapse phenomena. However, such transport protocols may suffer from link underutilization when there are wireless links along the path: the inability to distinguish wireless losses from congestion losses often results in unnecessary throughput decreases. Since networks are becoming increasingly heterogeneous, a mix of wired and wireless links, it is important to have some means of identifying the cause of a packet loss and to adapt the response of the transport protocol. In this paper, we present an end-to-end solution to this problem and propose a novel method for wireless loss estimation in IP DiffServ networks (WLED). WLED aims at enhancing the performance of equation-based rate control protocols for multimedia applications, by means of directly estimating the wireless and congestion loss rates. We discuss the integration of WLED with a recent, equation-based rate control protocol and we evaluate WLED's performance through simulation. Our results illustrate a new benefit of using DiffServ-like differentiation mechanisms for multimedia streaming in wireless networks, like UMTS, that support quality of service (QoS).

In recent years, the notion of a service offering a degraded performance with respect to the best-effort service traditionally found in IP networks has gained acceptance among network researchers. Such a less-than-best-effort (LBE) service may he considered as another way of providing a differentiated quality of service, following A. Odlyzko's "damaged goods for the Internet" approach. In this paper we are interested in evaluating, from a pricing perspective, the implications of the two scheduling models commonly proposed for building a LBE service-namely, priority queueing and generalized processor sharing (GPS). In particular, we focus on the network operator's issue of maximizing her revenue. We wish to study, for each scheduler, how to set prices and, especially, the impact that a given queueing model may have on revenues when users are mostly sensitive to delay. Drawing on previous work by Mandjes (2003), we present analytical expressions of the revenue earned by the network operator, when a GPS scheduler is used. A comparison of optimal revenues shows that: (a) Priority Queueing is more efficient, in economic terms, than both a GPS scheduler and a simple FIFO queue, that is, a network with no service differentiation; (b) revenues are lower with a GPS scheduler than with a FIFO queue. These results may have implications both on the practical implementation of LBE services and on the Paris Metro Pricing proposal by Odlyzko (1999).