Since the definition of the bufferbloat phenomenon, several Linux kernel modules have been introduced in its TCP/IP stack, and there is a lack of experimental studies on their effects when coupled with WLAN technologies, in particular, IEEE 802.11n and IEEE 802.11ac. One essential algorithm introduced is named TCP Small Queues (TSQ) and has the role of limiting the number of packets that a TCP socket can enqueue in the stack, waiting for the physical layer to send the packets before enqueueing extra data. A second significant TCP algorithm is named TCP Pacing (TP) and regulates the pace used by the socket to enqueue packets in the stack, regulating the formation of bursts of data. These mechanisms affect the frame aggregation logic on WLAN networks and compromise the throughput-latency tradeoff of all the TCP variants. This paper presents an experimental evaluation of these techniques investigating the wireless network performance of several TCP congestion control variants under the presence of different TSQ and TP policies, modeling also their interaction.

This paper shows the inefficiency of TCP BBR in exploiting the Wi-Fi bandwidth. This limitation of BBR has been observed with both IEEE 802.1ln and IEEE 802.11ac, where the mechanism of frame aggregation is used to boost the throughput of data transmission. In the last years, many TCP variants have been introduced to limit the bufferbloat phenomena and bound the latency, through a reduction of the queue backlog injection rate. However, this mechanism impacts on the Wi-Fi frame aggregation logic, impeding TCP congestion controls to reach the full throughput potential of a Wi-Fi interface. While this problem can be solved with TCP CUBIC allowing the sender node to enqueue more packets, for TCP BBR the fix is not the same, as it has a customized pacing algorithm. With this contribution, we propose BBR+, a new BBR version that allows to fine tune the congestion control pace, achieving 6 times more throughput over IEEE 802.1ln channels and 16 times more throughout over IEEE 802.11ac channels, at the cost of an increased latency that is however always less than the latency obtainable with TCP CUBIC.

A physical layer (PHY) abstraction model estimates the PHY performance in system-level simulators to speed up the simulations. This paper presents a PHY abstraction model for 5G New Radio (NR) and its integration into an open-source ns-3 based NR system-level simulator. The model capitalizes on the exponential effective signal-to-interference-plus-noise ratio (SINR) mapping (EESM) and considers the latest NR specification. To generate it, we used an NR-compliant link-level simulator to calibrate the EESM method as well as to obtain SINR-block error rate (BLER) lookup tables for various NR configurations. We also illustrate the usability of the developed model through end-to-end simulations in ns-3, under different NR settings of modulation and coding schemes, hybrid automatic repeat request combining methods, and link adaptation approaches.

In this work, we propose an innovative approach to contrast the bufferbloat, one of the primary sources of queueing latency, in 3GPP mobile networks. Our purpose is to validate an all-software, in-node traffic-control infrastructure, similar to what the Linux operating system already employs for Ethernet and WiFi networks. We introduce the IPtraffic-control infrastructure in the RAN part of an LTE network, for both UE and eNB, that can perform packet scheduling on flows directed to the same Data Radio Bearer. The effectiveness of such strategy leverages on a reduced usable size of the in-node RLC buffer, thanks to the BQL algorithm, which we evaluate for the first time in the context of LTE. By using BQL, we can create a backlog of packets that stays in the Traffic Control layer, on which it is possible to perform packet scheduling without bloating the RLC buffer. We compare the performance of BQL to another algorithm, called DynRLC, proposed to dynamically size the RLC buffer eNB side. The preliminary evaluation of the proposed approach is done on the ns-3 network simulator and it shows that the proposed approach with BQL significantly improves the flow delay, by controlling the queueing latency, as well as enhancing the flow isolation.

One of the main design targets of New Radio (NR) is to support multiple applications, including low-latency data transmissions. To achieve that, multiple features have been introduced, among which dynamic scheduling timings (denoted by K0, K1, K2) is the one which determines the delay between the different paired control and data transmissions. For example, K2 is the delay (in unit of slots) between an uplink grant reception and the corresponding uplink data transmission. The End-to-End (E2E) latency would be highly impacted by these scheduling timings. Based on the common observation, lower scheduling timings would lead to low E2E latency. However, in this paper, especially for uplink traffic, we show that this is not always true, and there are cases in which increasing the scheduling timings provides better E2E delay performance. This is due to the interplay between traffic patterns, gNB-UE process, and the NR numerology. To show such impact of scheduling timings on E2E latency with different uplink traffic patterns and NR numerologies, we implement an E2E ns-3 based NR simulator.

Future 5G networks will rely on heterogeneous deployment of micro eNodeBs, and is expected that wireless transport interfaces, deployed together with each eNodeB, will provide high-speed backhaul connectivity through the possibility to create an ad-hoc wireless mesh between the deployed micro eNodeBs. Therefore, the backhaul routing strategy will play a fundamental role in exploiting the backhaul resources. In this context, we aim to investigate the advantages and limitations of qualitatively different backhaul routing strategies as UEs download TCP and UDP content. In this paper, we compare a backpressure-based strategy that takes per-packet or per-flow basis (BP-MR) and traditional single path routing (OLSR). Ns-3 simulations show that the per-packet BP-MR variant enhances UDP performance by circumventing congestion but degrades TCP because of reordering. The per-flow BP-MR variant minimizes retransmissions by limiting the excessive reordering, and outperforms OLSR performance for both UDP backhaul traffic and TCP traffic.

TCP Wave changes the typical TCP transmission paradigm by replacing the ACK-clocked sliding window with self-scheduled bursts. In response to bursts, unmodified TCP receivers generate ACK trains, which carry useful information about the end-to-end link characteristics. TCP Wave inspects ACK-train flow to measure the following parameters: (i) ACK train spread (namely ACK train dispersion) and (ii) RTT variations. The former is expected to provide the overall “service capacity”, meant as the maximum capacity allowed over the end-to-end path, while the latter is considered as a congestion indicator. The joint use of such measurements allows TCP Wave to fine-tune the transmission rate to accurately match current network resource availability in the bottleneck link. This paper analysis TCP Wave ACK train-based measurements on a broad set of simulated links compliant to characteristics of today's real networks. To this scope, a testbed with TCP Wave implementation on Linux OS is used to perform tests varying both bottleneck capacity and physical latency.

In recent years, the Linux kernel has adopted an algorithm called TCP Small Queues (TSQ) for reducing queueing latency by controlling buffering in the networking stack. This solution consists of a back-pressure mechanism that limits the number of TCP segments within the sender TCP/IP stack, waiting for packets to actually be transmitted onto the wire before enqueueing further segments. Unfortunately, TSQ prevents the frame aggregation mechanism in the IEEE 802.11n/ac standards from achieving its maximum aggregation, because not enough packets are available in the queue to build aggregates from, which severely limits achievable throughput over wireless links. This paper demonstrates this limitation of TSQ in wireless networks and proposes Controlled TSQ (CoTSQ), a solution that improves TSQ so that it controls the amount of data buffered while allowing the IEEE 802.11n/ac aggregation logic to fully exploit the available channel and achieve high throughput. Results on a real testbed show that CoTSQ leads to a doubling of throughput on 802.11n and up to an order of magnitude improvement in 802.11ac networks, with a negligible latency increase.

Satellite networking is known to suffer from specific network performance issues, such as high latency and low throughput stability; this derives mainly from the high propagation delay, in particular with GEO satellites. This characteristic poses limitations on the benefits that general AQM solutions could introduce; in fact, reducing congestion and mitigating queueing delay is a vital feature that could boost the performance of satellite networks. This paper investigates PINK (Passive INverse feedbacK), a queue management algorithm designed to indirectly impose an individual resource allocation policy in order to mitigate the bufferbloat effect and the network congestion while exploiting the channel throughput and guaranteeing optimal flow fairness without forcing any packet drop. PINK modifies the TCP Acknowledgements (ACKs) segments passing through the Satellite access gateway. The modification consists in replacing the advertised Receive Windows field (RCV.WNDs) with custom values, in order to enforce a particular bandwidth utilization upper bound. To compute new RCV.WND values, PINK needs only the number of active connections, the flows RTT and the transmission channel bandwidth. The design characteristics of this new AQM let it works efficiently in satellite networks, and we validate this statement through several simulations performed with the ns-3 network simulator.

This paper proposes PINK (Passive INverse feedbacK), a queue management algorithm designed to indirectly impose a certain resource allocation policy on defined sets of client hosts. PINK adds intelligence at intermediate nodes that connect client hosts to bottleneck links or to external networks in general, allowing these nodes to dynamically modify the TCP Acknowledgments (ACKs) segments passing through. This is made by setting TCP ACK advertised Receive Windows field (RCV.WNDs) to custom values, in order to enforce a specific bandwidth utilization upper bound. To compute new RCV.WND values, PINK needs only the number of active connections, the flows RTTs and the transmission channel bandwidth. It follows that PINK permits to impose a centralized bandwidth management without the cooperation of clients, which means that no modification or addition whatsoever to end hosts is needed with the goal of enabling a smart environment by simply configuring the access node. Furthermore, as demonstrated in this paper, our proposal does not constraints client hosts performance without purpose; on the contrary, PINK improves efficiency on multiplexed channels by exploiting channel throughput, maintaining a low queuing delay, and guaranteeing optimal flow fairness without forcing any packet drop.

After a disaster, it is unlikely to expect current terrestrial infrastructures to provide the coverage and the Quality of Service required by emergency operators. Currently, PPDR operators go towards a portable LTE infrastructure that provides an access network to field operators, backhauled by a wireless mesh network that can route traffic from and towards Internet through a satellite gateway. We firstly propose a new per-flow strategy for a backpressure-based routing protocol, BP-MR, and then in the aforementioned environment we analyse the original BP-MR per-packet, the proposed per-flow, and OLSR as routing protocol for the mesh network coupled with three TCP congestion control (Cubic, Hybla, Vegas) performing a typical HTTP transfer. We analysed the throughput, the latency, and the scalability over the number of the flows/operators. An extensive simulation set allows to conclude that employing BP-MR per-flow in the mesh network coupled with TCP Vegas for performing the end-to-end transfer improves the performance for PPDR professionals.

In scenarios where a number of client nodes share a gateway in order to access a high-delay link, the original TCP as well as TCP variants specifically designed to improve performance on those links fail to provide optimal performance; this is due to the indirect inference of both end-to-end congestion and local channel conditions. To mitigate the former, C2ML (Congestion Control Middleware Layer) has been proposed as a tool for centralized and collaborative management of resources that can strongly improve performance for client hosts without affecting network resilience and connection survivability. To complete the picture, this paper proposes DyBRA, an improved version of C2ML with a bandwidth management algorithm designed to dynamically allocate bandwidth among client hosts on the basis of channel conditions, leading to improved performance regardless of the TCP variant employed by clients. Through simulations performed with the ns-3 network simulator, we show how DyBRA strongly improves throughput and TCP friendliness while reducing queue usage and latency.

When multiple nodes share the same gateway or access the same link, the throughput of wireless communications has the unfortunate problem of being limited by traffic congestion; the situation may also further degrade if they establish multiple TCP connections mapped over the same access channel. In end-to- end connections, TCP congestion control plays a critical role in order to not overflow the channel, although most of the network bandwidth bottlenecks are caused by the wireless systems. When TCP is coupled with high-delay channels, the protocol underperform; this is mainly due to long propagation delay, as it is the case with satellite links. To overcome the limitations of TCP on these links, various variants have been proposed, both rate-based or or burst-based. In this paper we present a cooperative transmission control that enhances the effectiveness of TCP asking only for a narrow sender-side modification, operating exclusively at the access stage. It allows to boost throughput over high delay channels, increasing also fairness and friendliness between different TCP algorithms. After the description, we validate its effectiveness and show its properties by means of experimental results, derived from simulations performed on the ns-3 network simulator.

This paper presents preliminary work on PINK (Proactive INjection into acK), an AQM algorithm able to enhance TCP congestion control properties without dropping packets. PINK is a completely transparent solution that does not require any modification to the existing protocol stack of end hosts, and it is particularly suitable for high-delay PPDR systems. The algorithm is based on an explicit feedback scheme, able to enforce a fair bandwidth sharing among clients by modifying the Receive Window in TCP acknowledgements returning to them; such a feedback is computed for each flow, and it only needs the number of active connections, the flows RTTs and the shared link bandwidth. Therefore, PINK is a per-flow stateless AQM that works independently from the TCP algorithm used by clients, and it is fair regardless the flows RTTs, which is a key feature for PPDR systems. The effectiveness of the proposed algorithm is demonstrated using the ns-3 simulator.

Network-enabled services for public protection and disaster relief (PPDR) professionals have been more than necessary in today's emergency situations. Under the extreme circumstances of an emergency, it is essential to have networks which support the required data throughput as well as high availability in spite of high traffic volume, and which minimize the end-to-end delay for applications: however, the choice between the existing technologies is not so easy for PPDR entities, given the high number of parameters associated to satisfying the stringent PPDR requirements, high investments required to permit desirable availability as well as modernization of the existing services (i.e. voice and data) and design constraints posed by network providers as current deployed network reach their boundaries when emergencies occur. This paper analyzes three different emergency scenarios and then presents a subset of the results obtained in terms of financial and economic recommendations, along with technical reports on throughput and end-to-end delay.

The vast majority of efforts aimed to improve network performance are focused on the increase of application throughput. The same holds in the context of Emergency Networks, where operators ask for more bandwidth in order to exploit data-intensive services. Following a steady growth in network capacities and number of users, large buffers have been inserted all over the Internet. Their effects on networks are non-trivial: while they may effectively serve the purpose of exploiting the channel potential, they also create unnecessary delays by damaging the behavior of the most common transport protocol, TCP. Nevertheless, they are being assumed by new congestion control algorithms, especially those tailored for high-latency links such as satellite ones. According to the anywhere-anytime paradigm, these channels represents a key technology for both Emergency and General-Purpose networks. In this paper we first show how buffer lengths impact the perceived delay over satellite links that employ a recently proposed burst-based TCP protocol, and then present a cooperative transmission control that reduce buffers usage and latency while keeping high throughput and flow fairness, thus allowing for a better service provision through satellite links.

Wireless communication technologies are critical for public protection and disaster relief (PPDR) professionals during the emergency operations that follow natural or man-made disasters, scenarios in which commercial terrestrial networks often fail to provide the necessary support. The reason is threefold: they simply get disrupted by the disaster, they cannot sustain the sudden surge of network demand or they fail to deliver the necessary bandwidth and/or other QoS guarantees. In every PPDR operation reliable voice communications are critical, especially in the very early stages of the response; nevertheless, there is an increasing demand from the PPDR community for a wider range of data-centric services. While current technologies used for PPDR operations provide a rich set of voice-centric services, they are unable to sustain high-bandwidth data-oriented applications. As the PPDR-TC EU consortium, we propose a hybrid approach to tackle the question of determining the future architecture for Pan-European PPDR networks based on the integration of Terrestrial and Satellite technologies, presenting our first simulation results on the integration of LTE and Satellite Networks.