TCP Congestion Research Articles

A Weighted AIMD Congestion Control Algorithm for Device-to-Device Communication in 5G Network

ABSTRACT Device-to-Device (D2D) communication offers promising benefits in 5G networks in terms of improved throughput and reduced latency. However, efficient congestion control remains a critical challenge to ensure optimal performance and resource utilization. Existing congestion control mechanisms often struggle with adapting to dynamic network conditions, leading to inefficiencies, such as underutilization of network resources or excessive packet loss. This paper addresses these challenges by proposing a novel Weighted AIMD congestion control algorithm tailored specifically for D2D communication in 5G networks (WACC-D2DC-5GN). Here, the Weighted AIMD Congestion Control Algorithm is proposed for estimating available bandwidth and Hunger Games Search Optimization to optimize the weight parameter of WACC. The proposed algorithm integrates weighted factors that dynamically adjust congestion window sizes based on real-time network feedback. Simulations are run on the NS3-mmWave module, NS3 LTE-A module. The simulation outcomes show that the proposed WACC-D2DC-5GN method attains 27.26%, 20.41%, and 23.26% higher SNR and 16.20%, 26.97%, and 30.34% higher Throughput when analysed with existing techniques like NexGen D-TCP: next generation dynamic TCP congestion control approach (NGD-TCP-CCA), user-centric channel allocation system for 5G higher-speed users by using ML to decrease handover rate (UCC-HSU-MLA), and efficient with reliable hybrid deep learning-allowed method for congestion control in 5G/6G networks (HDL-EMC-CN) respectively.

UNDERSTANDING THE TRADE-OFF IN TCP'S CUBIC WINDOWS CONGESTION: A COMPREHENSIVE ANALYSIS

This study investigates the difficulties of the CUBIC TCP algorithm, focusing on the trade-off between efficiency and fairness in window congestion control. The CUBIC approach, which was devised to optimize high-bandwidth and long-distance networks, is currently widely used as the default TCP congestion control algorithm in Linux and other systems. Our work investigates the dynamics of CUBIC's window expansion, specifically the interaction between the cubic function and the most recent congestion event. We look at how this interplay impacts the distribution of window widths, throughput, round-trip time (RTT), loss rate, and queue size under different network configurations. Keywords: TCP CUBIC, Window Congestion, Efficiency-Fairness Trade-Off, TCP’s Cubic Function.

SOHO-FL: A Fast Reconvergent Intra-Domain Routing Scheme Using Federated Learning

The development of machine learning provides a new paradigm for network optimizations, e.g., reinforcement learning (RL) has brought great improvements in many fields, such as adaptive video streaming, congestion control of TCP. The fundamental mechanism of such RL-based architectures is that the neural network decision model converges to a stable state by continuously interacting with network environment. However, for network routing problem, such RL-based strategies do not work well due to topology change. This is because topological changes would require the existing RL models to be retrained, while these models may stop making routing decisions or provide non-optimal decisions during the slow reconverging process of retraining, seriously affected transmission performance. To solve this problem, we proposed a fast convergent RL-model (SOHOFL), which can alleviate the performance degradation caused by the slow retraining process by federated learning. The experimental results based on real-world network topologies demonstrate that SOHO-FL outperforms the state-of-the-art algorithms in reconvergence time by 22.3% on average.

TCP BBR-n: Increased throughput for wireless-AC networks

Google proposed a new TCP congestion control algorithm (CCA), Bottleneck Bandwidth and Round-trip propagation time (BBR) which has opened up new dimensions in congestion control. BBR tries to operate near Kleinrock’s operating point to avoid excessive queue formation at the bottleneck and to use the link bandwidth optimally. BBR creates a model of the network path by measuring the bottleneck bandwidth and minimum round-trip time (RTT) to maximize the delivery rate and minimize latency. BBR v2 is an updated version of BBR which addresses many shortcomings of the original BBR (BBR v1) such as interprotocol fairness, RTT fairness, and excessive retransmissions. However, BBR v2 has certain limitations in its operation in IEEE 802.11ac (Wi-Fi 5) networks. The default BBR v2 limits the throughput of Wi-Fi 5 and an increased latency has been observed. This is because the Wi-Fi 5 frame aggregation logic is underutilized and fewer frames are being sent to the Wi-Fi 5 interface. In this paper, we have proposed BBR-n (BBR new) which provides better throughput than the generic BBR v2 in the Wi-Fi 5 networks. Real-time experiments were performed over a physical testbed using Flent to confirm that BBR-n achieves over double throughput as compared to generic BBR v2 and reduced latency in networks as compared to pure loss-based variants such as Cubic and Reno.

Evaluation of TCP Congestion Control Algorithms with traffic control policies in a PEP-based geosynchronous satellite scenario

Much work has been done in recent years comparing TCP variants in many scenarios, including the satellite one. Some of these studies were carried out with the ns-2 and ns-3 simulators, which may give a good idea of the behavior of these TCP variants, although the results cannot be considered completely realistic. Some of the latest studies use virtualization to derive a variety of measures to test the suitability of Congestion Control Algorithms in a more realistic way. Such studies also take into account new variants of TCP, such as Google’s BBR. However, few of these works have evaluated the behavior of these variants in a scenario based on Performance Enhancing Proxies. Furthermore, no emphasis has been placed on those variants of TCP that try to maximize throughput and minimize delay, which is crucial in many services with low latency requirements. Our paper aims to fill this gap, offering a comparative assessment of the performance of these three TCP flavors: BBR, YeAH and CUBIC. We have created four different approaches to mix these TCP flavors with two traffic control techniques, including Active Queue Management (AQM) policies. To perform the testings, we have developed a TCP-Splitting satellite Emulation Framework (TSEF), a platform that allows us to evaluate the performance of TCP flavors on the forward channel of a geosynchronous satellite scenario based on Performance Enhancing Proxies (PEPs). According to our results, we can conclude that the application of AQM policies and ECN marks on aggressive TCP flavors like CUBIC is mandatory to maximize throughput and minimize delay.

DECC: Achieving Low Latency in Data Center Networks With Deep Reinforcement Learning

Data Center Networks (DCNs) suffer from synchronized bursts for segmentation and aggregation modes, leading to buffer overflows at switches and increasing network delay. To overcome this problem, some congestion control algorithms like DCTCP use Explicit Congestion Notification (ECN) to notify in-network congestion and reduce switch buffer occupancy. However, the traditional Additive Increase Multiplicative Decrease (AIMD) method causes high fluctuation of round-trip time (RTT) in DCNs. Some intelligent congestion control algorithms designed for Internet can achieve great flexibility, but are not applicable in DCNs for a lack of accurate congestion feedback. In this paper, we analyze the deficiencies of utilizing RTT as congestion signals and the applicability of learning algorithms in DCNs. Then, we propose DECC, a smart TCP congestion control algorithm for DCNs, which combines Deep Reinforcement Learning (DRL) with ECN to achieve high bandwidth utilization as well as low queuing delay. DECC fully utilizes precise in-network feedback and formulates several QoS requirements to a multi-objective function. Meanwhile, it decouples cwnd adjustment with DRL decision making to gradually learn the optimal congestion control policy in real-time. We evaluate the performance of DECC in various scenarios. Simulation results show that DECC can reduce the queue length at bottleneck switches by more than 50% compared to DCTCP, while maintaining high bandwidth utilization and reducing Flow Completion Time (FCTs) under burst traffic.

An Accurate Platform for Investigating TCP Performance in Wi-Fi Networks

An increasing number of devices are connecting to the Internet via Wi-Fi networks, ranging from mobile phones to Internet of Things (IoT) devices. Moreover, Wi-Fi technology has undergone gradual development, with various standards and implementations. In a Wi-Fi network, a Wi-Fi client typically uses the Transmission Control Protocol (TCP) for its applications. Hence, it is essential to understand and quantify the TCP performance in such an environment. This work presents an emulator-based approach for investigating the TCP performance in Wi-Fi networks in a time- and cost-efficient manner. We introduce a new platform, which leverages the Mininet-WiFi emulator to construct various Wi-Fi networks for investigation while considering actual TCP implementations. The platform uniquely includes tools and scripts to assess TCP performance in the Wi-Fi networks quickly. First, to confirm the accuracy of our platform, we compare the emulated results to the results in a real Wi-Fi network, where the bufferbloat problem may occur. The two results are not only similar but also usable for finding the bufferbloat condition under different methods of TCP congestion control. Second, we conduct a similar evaluation in scenarios with the Wi-Fi link as a bottleneck and those with varying signal strengths. Third, we use the platform to compare the fairness performance of TCP congestion control algorithms in a Wi-Fi network with multiple clients. The results show the efficiency and convenience of our platform in recognizing TCP behaviors.

A Detection and Mitigation Scheme of LDoS Attacks via SDN Based on the FSS-RSR Algorithm

Software-defined networking (SDN) is a new network architecture that separates the data plane from the control plane and provides network programmability, dynamic deployment, and management of network traffic. However, its security also faces many threats, such as low-rate denial of service (LDoS) attacks. The LDoS attack can use the vulnerability of the TCP congestion control mechanism on the network and requires only a small amount of attack data to achieve the attack effect, which make it harder to detect and mitigate than traditional DoS attacks. In this paper, we propose an LDoS attack detection and mitigation scheme. Our detection scheme collects the features of network traffic and uses the FSS-RSR algorithm to detect LDoS attacks in real time. Our mitigation scheme monitors traffic fluctuations to locate attack source IP addresses and delivers traffic rules to mitigate attacks. The experimental results show that our scheme has high accuracy and low complexity for LDoS attack detection, and a short response time for LDoS attack mitigation of fewer than 8 seconds.

A Comparative Analysis of Ookla Speedtest and Measurement Labs Network Diagnostic Test (NDT7)

Consumers, regulators, and ISPs all use client-based "speed tests" to measure network performance, both in single-user settings and in aggregate. Two prevalent speed tests, Ookla's Speedtest and Measurement Lab's Network Diagnostic Test (NDT), are often used for similar purposes, despite having significant differences in both the test design and implementation, and in the infrastructure used to perform measurements. In this paper, we present the first-ever comparative evaluation of Ookla and NDT7 (the latest version of NDT), both in controlled and wide-area settings. Our goal is to characterize when and to what extent these two speed tests yield different results, as well as the factors that contribute to the differences. To study the effects of the test design, we conduct a series of controlled, in-lab experiments under a comprehensive set of network conditions and usage modes (e.g., TCP congestion control, native vs. browser client). Our results show that Ookla and NDT7 report similar speeds under most in-lab conditions, with the exception of networks that experience high latency, where Ookla consistently reports higher throughput. To characterize the behavior of these tools in wide-area deployment, we collect more than 80,000 pairs of Ookla and NDT7 measurements across nine months and 126 households, with a range of ISPs and speed tiers. This first-of-its-kind paired-test analysis reveals many previously unknown systemic issues, including high variability in NDT7 test results and systematically under-performing servers in the Ookla network.

Intelligent TCP Congestion Control Policy Optimization

Network congestion control is an important means to improve network throughput and reduce data transmission delay. To further optimize the network data transmission capability, this research suggests a proximal policy optimization-based intelligent TCP congestion management method, creates a proxy that can communicate with the real-time network environment, and abstracts the TCP congestion control mechanism into a partially observable Markov decision process. Changes in the real-time state of the network are fed back to the agent, and the agent makes action commands to control the size of the congestion window, which will produce a new network state, and the agent will immediately receive a feedback reward value. To guarantee that the actions taken are optimum, the agent’s goal is to obtain the highest feedback reward value. The state space of network characteristics should be designed so that agents can observe enough information to make appropriate decisions. The reward function is designed through a weighted algorithm that enables the agent to balance and optimize throughput and latency. The model parameters of the agent are updated by the proximal policy optimization algorithm, and the truncation function keeps the parameters within a certain range, reducing the possibility of oscillation during gradient descent and ensuring that the training process can converge quickly. Compared to the traditional CUBIC control method, the results show that the TCP-PPO2 policy reduces latency by 11.7–87.5%.

Intelligent TCP Congestion Control Scheme in Internet of Deep Space Things Communication

In deep space missions, large-scale deep space information and continuous sensory data are generated, processed, and exchanged over Internet of deep space things (IoDST). Ensuring reliable communication in IoDST requires resolution of challenges engendered by the unique characteristics of IoDST, including long propagation delays, link outages, high error rates, and asymmetric bandwidths. Thus, transmission control protocol (TCP) layer functionalities are crucial for ensuring reliable IoDST communications. However, existing TCP congestion control (CC) protocols present poor performance in IoDST communications, owing to the dependence of traditional window-based CC approaches on pre-configured rules to adjust transmission rate against the aforementioned unique characteristics. In this paper, we propose an intelligent CC scheme called optimistic actor–critic-based TCP congestion control (OAC-TCPCC) to solve the problems of the challenging link conditions in IoDST. OAC-TCPCC dynamically determines optimal congestion window for data transmission over IoDST communication links to maximize the TCP throughput performance and minimize file transfer time. Simulation results reveal that OAC-TCPCC improves the TCP throughput performance by up to 27%, 79%, 92%, 477%, 643%, and 766%; and reduces file transfer time by up to 16%, 37%, 50%, 48%, 58%, and 79%, compared to DRL-CC, TP-Planet, FAST TCP, TCP Peach, TCP BBR, and TCP CUBIC, respectively.

RSCAT: Towards zero touch congestion control based on actor–critic reinforcement learning and software-defined networking

Network congestion is a phenomenon present in contemporaneous data centers (DCs) independently of scale and underlying technologies. The small-scale presence of congestion causes information delay for the DC hosted services, while in large-scale data losses or even service downtimes are not uncommon. Several congestion control algorithms out in the market have no network support, and as such, they must act and rely only on the data provided by the servers they are running on. Some algorithms have used network’s feedback to improve the performance, however these solutions depend on tuning parameters for marking queued packets, an arduous task on DCs shared by multiple applications with different workloads. In turn, Software-Defined Networking (SDN) created an opportunity to avoid congestion once the centralized controller can gather ongoing and historical information from all network switches and flows. The data gathered is enormous, and fast-computing algorithms are crucial for decision-making. In this sense, this work proposes Reinforcement Learning and SDN-aided Congestion Avoidance Tool (RSCAT), which uses data classification to determine if the network is congested and actor–critic reinforcement learning to find better Transmission Control Protocol (TCP) parameters. We evaluated RSCAT based on three workloads: a diversified DC traffic simulation; high-performance computing from NAS benchmark; and Hadoop TeraSort application. Our results show RSCAT fundamentals (reinforcement learning, data classification, and SDN) are potential candidates to achieve zero touch network congestion control. Specifically, our experimental analysis shows RSCAT, running alongside with traditional TCP congestion control algorithms like DCTCP and CUBIC, can decrease the flow completion time and applications’ runtimes in several cases, without requiring any software update on DC end-points.

Improving the TCP Newreno Congestion Avoidance Algorithm on 5G Networks

Enhancing Transmission Control Protocol (TCP) performance is one of the key solutions for improving the performance of modern wireless communication networks. It is a highly dependable protocol for communication between network hosts based on an internet protocol (IP). It uses packet sequence numbering and acknowledgment (ACK) packets to ensure that data is recoverable in the event of problems like data loss, and corruption. Loss-based Congestion Control (CC) algorithms overreact and underperform in the presence of rapid channel oscillations, resulting in buffer bloat and significant delays. The goal of this paper is to develop an efficient congestion control mechanism for the new TCP agent Newreno over the fifth generation (5G) network, as well as to improve the increment and decrement strategy for adjusting the congestion window (cwnd) during the congestion avoidance phase, It also fast adapts to the current network requirements and adjusts the transmission rate accordingly. TCP enhancements can improve performance under dynamic traffic loads and over long product channels with high bandwidth. As a result, when such a TCP congestion control mechanism is improved, the network’s throughput increases, and low packet losses, and delay are reduced.

End-to-End TCP Congestion Control as a Classification Problem

The traditional rule-based congestion control algorithms cannot set congestion window size flexibly, resulting in the inadaptation of the dynamic networks. This article presents a method to model end-to-end TCP congestion control problem using classification techniques. The network status parameters as the input and the type of network status as the output are defined through the analysis of some existing congestion control algorithms. NewReno, CUBIC, and Compound are used as feedback to produce the training data of the XGBoost classifier. The experimental results show that the classifier effectively shapes the strategies of three outstanding congestion control algorithms and almost achieves the same throughput, delay, and fairness. The proposed method makes the congestion control algorithm able to learn from data produced by network.

A Comparative Analysis of Ookla Speedtest and Measurement Labs Network Diagnostic Test (NDT7)

Consumers, regulators, and ISPs all use client-based "speed tests" to measure network performance, both in single-user settings and in aggregate. Two prevalent speed tests, Ookla's Speedtest and Measurement Lab's Network Diagnostic Test (NDT), are often used for similar purposes, despite having significant differences in both the test design and implementation, and in the infrastructure used to perform measurements. In this paper, we present the first-ever comparative evaluation of Ookla and NDT7 (the latest version of NDT), both in controlled and wide-area settings. Our goal is to characterize when and to what extent these two speed tests yield different results, as well as the factors that contribute to the differences. To study the effects of the test design, we conduct a series of controlled, in-lab experiments under a comprehensive set of network conditions and usage modes (e.g., TCP congestion control, native vs. browser client). Our results show that Ookla and NDT7 report similar speeds under most in-lab conditions, with the exception of networks that experience high latency, where Ookla consistently reports higher throughput. To characterize the behavior of these tools in wide-area deployment, we collect more than 80,000 pairs of Ookla and NDT7 measurements across nine months and 126 households, with a range of ISPs and speed tiers. This first-of-its-kind paired-test analysis reveals many previously unknown systemic issues, including high variability in NDT7 test results and systematically under-performing servers in the Ookla network.

K-means Clustering, Unsupervised Classification, K-NN, Euclidean Distance, Genetic Algorithm

In recent days, the need to provide reliable data transmission over Internet traffics or cellular mobile systems becomes very important. Transmission Control Protocol (TCP) represents the prevailing protocol that provide reliability to data transferring in all end-to-end data stream services on the Internet and many of new networks. TCP congestion control has become the key factor manipulating the behavior and performance of the networks. TCP sender can regulates the size of the congestion window (CWND) using the congestion control mechanism and TCP dynamically adjust the window size depending on the packets acknowledgment (ACK) or by indicates the packets losses when occur. TCP congestion control includes two main phases, slow-start and congestion avoidance and these two phases even work separately, but the combination of them controls CWND and the packet injection to the network pipe. Congestion avoidance and slow-start are liberated mechanisms and using unlike objectives, but if the congestion happens, they are executed together. This article provides an efficient and reliable congestion avoidance mechanism to enhancing the TCP performance in large-bandwidth low-latency networks. The proposed mechanism also includes a facility to send multiple flows over same connection with a novel technique to estimate the number of available flows dynamically, where the all experiments to approving the proposed techniques are performed over the network simulation NS-2.

A Comparative Performance Analysis of TCP Congestion Control Algorithm: Elastic TCP vs e-Elastic TCP

A Comparative Performance Analysis of TCP Congestion Control Algorithm: Elastic TCP vs e-Elastic TCP

A comparative performance analysis of TCP congestion control algorithm: elastic TCP vs. e-Elastic TCP

A comparative performance analysis of TCP congestion control algorithm: elastic TCP vs. e-Elastic TCP

BBR-With Enhanced Fairness (BBR-EFRA): A new enhanced RTT fairness for BBR congestion control algorithm

Towards the end of 2016, the Google research team proposed and developed a new state-of-the-art TCP congestion control algorithm called Bottleneck Bandwidth and Round-trip propagation time (BBR). When deployed on various Google internal servers, BBR attained higher throughput and low latency performances on modern-day and sophisticated networks than traditional congestion control algorithms like Cubic, Reno, or Vegas. Unlike conventional congestion control algorithms, BBR controls data transmission by maximizing delivery rate and minimizing the round-trip time (RTT), therefore maximizing bandwidth utilization and improving throughput and latency delay performances. However, some experiments have reported persistent queue formation and massive packet retransmissions rate in the bottleneck link, which happens to be the main cause of unfairness between different RTT flows. BBR prefers and favors long RTT over short RTT flows; therefore, long RTT flows are allocated more bandwidth when competing with short RTT flows in the bottleneck link. It was also noted that even the minor difference between the two competing flows could be a source of throughput imbalance and unfairness. The dominance of long RTT flows is the central origin of the high queuing delay, packet loss and retransmission rates, and even severe BBR vulnerability that malicious users can exploit to obtain a larger share of bandwidth simply by increasing the RTT (delay). Therefore, we propose a BBR-With Enhanced Fairness (BBR-EFRA) to mitigate this major concern challenge. Our proposed algorithm adaptively controls the congestion window (CWND) by adjusting the bandwidth delay product (BDP) values of each BBR flow during data transmission based on buffer queue status computation. Our proposed approach guarantees different RTT flows to compete for the available bottleneck bandwidth more equally, hence improving the BBR fairness issue. We evaluated our algorithm on the NS3 simulating environment. BBR-EFRA shows improved RTT fairness by more than 16%, reduced retransmission rate by more than 20%, and queuing delay by more than 18%, and finally increased Jain’s fairness index by more than 1.3 times compared with other recently published BBR variants like an adaptive congestion window of BBR(BBR-ACW). Therefore, with BBR-EFRA, short RTT flows can fairly compete for the available bandwidth even with higher RTT differences under various network scenarios.

Yinker: A flexible BBR to achieve the high-throughput and low-latency data transmission over Wi-Fi and 5G networks

The popularization and development of Wi-Fi and 5G networks have introduced new applications requiring high data rates and low latency. However, the vast random packet loss caused by mobility and channel conditions in wireless networks can worsen the performance of traditional TCP congestion control algorithms. Therein, BBR [1] was proposed in 2016, and claims to operate at the optimum point. BBR strives to match the congestion window to the bandwidth-delay product, calculated from the measured bottleneck bandwidth and round trip times. Through simulations, we found that BBR suffers from different degrees of throughput degradation for different loss rates. The improved BBR, Yinker, is proposed to address this issue. Precisely, we dynamically adjust BBR’s pacing_gain based on the network conditions, including loss rate and congestion degree. We have evaluated Yinker in both real-world environments and trace-based emulations and compared its performance with different BBR variants and state-of-the-art schemes, including Cubic, Verus, and Copa. On average, TCP D*, BBR v2, Copa, and BBR have 4.24×, 2.34×, 2.01×, and 1.48× lower throughput compared to Yinker, respectively. This outstanding delay performance comes at little cost in latency. For instance, compared to TCP D* (which achieves the lowest latency), Yinker’s latency is only about 3% more.