Congestion Control In Networks Research Articles

Troubleshooting solution for traffic congestion control

The Internet has existed since the 1970s as a means of data exchange between network devices in small networks. In the early stage, there was a small number of devices, but today there is an ever-increasing number of devices, leading to congestion in the network. Therefore, congestion control has attracted so much attention in the academic community and the industry for the past 30 years. Recently, Google has developed BBR (Bottleneck Bandwidth and Round-Trip Time), a rate-based congestion control algorithm. BBR controls transmission rates based on delivery rate and round-trip time (RTT). However, such a static congestion control algorithm (e.g., BBR, etc.) cannot achieve high performance in various network conditions (e.g., low bandwidth, etc.). Concretely, these static algorithms cannot adapt to the dynamic changes of the network environment. Therefore, in this paper, we propose an adaptive algorithm (called ABBR) for congestion control in next-generation networks. ABBR takes into account the reinforcement learning algorithm to learn relevant policies to change the transmission rate corresponding to each congestion control algorithm to optimize long-term performance. The experimental results show that our proposal can achieve good performance in terms of throughput, RTT, and fairness compared to the benchmarks.

Novel Geodetic Fuzzy Subgraph-Based Ranking for Congestion Control in RPL-IoT Network

Novel Geodetic Fuzzy Subgraph-Based Ranking for Congestion Control in RPL-IoT Network

Efficient fuzzy methodology for congestion control in wireless sensor networks

In wireless sensor networks (WSNs), ensuring reliable and efficient communication between nodes requires effective congestion control. This study presents an innovative approach, called FSFG, which combines FCM fuzzy clustering, Fuzzy Inference System (FIS), Support Vector Regression (SVR), and Genetic Algorithm (GA) to address congestion in WSNs. FSFG consists of two phases: Structure Identification and Parameter Identification. In the first phase, fuzzy clustering techniques are employed to determine the desirability of outputs in different classes. Subsequently, support vector regressions are used to determine membership functions with a minimal error rate. In the parameter identification phase, the obtained rules from the FCM method are used to design the desired system, enabling the estimation of necessary outputs based on relevant inputs. Through simulations, the results demonstrate that the FSFG algorithm excels in congestion control, exhibiting a lower error rate and outperforming other algorithms in terms of delay and packet delivery rate.

Efficient and secure routing with UAV: GuidedPheromone update based on improved Ant colony optimization and fuzzy logic for congestion control in vehicular ad-hoc network

Efficient and secure routing with UAV: GuidedPheromone update based on improved Ant colony optimization and fuzzy logic for congestion control in vehicular ad-hoc network

Mustang: Improving QoE for Real-Time Video in Cellular Networks by Masking Jitter

The advent of 5G and interactive live broadcasting has led to a growing trend of people preferring real-time interactive video services on mobile devices, particularly mobile phones. In this work, we measure the performance of Google congestion control (GCC) in cellular networks, which is the default congestion control algorithm for Web Real-Time Communications (WebRTC). Our measurements show that GCC sometimes makes bitrate decisions which are harmful to quality of experience (QoE) in cellular networks with high jitter. We further find that the frame delivery time (FDT) in the player can mitigate network jitter and maintain QoE. Moreover, the receiving rate is better to reflect the network congestion than RTT in cellular networks. Based on these measurements and findings, we propose Mustang, an algorithm designed to overcome the jitter in cellular networks. Mustang makes use of the FDT and receiving rate as feedback information to the sender. Then the sender adjusts its sending rate based on the information to guarantee QoE. We have implemented Mustang in WebRTC and evaluated it in both emulated and real cellular networks. The experimental results show that Mustang can improve WebRTC’s both QoS and QoE performance. For QoS, Mustang increases the sending rate by 72.1% and has similar RTT and packet loss when compared with GCC, while it is about 30% better for QoE.

Deep Reinforcement Learning for Communication Networks

This research explores optimizing communication tasks with (Multi-Agent) Reinforcement Learning (RL/MARL) in Point-to-Point and Group Communication (GC) networks. The study initially applied RL for Congestion Control in networks with dynamic link properties, yielding competitive results. Then, it focused on the challenge of effective message dissemination in GC networks, by framing a novel game-theoretic formulation and designing methods to solve the task based on MARL and Graph Convolution. Future research will deepen the exploration of MARL in GC. This will contribute to both academic knowledge and practical advancements in the next generation of communication protocols.

SDN-Based Congestion Control and Bandwidth Allocation Scheme in 5G Networks

5G cellular networks are already more than six times faster than 4G networks, and their packet loss rate, especially in the Internet of Vehicles (IoV), can reach 0.5% in many cases, such as when there is high-speed movement or obstacles nearby. In such high bandwidth and high packet loss network environments, traditional congestion control algorithms, such as CUBIC and bottleneck bandwidth and round-trip propagation time (BBR), have been unable to balance flow fairness and high performance, and their flow rate often takes a long time to converge. We propose a congestion control algorithm based on bottleneck routing feedback using an in-network control mode called bottleneck routing feedback (BRF). We use SDN technology (OpenFlow protocol) to collect network bandwidth information, and BRF controls the data transmission rate of the sender. By adding the bandwidth information of the bottleneck in the option field in the ACK packet, considering the flow fairness and the flow convergence rate, a bandwidth allocation scheme compatible with multiple congestion control algorithms is proposed to ensure the fairness of all flows and make them converge faster. The performance of BRF is evaluated via Mininet. The experimental results show that BRF provides higher bandwidth utilization, faster convergence rate, and fairer bandwidth allocation than existing congestion control algorithms in 5G cellular networks.

Congestion Control in Wi-Fi Networks—State of the Art, Performance Evaluation, and Key Research Directions

Congestion Control in Wi-Fi Networks—State of the Art, Performance Evaluation, and Key Research Directions

Harnessing PSO and GA for Congestion Control in High-speed Wireless Sensor Networks

Sensor and sink nodes create wireless sensor networks. Traffic congestion caused by WSN data transfer causes bigger packet losses, low throughput and excessive energy consumption. This work proposes a hybrid congestion management approach using genetic algorithm and particle swarm optimization. The new approach is simulated and compared to established methods. The suggested system dramatically improved performance metrics. The suggested system of interest increased to 94% detection efficiency, 91% network lifespan, 106 J energy usage, and 22 packet loss rate. The hybrid approach avoided wireless sensor network congestion and managed traffic.

Plausibility of BBR as CUBIC’S Replacement and proposed improvement to BBR using GENET

Recent advancements in deep reinforcement learning (RL) have opened new avenues for enhancing network congestion control algorithms. Our research builds upon these developments, particularly focusing on the BBR (Bottleneck Bandwidth and Round-trip propagation time) congestion control algorithm. We propose integrating GENET's reinforcement learning framework, a novel training paradigm that has demonstrated success in various network adaptation algorithms, including adaptive video streaming, congestion control, and load balancing. GENET leverages curriculum learning to effectively train RL models by progressively introducing more challenging network environments. This method counters the common pitfalls in RL training, such as suboptimal performance in a wide range of environments and poor generalization in narrowly defined training scenarios. Our approach exploits the strengths of GENET in identifying and emphasizing network conditions where the current RL model underperforms compared to traditional rule-based baselines, thereby facilitating significant improvements. This research aims to demonstrate that applying GENET's methodology to the BBR congestion control algorithm can yield RL policies that surpass both regularly trained RL policies and conventional baselines, thereby advancing the efficiency and reliability of network congestion control.

Through the RTT Lens : A Comparative study of CUBIC and BBR Congestion Control

As the demand for high-performance internet applications continues to surge, the efficiency of network congestion control algorithms becomes a critical factor in ensuring a seamless user experience. This research paper delves into the comparative analysis of two prominent congestion control mechanisms: BBR (Bottleneck Bandwidth and Round- trip propagation time) and Cubic[1]. Both algorithms play pivotal roles in regulating data flow within networks, but their approaches differ significantly. The comparison section highlights the adaptive behavior of each algorithm, emphasizing real-world implications for diverse network scenarios. The discussion interprets the findings, offering a nuanced understanding of the strengths and weaknesses of both BBR and Cubic, thereby contributing to the broader discourse on congestion control strategies.

Deep-Reinforcement-Learning-Aided Loss-Tolerant Congestion Control for 6LoWPAN Networks

The IPv6 over Low-power Wireless Personal Area Network (6LoWPAN) protocol stack is a promising solution to connect Wireless Sensor Networks (WSNs) with the Internet for realizing a ubiquitous network interconnection of all things. However, 6LoWPAN networks face a critical challenge to control congestion caused by the burst of data traffic from wireless sensors. Packet loss will occur when buffer overflows. This paper focuses on the loss-tolerant congestion control problem in 6LoWPAN networks, which has not been addressed in existing works. We formulate the congestion control problem as a non-cooperative Markov game framework and conceive a novel congestion control method, namely Deep reinforcement learning aided Loss-tolerant Congestion Control (DLCC), to alleviate congestion while maintaining a tolerable packet loss imposed by buffer overflow. The proposed DLCC employs Deep Reinforcement Learning (DRL) to solve the curse of state dimensionality, while packet loss constraints are handled by utilizing Lagrange multipliers to integrate the reward with loss constraints. By dynamically updating Lagrange multipliers in an online learning procedure, DLCC finds the optimal congestion control policy. Our simulation results show that DLCC maintains the packet loss rate below the tolerable threshold at the presence of congestion. In contrast to existing hybrid congestion control algorithms, the proposed DLCC algorithm is more energy-efficient and provides higher throughput, lower average delay and better fairness.

A hierarchical control framework for alleviating network traffic bottleneck congestion using vehicle trajectory data

Traffic bottlenecks significantly influence the operation efficiency of large-scale road networks. Developing advanced control strategies for bottleneck optimization is a cost-efficient and critical way to deal with network congestion. However, the state-of-the-art studies on network congestion control focus on the topology level, which may fail to relieve congestion by addressing the root cause of bottleneck. This study proposed a hierarchical control framework for alleviating network traffic bottleneck congestion using vehicle trajectory data. First, the bottleneck-related sub-network (BRS) was identified by tracing vehicle trajectories upstream and downstream of the bottleneck based on the traffic flow propagation. Then, a hierarchical control framework was proposed for BRS optimization. Specifically, in the outer layer, i.e., the gating control layer, the multigated intersections in BRS were controlled via a multimemory deep Q-network approach to optimize the network traffic distribution. In the inner layer, i.e., the coordinated control layer, local intersection controllers were coordinated by adjusting the dynamic input and output streams of the bottleneck under the guidance of the outer layer controller, which helps balance the traffic pressure within BRS and avoids congestion transferring in the network. Both simulation and field experiments were conducted to verify the performance of the proposed hierarchical framework. Results reveal that the framework can effectively relieve network traffic congestion with decreased queue length and travel time.

Routing Optimizing Decisions in MANET: The Enhanced Hybrid Routing Protocol (EHRP) with Adaptive Routing based on Network Situation

Mobile ad hoc networks (MANETs) are wireless networks that operate without a fixed infrastructure or base station. In MANETs, each node acts as a data source and a router, establishing connections with its neighboring nodes to facilitate communication. This research has introduced the Enhanced Hybrid Routing Protocol (EHRP), which combines the OLSR, AOMDV, and AODV routing protocols while considering the network situation for improved performance. The EHRP protocol begins by broadcasting a RREP (Route Reply) packet to discover a route. The selection of routing options is based on the current network situation. To determine the distance between the source and destination nodes, the proposed EHRP initiates a RREQ (Route Request) packet. In situations where network mobility exceeds the capabilities of the AODV protocol, the EHRP protocol can utilize the OLSR routing protocol for route selection and data transmission, provided that at least 70% of the network nodes remain stable. Additionally, the EHRP protocol effectively handles network load and congestion control through the utilization of the AOMDV routing protocol. Compared to the hybrid routing protocol, the enhanced hybrid routing protocol (EHRP) demonstrates superior performance. Its incorporation of the OLSR, AOMDV, and AODV protocols, along with its adaptive routing adaptation based on network conditions, allows for efficient network management and improved overall network performance.
 The analysis of packet delivery ratio for EHRP and ZRP reveals that EHRP achieves a packet delivery ratio of 98.01%, while ZRP achieves a packet delivery ratio of 89.99%. These results indicate that the enhanced hybrid routing protocol (EHRP) outperforms the hybrid routing protocol (ZRP) in terms of packet delivery ratio. EHRP demonstrates a higher level of success in delivering packets to their intended destinations compared to ZRP.
 The analysis of normal routing load for EHRP and ZRP reveals that EHRP exhibits a normal routing load of 0.13%, while ZRP exhibits a higher normal routing load of 0.50%. Based on these results, it can be concluded that the performance of the Enhanced Hybrid Routing Protocol (EHRP) is significantly better than that of the Hybrid Routing Protocol (ZRP) when considering the normal routing load. EHRP demonstrates a lower level of routing overhead and more efficient resource utilization compared to ZRP in scenarios with normal routing load.
 When comparing the average end-to-end delay between the Enhanced Hybrid Routing Protocol (EHRP) and ZRP, the analysis reveals that EHRP achieves an average delay of 0.06, while ZRP exhibits a higher average delay of 0.23. These findings indicate that the Enhanced Hybrid Routing Protocol (EHRP) performs better than ZRP in terms of average end-to-end delay. EHRP exhibits lower delay, resulting in faster and more efficient transmission of data packets from source to destination compared to ZRP.
 After considering the overall parameter matrix, which includes factors such as normal routing load, data send and receive throughput, packet delivery ratio, and average end-to-end delay, it becomes evident that the performance of the Enhanced Hybrid Routing Protocol (EHRP) surpasses that of the current hybrid routing protocol (ZRP). Across these metrics, EHRP consistently outperforms ZRP, demonstrating superior performance and efficiency. The Enhanced Hybrid Routing Protocol (EHRP) exhibits better results in terms of normal routing load, higher throughput for data transmission and reception, improved packet delivery ratio, and lower average end-to-end delay. Overall, EHRP offers enhanced performance and effectiveness compared to the existing hybrid routing protocol (ZRP).

High traffic communication congestion control for wireless sensor networks based on harmony search optimization

In wireless sensor networks (WSNs), communication between the wireless nodes requires minimum response delay, minimum congestion and communication reliability. A wide variety of sensors produces a mixture of heterogeneous traffics with different reliability requirements. The article focuses on high traffic congestion which affects communication and produces latency. In the existing approaches, the congestion was controlled and the optimization was done during the time of node deployment. In the proposed method, high traffic congestion was controlled by a hop-by-hop approach which was applied in the statically deployed sensor nodes, the optimization was performed at the time of communication. To provide a uninterrupted communication to the WSNs the proposed approach analyses the occupancy ratio of the buffer and evaluates the downstream node congestion level. Here, the Harmony Search Algorithm is considered for design the optimal sensor network with Support Vector Machine (SVM). The experimental result shows the effectiveness and feasibility of the HSA-SVM environment. Also, it significantly enhances communication in diverse traffic conditions, specifically in heavy traffic areas with limited data.

Enhanced Congestion Control in Wireless Networks: A Joint Random Early Detection and Drop Tail Mechanism

Enhanced Congestion Control in Wireless Networks: A Joint Random Early Detection and Drop Tail Mechanism

Joint congestion and contention avoidance in a scalable QoS-aware opportunistic routing in wireless ad-hoc networks.

Opportunistic routing (OR) can greatly increase transmission reliability and network throughput in wireless ad-hoc networks by taking advantage of the broadcast nature of the wireless medium. However, network congestion is a barrier in the way of OR's performance improvement, and network congestion control is a challenge in OR algorithms, because only the pure physical channel conditions of the links are considered in forwarding decisions. This paper proposes a new method to control network congestion in OR, considering three types of parameters, namely, the backlogged traffic, the traffic flows' Quality of Service (QoS) level, and the channel occupancy rate. Simulation results show that the proposed algorithm outperforms the state-of-the-art algorithms in the context of OR congestion control in terms of average throughput, end-to-end delay, and Packet Delivery Ratio (PDR). Due to the higher PDR at different traffic loads and different node densities, it can be concluded that the proposed algorithm also improves network scalability, which is very desirable given the recent changes in wireless networks.

Fine Tuning of Interval Configuration for Deep Reinforcement Learning Based Congestion Control

It is apparent that various internet services in today’s digital ecosystem effectuate different types of networks’ quality of services (QoS) requirements. This condition, in fact, adds another level of complexity to the current network congestion control protocols. Therefore, it drives the adoption of deep reinforcement learning to improve the protocols’ adaptability to the dynamic networks’ QoS requirements. In this case, the state-of-the-art works on congestion control protocols, formulate the markov decision process (MDP) by transforming the congestion control pattern from the saw tooth congestion window to the staircase sending rate per-interval cycles. This approach treats congestion control as a sequential decision-making process that fits reinforcement learning. However, the interval configuration parameter that gives the optimum QoS has not been empirically studied. In this work, we present an extensive study on various interval configuration parameters for the deep reinforcement learning-based congestion control agent. Our work shows that various interval configuration, which consists of the RTT estimator and the n parameter, results in different QoS. The experiment shows that the RTTjk has significantly higher throughput than RTTewma and RTTmin−filtered in various network conditions. Furthermore, we found that the RTTjk with n = 2.0 is superior to other configurations in almost all networking scenarios. Whereas the RTTjk with n = 1.0 is optimal for a network environment with fixed bandwidth scenario.

Consensus Congestion Control for Ad Hoc Networks: Time-Delay and Saturation

This work focuses on the congestion control design for ad hoc networks affected by time-delay and asymmetrical input saturation. First, an innovative general TCP/AQM model for mobile and wireless ad hoc networks (MANET/WANET) is proposed using graph theory and TCP/AQM system dynamics in a dumbbell topology. Next, a new characteristic condition for sector linearity is developed, which plays the role of a specification to be checked and allows us to assume that the dynamics of the nodes are not saturated. Finally, we propose a new consensus-based congestion control protocol while satisfying the input constraints based on the Lyapunov method, the new sector of linearity condition, the Bessel Legendre-based integral inequality, and the reciprocal convex combination. The proposed results have the advantage of being given as a set of linear matrix inequality (LMI) conditions, which can be solved using convex optimization toolboxes. The merits of the theory are evaluated with the help of various real and challenging scenarios with the use of MATLAB and NS3.

A Weighted Optimal Scheduling Scheme for Congestion Control in Cloud Data Center Networks

Cloud services are experiencing a remarkable increase in the number of users and the resource required over the past few years. Thus, it has become a great challenge for the internet vendors to make a robust framework to serve the customers with low cost and delay. Congestion control is one of the essential topics of routing algorithms in cloud data center networks. In this paper, we propose a weighted optimal scheduling scheme WSPR for congestion control in cloud data center networks which prevents the congestion in advance with the global view so that it can make good use of vacant network resources. We choose BCube as our network model and modify the network topology to fit software-defined networks so as to have a full view of the topology. First, we design the SP graph which contains all shortest paths between a source server and a destination server. Second, we propose WSPR to allocate the most appropriate path to each flow for congestion control. We implement a system to simulate a data center, and evaluate our proposed scheme WSPR by comparing WSPR with other classical methods. The experimental results demonstrate that our proposed scheme WSPR has the best performance in terms of the maximum delay, average delay, and throughput among all compared methods.