Packet Loss Constraints Research Articles

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.

Performance limits of network delay systems based on bandwidth and packet loss constraints

AbstractIn this work, we study the error limits of the network delay systems in terms of bandwidth and packet loss. The tracking error index of networked delay systems is identified by the tracking error energy. The error limits of the network delay systems are achieved by the spectral decomposition and partial decomposition techniques. The results of this work show that the tracking error limits are relevant to the inherent characteristics of the provided object (non‐minimum phase zero, time delay, and unstable pole). In addition, the packet loss, channel noise, and bandwidth also determine the error limits of the network delay systems. The simulation results show that the theory presented is valid.

Distributed crowd filtering mechanism based on heterogeneous network delay and data packet loss constraint

The differences among heterogeneous networks, network delay, and data packet error could decline network performance. In order to solve the above problems, we study a kind of distributed crowd-filtering mechanism. This mechanism can effectively control the heterogeneous network diversity delay by updating the coordination signal matrix, the aggregation signal matrix, and the control signal matrix. The mechanism can effectively improve network reliability based on the packet loss constraint. Considering the influence of heterogeneous network latency on the reliability of data mining results, we studied the crowd-filtering scheme for eliminating the effect of packet loss on the efficiency of data filtering. Experimental results show that, compared with the ant colony algorithm, the distributed crowd filter can accurately predict the data position, reduce the filtering error, and improve the classification accuracy and the data characteristics.

On the Relationship Between QoS & QoE: Why Differential Traffic Management on the Internet Is Not a Zero-Sum Practice

Is differential traffic management (i.e., prioritizing some forms of Internet traffic over others) a zero-sum practice? Do the benefits of favoring Internet traffic from some edge providers come only at the expense of other edge providers? Or, rather, do the benefits of prioritization exceed the corresponding harms, making this a positive-sum practice that enhances aggregate welfare?In paragraphs 125-32 of the 2015 Open Internet Order, the FCC laid out its case for why “paid prioritization” — defined as “the management of a broadband provider’s network to directly or indirectly favor some traffic over other traffic, including through use of techniques such as traffic shaping, prioritization, resource reservation, or other forms of preferential traffic management, either (a) in exchange for consideration (monetary or otherwise) from a third party, or (b) to benefit an affiliated entity” — should be banned, ex ante, as a per se unreasonable practice. To justify this proscription, the FCC pointed to support in the record from commenters asserting that differential traffic management is “inherently a zero-sum practice,” which will inevitably lead to a bifurcated Internet wherein certain edge providers are relegated to a “slow lane” with degraded performance. The FCC also cited to multiple opposing comments in the record that argued traffic prioritization is not a zero-sum game, but, evidently, did not find these arguments persuasive.Drawing on principles of game theory, cognitive psychology, and statistics, this paper examines this discrete issue and posits that — contrary to the arguments accepted by the FCC — differential traffic management on the Internet is not a zero-sum practice. Specifically, this paper recognizes the relationship between the quality of service (QoS) for edge services and quality of experience (QoE) for users, and observes that while these two variables are positively correlated for all edge services, the strength of such correlation differs significantly. For edge services that are particularly sensitive to latency, packet loss, or bandwidth constraints (e.g., live video, online multiplayer gaming), the users’ QoE will have a strong direct correlation to the services’ QoS. Conversely, for edge services that are comparatively insensitive to latency, packet loss, and bandwidth constraints (e.g., email, software updates), the direct correlation between QoS and QoE will be significantly weaker.Thus, if an ISP were to prioritize services that fall into the former category, and de-prioritize services that fall into the latter category, the benefits yielded to the former (i.e., higher user QoE) would outweigh the harms done to the latter (i.e., slightly lower user QoE), so the network as a whole would work better and aggregate consumer welfare would be increased. For example, in such a system, users streaming live video would be much happier, while users checking their email would be only a little worse off, and may see no significant reduction in QoE. This suggests that the FCC should perhaps reconsider its view of such traffic management practices going forward, particularly as new services with greater QoS demands (e.g., VR-based services) begin to enter the market.

Stability and Capacity of Regular Information Theory and Communication Networks

The problem of throughput optimization in decentralized wireless networks with spatial randomness under queue stability and packet loss constraints are investigated. Two key performance measures are analyzed, namely the effective link throughput and the network spatial throughput. Specifically, the double of medium access probability, coding rate, and maximum number of retransmissions that maximize each throughput metric is analytically derived for a class of Poisson networks, in which packets arrive at the transmitters following a geometrical distribution. Necessary conditions so that the effective link throughput and the network spatial throughput are stable and achievable under bounded packet loss are determined, as well as upper bounds for both cases by considering the unconstrained optimization problem. It results show in which system configuration stable achievable throughput capacity obtained as a function of the network density and the arrival rate. Finally the time delay has get reduced at high rate. The throughput for the two nodes has been found efficiently.

Throughput Optimization in Wireless Networks Under Stability and Packet Loss Constraints

The problem of throughput optimization in decentralized wireless networks with spatial randomness under queue stability and packet loss constraints is investigated in this paper. Two key performance measures are analyzed, namely the effective link throughput and the network spatial throughput. Specifically, the tuple of medium access probability, coding rate, and maximum number of retransmissions that maximize each throughput metric is analytically derived for a class of Poisson networks, in which packets arrive at the transmitters following a geometrical distribution. Necessary conditions so that the effective link throughput and the network spatial throughput are stable and achievable under bounded packet loss are determined, as well as upper bounds for both cases by considering the unconstrained optimization problem. Our results show in which system configuration stable achievable throughput can be obtained as a function of the network density and the arrival rate. They also evince conditions for which the per-link throughput-maximizing operating points coincide or not with the aggregate network throughput-maximizing operating regime.

Maximizing System Energy Efficiency by Exploiting Multiuser Diversity and Loss Tolerance of the Applications

We address the problem of energy efficient scheduling over fading channels for the loss tolerant applications. The proposed scheduling scheme allows dropping of a certain predefined proportion of data packets on the transmitter side. However, there is a hard constraint on the maximum number of successively dropped packets. The scheduler exploits average data loss tolerance to reduce the average system energy expenditure while fulfills the hard constraint on the number of successively dropped packets. We explore the effect of average and successive packet loss constraints on the system energy and characterize the regions where one parameter is more critical as compared to the other in the sense of achieving better energy efficiency. The scheme is analyzed using asymptotically large user limit and the optimized channel¿-dependent dropping thresholds are computed by using combinatorial optimization techniques. The numerical results illustrate the energy efficiency of the scheme as a function of the average and successive packet drop parameters.

Development of a Network Adaptive H.264/AVC Medical Video Transmission System

Network high definition video applications such as video calls and videoconferences have been increasingly developed recent years. When congestions occur and network environment becomes unstable, the network applications have to deal with such issues like delay, packet loss and bandwidth constraints. In this paper, we describe a high definition H.264/AVC video transmission system with network adaptive function based on real-time transport protocol (RTP) and real-time transport control protocol (RTCP). The adaptive system is composed of three components: 1) transmission management (TM) which focuses on network state monitoring, bandwidth allocating and transmission rate control; 2) a feedback-based adaptive forward error correction (AFEC) scheme using Reed Solomon (RS) code; 3) adaptive video encoding (AVE) that adjusts the bitrate of video streaming according to the bandwidth constraint information offered by TM. In our paper, delay, packet loss and bandwidth constraints are generated by the network bottleneck and congestions. Hence, the main idea is to enhance the bitrate of video streaming adapt to the bandwidth constraints. In addition, we integrate AFEC with RTCP to feedback bandwidth status and make the adaptive video streaming more robust under packet loss environment. Here we call it interactive adaptive forward error correction (IAFEC). The test result shows IAFEC can be adaptive to strict bandwidth constraints and high packet loss rates.

Cross-Layer Design for MRT Systems with Channel Estimation Error and Feedback Delay

In this paper, we analyze the effect of channel estimation error and delayed feedback on the performance of a cross-layer scheme combining adaptive modulation for maximum ratio transmission (MRT) at the physical layer and truncated automatic repeat request (T-ARQ) at the link layer. The cross-layer attempts to maximize the spectral efficiency under predetermined quality of service requirements, namely, delay and packet-loss constraints. In particular, we derive closed-form expressions for the average packet error rate, average achievable spectral efficiency, and the outage probability, hence, enabling to evaluate the performance of the cross-layer scheme in the presence of imperfect channel estimation at the receiver and delayed feedback path between the receiver and transmitter. The final expressions are given in terms of the minimum and maximum number of transceiver antennas as well as the channel estimation error and feedback delay parameters. We further show that with a small amount of channel estimation error and/or feedback delay the system performance is severely degraded. Specifically, numerical results demonstrate that with channel estimation error and feedback delay the performance of the cross-layer scheme is even worse than that of T-ARQ only. We also observe an irreducible outage probability floor appearing in the high signal-to-noise regime.

End-to-end quality of service constrained routing and admission control for MPLS networks

Multiprotocol label switching (MPLS) networks require dynamic flow admission control to guarantee end-to-end quality of service (QoS) for each Internet protocol (IP) traffic flow. In this paper, we propose to tackle the joint routing and admission control problem for the IP traffic flows in MPLS networks without rerouting already admitted flows. We propose two mathematical programming models for this problem. The first model includes end-to-end delay constraints and the second one, end-to-end packet loss constraints. These end-to-end QoS constraints are imposed not only for the new traffic flow, but also for all already admitted flows in the network. The objective function of both models is to minimize the end-to-end delay for the new flow. Numerical results show that considering end-to-end delay (or packet loss) constraints for all flows has a small impact on the flow blocking rate. Moreover, we reduces significantly the mean end-to-end delay (or the mean packet loss rate) and the proposed approach is able to make its decision within 250 msec.

Dynamic Video Transcoding in Mobile Environments

A transcoder with dynamic feedback addresses interactivity, packet loss, and client power constraints in mobile communication systems.

Routing and admission control with multiconstrained end-to-end quality of service in MPLS networks

Multiprotocol label switching (MPLS) networks require dynamic flow admission control to guarantee end-to-end quality of service (QoS) for each Internet protocol (IP) traffic flow. In this paper, we propose to tackle the joint routing and admission control problem for the IP traffic flows in MPLS networks with multiconstrained end-to-end QoS. We propose a mathematical programming model for this problem that includes end-to-end delay and packet loss constraints. These constraints are imposed not only for the new traffic flow, but also for all already admitted flows in the network. Three objective functions are proposed and analyzed: the end-to-end delay, the end-to-end packet loss and the cost of the path used by the new flow. Numerical results show that considering end-to-end delay and packet loss constraints for all flows has a small impact on the flow blocking rate, but improves significantly the end-to-end QoS parameters. Moreover, the proposed approach is able to make the decision to admit or not a new flow in less than 270 ms of computational time.

Two algorithms for multi‐constrained optimal multicast routing

AbstractMultimedia applications, such as video‐conferencing and video‐on‐demand, often require quality of service (QoS) guarantees from the network, typically in the form of minimum bandwidth, maximum delay, jitter and packet loss constraints, among others. The problem of multicast routing subject to various forms of QoS constraints has been studied extensively. However, most previous efforts have focused on special situations where a single or a pair of constraints is considered. In general, routing under multiple constraints, even in the unicast case is an NP‐complete problem. We present in this paper two practical and efficient algorithms, called multi‐constrained QoS dependent multicast routing (M_QDMR) and (multicasting routing with multi‐constrained optimal path selection (M_MCOP)), for QoS‐based multicast routing under multiple constraints with cost optimization. We provide proof in the paper that our algorithms are correct. Furthermore, through extensive simulations, we illustrate the effectiveness and efficiency of our proposals and demonstrate their significant performance improvement in creating multicast trees with lower cost and higher success probability. Copyright © 2003 John Wiley & Sons, Ltd.