Bandwidth Competition Research Articles

Resource assignment in short video accelerator system

Video accelerator is developed for better user experience in video sharing websites such as YouTube. PPLive video accelerator (PPVA), which has the largest number of users in China, is based on peer-to-peer (P2P) system. The number of videos and peers in PPVA is by orders of magnitude many times larger than which in traditional P2P video on demand (VoD) system. As a result, even though the resource is sufficient, due to unfairness assignment, the quality of service can hardly satisfy all users. In this paper, we concentrate on the assignment of the fundamental resources in PPVA: storage and bandwidth. The problem of storage assignment is formulated as a nonlinear program (NLP) regarding the number of request as a random variable. The results show that the influence of the variance of requests is not negligible and proportional approach is appropriate only when the mean is much larger than the variance. The criteria about how to locate videos to appropriate peers are also presented, taking into account constrains such as the utilization of total bandwidth, the probability of bandwidth competition and the fairness between videos. Furthermore, the heuristic algorithms of allocating upload bandwidth in centralized and distributed fashion are proposed and evaluated against a widely used strategy (equal allocation) with respect to the balance among videos. Simulation results demonstrate that both algorithms can lead to significant performance improvement.

A new perspective on behavioral inconsistency and neural noise in aging: compensatory speeding of neural communication

This paper seeks to present a new perspective on the aging brain. Here, we make connections between two key phenomena of brain aging: (1) increased neural noise or random background activity; and (2) slowing of brain activity. Our perspective proposes the possibility that the slowing of neural processing due to decreasing nerve conduction velocities leads to a compensatory speeding of neuron firing rates. These increased firing rates lead to a broader distribution of power in the frequency spectrum of neural oscillations, which we propose, can just as easily be interpreted as neural noise. Compensatory speeding of neural activity, as we present, is constrained by the: (A) availability of metabolic energy sources; and (B) competition for frequency bandwidth needed for neural communication. We propose that these constraints lead to the eventual inability to compensate for age-related declines in neural function that are manifested clinically as deficits in cognition, affect, and motor behavior.

Bandwidth scheduling and optimization using non-cooperative game model-based shuffled frog leaping algorithm in a networked learning control system

In this paper, under a general framework of Networked two-layer Learning Control Systems (NLCSs), optimal and fair network scheduling is studied. Multi-networked feedback control loops called subsystems in an NLCS share a common communication media. Therefore, there is a competition for the available bandwidth. A non-cooperative game fairness model is first formulated, and then the utility function of subsystems is designed. This takes into account of a number of factors, namely transmission data rate, sampling, the feature of scheduling pattern adopted, and networked control. For the problem defined above, the existence and uniqueness of Nash equilibrium point are proved. Following this, an evolutionary algorithm appeared in the literature recently, shuffled frog leaping algorithm, is improved applying to obtain an optimal solution. The algorithm has a high convergence rate, and the comparison simulation results have demonstrated the effectiveness of the proposed theoretical approach and the algorithm applied.

Improving multi-view peer-to-peer live streaming systems with the divide-and-conquer strategy

Multi-view peer-to-peer (P2P) live streaming systems have recently emerged, where a user is allowed to simultaneously watch one or multiple channels. Previous work on building multi-view P2P streaming systems requires network coding in data block scheduling within each channel, since it solves the fundamental inter-channel bandwidth competition problem at the individual peer level. Therefore, it limits the migration from existing single-view systems to multi-view systems. In this paper, we propose a new protocol, called divide-and-conquer (DAC), which provides a flexible way of building multi-view P2P streaming systems based on existing single-view systems. Specifically, the DAC protocol solves the inter-channel bandwidth competition problem at the channel level based on a divide-and-conquer strategy. Our extensive packet level simulations show that our DAC protocol achieves the three design goals: flexibility, efficiency and scalability.

Efficient scheduling algorithms for mixed services in wireless OFDMA system

Orthogonal Frequency Division Multiple Access (OFDMA) is an attracting system for combating the frequency selective fading. It's widely considered as a promising candidate modulation and access scheme for next generation mobile communication systems. However, the explosive growth of multimedia applications raises higher performance requirements for Radio Resource Management (RRM) in OFDMA system. In particular, limited bandwidth and fierce resource competition impose a challenging issue on the design of packet scheduler. In this paper, we propose two scheduling algorithms applicable to mixed services (i.e., real-time and non-real-time services), that is QoS-oriented Dynamic Threshold Control (DTC) algorithm and fairness-oriented Fairness Aware and QoS Aware (FAQA) algorithm. These two algorithms are both composed of two stages, i.e., initial subcarrier allocation and subcarrier reallocation. For the DTC algorithm, in the stage of initial subcarrier allocation, based on the different metrics to weigh QoS between both services, we design a unique scheduling strategy for each type of service. A dynamic threshold is adopted to help users quickly recover from starvation, so that any one user in system can escape from severely degraded QoS. In the stage of subcarrier reallocation, we will reallocate the surplus subcarriers from the buffer-empty users to the buffer-nonempty users so as to increase spectrum efficiency. For FAQA algorithm, in the stage of initial subcarrier allocation, for the purpose of achieving Proportional Fairness (PF) with lower complexity, we deduce a simple scheduling strategy satisfying PF criterion by means of Greedy algorithm. In the stage of subcarrier reallocation, in order to increase the number of users satisfying QoS requirements, we'll reallocate the surplus subcarriers from the QoS-satisfied users to the QoS-unsatisfied users. Simulation results demonstrate that, on one hand, the DTC algorithm not only guarantees the quality of both services, but also increases the spectrum efficiency; on the other hand, the FAQA algorithm well maintains the fairness among users, and increases the QoS satisfaction degree at the same time.

Dynamic Bandwidth Auctions in Multioverlay P2P Streaming with Network Coding

In peer-to-peer (P2P) live streaming applications such as IPTV, it is natural to accommodate multiple coexisting streaming overlays, corresponding to channels of programming. In the case of multiple overlays, it is a challenging task to design an appropriate bandwidth allocation protocol, such that these overlays efficiently share the available upload bandwidth on peers, media content is efficiently distributed to achieve the required streaming rate, as well as the streaming costs are minimized. In this paper, we seek to design simple, effective, and decentralized strategies to resolve conflicts among coexisting streaming overlays in their bandwidth competition and combine such strategies with network-coding-based media distribution to achieve efficient multioverlay streaming. Since such strategies of conflict are game theoretic in nature, we characterize them as a decentralized collection of dynamic auction games, in which downstream peers bid for upload bandwidth at the upstream peers for the delivery of coded media blocks. With extensive theoretical analysis and performance evaluation, we show that these local games converge to an optimal topology for each overlay in realistic asynchronous environments. Together with network-coding-based media dissemination, these streaming overlays adapt to peer dynamics, fairly share peer upload bandwidth to achieve satisfactory streaming rates, and can be prioritized.

A Multi-channel HD Video Streaming in Digital Community Network: Video Rate Adaptation and Playback Adjustment

In this paper, we propose design and implementation of a practical end-to-end system which implements multi-channel digital television (DTV) multicasting to the home over broadband community access network using sender-driven video rate adaptation and receiver-driven playback adjustment. For video rate adaptation, we employ Program Clock Reference (PCR) embedded in the MPEG-2 transport stream (TS) to enhance packet timing control precision and regulate the transmission rate in a refined way. At the same time, we integrate Traffic Control (TC) ingeniously to separate streams of multiple channels at the network card of DTV server and thus avoid bandwidth competition between them. For playback adjustment, we introduce control theory and over-boundary strategy to reduce buffer jitter and maximize the visual quality of displayed video according to the receiver buffer occupancy. We test our system with standard definition (SD) and high definition (HD) MPEG-2 streams and find that the proposed algorithms can reduce overflow and underflow of sender and receiver buffer, and achieve better video quality and more concurrent channels than traditional solutions.

Modelling and analysis of bandwidth competition in 802.11 wireless networks

In a wireless network, bandwidth is shared and nodes are in fact 'competing' with each other for resources. Therefore, the Quality of Service (QoS) an individual node can offer to applications is determined by how intense the competition is and how much bandwidth it can win in the competition. In this paper, we propose a Markov chain model to study the bandwidth competition in 802.11 networks. A set of heuristic formulas are derived to analyse and predict bandwidth competition. The formulas can be used to compute the exact quantity of bandwidth allocated to competing nodes, given their demands and their traffic parameters. It is illustrated that nodes that demand the same amount of bandwidth may receive very different allocations, and nodes that demand more bandwidth do not always obtain more bandwidth than other nodes. The findings in this study imply that (1) nodes can share the bandwidth precisely according to their demands with full cooperation, (2) or a node can guarantee its own bandwidth while suppressing others with a deliberate consideration in composing packets.

Limited sharing with traffic prediction for dynamic bandwidth allocation and QoS provisioning over Ethernet passive optical networks

Feature Issue on Optical Access Networks (OAN) As an inexpensive, simple, and scalable solution for broadband access, Ethernet passive optical networks (EPONs) have the capability of delivering integrated broadband services to the end users. A critical issue of EPONs is the utility of a shared upstream channel among the local users, and thus an efficient bandwidth allocation mechanism is required to facilitate statistical multiplexing among the local network traffic. In this paper we propose a dynamic bandwidth allocation scheme, i.e., limited sharing with traffic prediction (LSTP), for upstream channel sharing over EPONs. LSTP enables dynamic bandwidth negotiation between the optical line terminal (OLT) and its associated optical network units (ONUs), alleviates data delay by predicting the traffic arrived during the waiting time and prereserving a portion of bandwidth for delivery, and avoids the aggressive bandwidth competition by upper bounding the allocated bandwidth to each ONU. Theoretical analysis and simulation results verify the feasibility of LSTP by showing that LSTP outperforms other existing schemes with respect to QoS metrics of data delay and data loss.

GSC: a generic source-based congestion control algorithm for reliable multicast

This paper presents a simple, generic source-based end-to-end multicast congestion control (GSC) algorithm for reliable multicast transport (RMT) protocols. The algorithm is completely implemented at the source and leverages the reverse control information flow in RMT protocols like PGM or RMTP. Specifically, it does not introduce any new control traffic or new fields in RMT protocol headers. It partially addresses the drop-to-zero problem by introducing a robust, adaptive time-filter based upon RTT (round trip time) estimates collected by observing NAK traffic. This solution allows it to scale for large multicast groups while being adaptive to congestion situation changes in any part of the tree. The algorithm is friendly to TCP in terms of competition for bandwidth shares. The scheme has minimal control traffic requirements and weak RTT estimation requirements.

A method for efficiently executing horn clause programs using multiple processors

Previous investigations have suggested the use of multiple communicating processors for executing logic programs. However, this strategy lacks efficiency due to competition for memory and communication bandwidth, and this is a problem that has been largely neglected. In this paper we propose a realistic model for executing logic programs with low overhead on multiple processors. Our proposal does not involve shared memory or copying computation state between processors. The model organises computations over the nondeterministic proof tree so that different processors explore unique deterministic computation paths independently, in order to exploit the “OR-parallelism” present in a program. We discuss the advantages of this approach over previous ones, and suggest control strategies for making it effective in practice.