We estimate end-to-end service metrics from network device statistics. Our approach is based upon statistical, supervised learning, whereby the mapping from device-level to service-level metrics is learned from observations, i.e., through monitoring the system. The approach enables end-to-end performance prediction without requiring an explicit model of the system, which is different from traditional engineering techniques that use stochastic modeling and simulation. The fact that end-to-end service metrics can be estimated from local network statistics with good accuracy in the scenarios we consider suggests that service-level properties are ''encoded'' in network-level statistics. We show that the set of network statistics needed for estimation can be reduced to a set of measurements along the network path between client and service backend, with little loss in estimation accuracy. The reported work is largely experimental and its results have been obtained through testbed measurements from a video streaming service and a KV store over an OpenFlow network .
In mesh-based peer-to-peer streaming systems data is distributed among the peers according to local scheduling decisions. The local decisions affect how packets get distributed in the mesh, the probability of duplicates and consequently, the probability of timely data delivery. In this paper we propose an analytic framework that allows the evaluation of scheduling algorithms. We consider four solutions in which scheduling is performed at the forwarding peer, based on the knowledge of the playout buffer content at the neighbors. We evaluate the effectiveness of the solutions in terms of the probability that a peer can play out a packet versus the playback delay, the sensitivity of the solutions to the accuracy of the knowledge of the neighbors' playout buffer contents, and the scalability of the solutions with respect to the size of the overlay. We also show how the model can be used to evaluate the effects of node arrivals and departures on the overlay's performance.
The emerging cognitive radio (CR) technology enables the introduction of hierarchical spectrum sharing in wireless networks, where the primary users (PUs) have transmission guarantees, but the coexisting secondary users (SUs) need to be cognitive toward primary activities and adjust their transmissions to conform to the primary constraints. We consider large-scale coexisting primary and secondary networks, where concurrent primary and secondary transmissions are allowed and where the SUs control the interference at the primary receivers by tuning the probability of transmitting and by forming a primary exclusive region (PER) around each primary receiver within which all SUs have to be silent. Moreover, the primary source-destination pairs utilize vertical cooperation by selecting a nearby SU to act as a cooperative relay. We define a unified analytic framework to model cognition and cooperative transmission in large-scale networks. We characterize the achievable gains considering the transmission density region and show that both of the networks have strong incentives to participate in the collaboration.Index Terms-Cognitive radio (CR), cooperative transmission, density region, outage probability, primary exclusion region (PER).
Energy efficiency has been the driving force behind the design of communication protocols for battery-constrained wireless sensor networks (WSNs). The energy efficiency and the performance of the proposed protocol stacks, however, degrade dramatically in case the low-powered WSNs are subject to interference from high-power wireless systems such as WLANs. In this paper we propose COG-MAC, a novel cognitive medium access control scheme (MAC) for IEEE 802.15.4-compliant WSNs that minimizes the energy cost for multihop communications, by deriving energy-optimal packet lengths and single-hop transmission distances based on the experienced interference from IEEE 802.11 WLANs. We evaluate COG-MAC by deriving a detailed analytic model for its performance and by comparing it with previous access control schemes. Numerical and simulation results show that a significant decrease in packet transmission energy cost, up to 66%, can be achieved in a wide range of scenarios, particularly under severe WLAN interference. COG-MAC is, also, lightweight and shows high robustness against WLAN model estimation errors and is, therefore, an effective, implementable solution to reduce the WSN performance impairment when coexisting with WLANs.
In this paper we propose and analyze a generalized multiple-tree-based overlay architecture for peer-to-peer live streaming that employs multipath transmission and forward error correction. We give mathematical models to describe the stability properties of the overlay and evaluate the error recovery in the presence of node dynamics and packet losses. We show how the stability of the overlay improves with the proper allocation of the outgoing bandwidths of the peers among the trees without compromising its error correcting capability.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.