Periodic Packet Research Articles

Design and performance evaluation of an improved TCP congestion avoidance scheme

TCP Vegas detects network congestion in its early stage and successfully prevents the periodic packet loss that usually occurs in traditional schemes. It has been demonstrated that TCP Vegas achieves a much higher throughput than TCP Reno. However, TCP Vegas cannot prevent unnecessary throughput degradation when congestion occurs on the backward path. A router-based congestion avoidance scheme for TCP Vegas is proposed. By distinguishing whether or not congestion occurs in the forward path, it significantly improves the connection throughput when the backward path is congested.

End-to-end available bandwidth: Measurement methodology, dynamics, and relation with TCP throughput

The available bandwidth (avail-bw) in a network path is of major importance in congestion control, streaming applications, quality-of-service verification, server selection, and overlay networks. We describe an end-to-end methodology, called self-loading periodic streams (SLoPS), for measuring avail-bw. The basic idea in SLoPS is that the one-way delays of a periodic packet stream show an increasing trend when the stream's rate is higher than the avail-bw. We have implemented SLoPS in a tool called pathload. The accuracy of the tool has been evaluated with both simulations and experiments over real-world Internet paths. Pathload is nonintrusive, meaning that it does not cause significant increases in the network utilization, delays, or losses. We used pathload to evaluate the variability ("dynamics") of the avail-bw in Internet paths. The avail-bw becomes significantly more variable in heavily utilized paths, as well as in paths with limited capacity (probably due to a lower degree of statistical multiplexing). We finally examine the relation between avail-bw and TCP throughput. A persistent TCP connection can be used to measure roughly the avail-bw in a path, but TCP saturates the path and increases significantly the path delays and jitter.

An enhanced congestion avoidance mechanism for TCP Vegas

TCP Vegas detects network congestion in the early stage and successfully prevents periodic packet loss that usually occurs in traditional schemes. It has been demonstrated that TCP Vegas achieves much higher throughput than TCP Reno. However, TCP Vegas cannot prevent unnecessary throughput degradation when congestion occurs in the backward path. In this letter, we propose an enhanced congestion avoidance mechanism for TCP Vegas. By distinguishing whether congestion occurs in the forward path or not, it significantly improves the connection throughput when the backward path is congested.

On the use of out-of-band signaling in ad hoc wireless networks

Ad hoc networks consist of a set of uncoordinated nodes which communicate over wireless links. Many of the existing routing protocols proposed for ad hoc networks have significant routing overhead, which includes periodic beacons, flooded RouteRequests, and other path reconfiguration packets. The overhead depends on the routing protocol, frequency of link breaks, the number of nodes in the network, and coverage area of the network. In general, any stability based routing protocol or QoS enhanced routing protocol needs periodic beacons in order to update stability or QoS information. The high routing overhead normally consumes a significant part of total bandwidth available for communication, causing a decrease in the packet delivery ratio and throughput. In this work, we use out-of-band signaling in which the bandwidth is partitioned into dedicated data and control channels and the control channel is used for control traffic (control packets at the network layer). We study the effects of using the out-of-band signaling on a simple model routing protocol. We also develop an analytical model for evaluating the control overhead generated by this routing protocol. We use out-of-band signaling for a protocol similar to the model routing protocol and study its impact. The results show that even though at light load in-band signaling is better than out-of-band signaling, at high network load out-of-band signaling outperforms in-band signaling. Packet delivery ratio is improved with out-of-band signaling at high mobility and high network load.

End-to-end available bandwidth

The available bandwidth (avail-bw) in a network path is of major importance in congestion control, streaming applications, QoS verification, server selection, and overlay networks. We describe an end-to-end methodology, called Self-Loading Periodic Streams (SLoPS), for measuring avail-bw. The basic idea in SLoPS is that the one-way delays of a periodic packet stream show an increasing trend when the stream's rate is higher than the avail-bw. We implemented SLoPS in a tool called pathload. The accuracy of the tool has been evaluated with both simulations and experiments over real-world Internet paths. Pathload is non-intrusive, meaning that it does not cause significant increases in the network utilization, delays, or losses. We used pathload to evaluate the variability ('dynamics') of the avail-bw in some paths that cross USA and Europe. The avail-bw becomes significantly more variable in heavily utilized paths, as well as in paths with limited capacity (probably due to a lower degree of statistical multiplexing). We finally examine the relation between avail-bw and TCP throughput. A persistent TCP connection can be used to roughly measure the avail-bw in a path, but TCP saturates the path, and increases significantly the path delays and jitter.

Control of reactor pulses by external neutron injection

Abstract It is shown that it is possible to regulate the energy of each pulse of a powerful pulsed periodic reactor using an injector of relatively low power. The target in the reactor core generates the neutron pulses forced by the injector. The correlations for determination of the moment of injection are obtained. The equation connecting the mean reactor power, the intensity integral of the target and the scatter multiplicity of the pulse energy without injection is obtained. In addition to its function as a regulator, the injector plays the role of an auxiliary emergency unit. It is shown that using an injector provides a regime in which the reactor can generate power pulses in the form of periodic packets.

On convergence of wavelet packet expansions

It is well known that the-Walsh-Fourier expansion of a function from the block spaceBq([0,1]), 1<q≤∞, converges pointwise a. e. We prove that the same result is true for the expansion of a function fromBq in certain periodized smooth periodic non-stationary wavelet packets bases based on the Haar filters. We also consider wavelet packets based on the Shannon filters and show that the expansion of Lp-functions, 1<p<∞, converges in norm and pointwise almost everywhere.

Roll waves in a gas–liquid medium

The one‐velocity motion of a gas–liquid medium with a variable mass fraction of the gas phase, which is equilibrium in terms of phase pressures, is considered. The existence conditions of nonlinear periodic wave packets similar in structure to roll waves in open inclined channels are found. The structure of travelling waves in the medium with continuous addition of energy to the gas phase is studied.

Control of reactor pulses by external neutron injection

It is shown that it is possible to regulate the energy of each pulse of a powerful pulsed periodic reactor using an injector of relatively low power. The target in the reactor core generates the neutron pulses forced by the injector. The correlations for determination of the moment of injection are obtained. The equation connecting the mean reactor power, the intensity integral of the target and the scatter multiplicity of the pulse energy without injection is obtained. In addition to its function as a regulator, the injector plays the role of an auxiliary emergency unit. It is shown that using an injector provides a regime in which the reactor can generate power pulses in the form of periodic packets.

SPATIO-TEMPORAL RECONSTRUCTION OF VORTEX DYNAMICS IN AXISYMMETRIC WAKES

With time recording Digital-Particle-Image Velocimetry and spatio-temporal reconstruction technique, we obtained detailed quantitative results of the evolution of the velocity and vorticity field in the wake of axisymmetric bluff bodies—a sphere and an axially oriented cylinder with an elliptic nose and a blunt base. Experiments were carried out for Reynolds numbers of Re=500, 700 and 1000 in the transition range from “regular” to “irregular” shedding. DPIV-recordings in radial cross-sections at several distances downstream of the bodies allowed us to reconstruct the dynamics of the streamwise vorticity over a large number of shedding cycles. Our results prove that the wake in this regime consists of a double-sided chain of oppositely oriented hairpin vortices. In addition, the results show a well-defined low-frequency modulation of the vortex shedding, with a distinct peak in the frequency spectrum at a Strouhal number of about Sr≈0·05 (in case of the sphere). The wake pattern in this “irregular” shedding regime typically exhibits periodic packets of 3–4 “regular” shedding cycles which are interrupted by phases with less action. The results indicate the coexistence of a long-wave instability of axisymmetric wakes against helical waves in addition to the primary instability causing the vortex shedding process.

Vegas plus: improving the service fairness

TCP Vegas is a congestion avoidance scheme designed to prevent the periodic packet loss which occurs in traditional schemes. Since Vegas successfully avoids such packet loss, it achieves much higher throughput than TCP Reno. However, it does not concern the fairness among source-destination pairs with different round-trip times (RTTs). We propose a different mechanism to adjust the window size, this allows TCP to provide much better fairness regardless the large variation of RTTs.

Synaptic origin and stimulus dependency of neuronal oscillatory activity in the primary visual cortex of the cat.

1. We have studied the oscillatory activity of single neurons (91 recorded extracellularly and 76 intracellularly) in the primary visual cortex of cats and kittens to characterize its origins and its stimulus dependency. A new method for the detection of oscillations was developed in order to maximize the range of detectable frequencies in both types of recordings. Three types of activity were examined: spontaneous background activity, responses to intracellular current steps and visual responses. 2. During spontaneous activity, persistent oscillatory activity was very rare in both types of recordings. However, when intracellular records were made using KCl-filled micropipettes, spontaneous activity appeared rhythmic and contained repeated depolarizing events at a variety of frequencies, suggestive of tonic periodic inhibitory input normally masked at resting potential. 3. Patterns of firing activity in response to intracellular current steps allowed us to classify neurons as regular spiking, intrinsically bursting, and fast-spiking types, as described in vitro. In the case of rhythmically firing cells, the spike frequency increased with the amount of injected current. Subthreshold current-induced oscillations were rarely observed (2 out of 76 cells). 4. Visual stimulation elicited oscillations in one-third of the neurons (55 out of 167), predominantly in the 7-20 Hz frequency range in 93% of the cases. Rhythmicity was observed in both simple and complex cells, and appeared to be more prominent at 5 and 6 weeks of age. 5. Intracellular recordings in bridge mode and voltage clamp revealed that visually evoked oscillations were driven by synaptic activity and did not depend primarily on the intrinsic properties of recorded neurons. Hyperpolarizing the membrane led to an increase in the size of the rhythmic depolarizing events without a change in frequency. In voltage-clamped cells, current responses showed large oscillations at the same frequency as in bridge mode, independently of the actual value of the holding potential. 6. In fourteen intracellularly recorded neurons, oscillations consisted of excitatory events that could be superimposed on a depolarizing or a hyperpolarizing slow wave. In two other neurons, visual responses consisted of excitatory and inhibitory events, alternating with a constant phase shift. 7. Drifting bars were much more efficient in evoking oscillatory responses than flashed bars. Except in three cells, the frequency of the oscillation did not depend on the physical characteristics of the stimulus that were tested (contrast, orientation, direction, ocularity and position in the receptive field). No significant correlation was found between the intensity of the visual response and the strength of the rhythmic component. 8. Although it cannot be excluded that the dominant frequency of oscillations might be related to the type of anaesthetics used, no correlation was found between local EEG and the oscillatory activity elicited by visual stimulation. 9. We conclude that the oscillations observed in the present work are generated by synaptic activity. It is likely that they represent an important mode of transmission in sensory processing, resulting from periodic packets of synchronized activity propagated across recurrent circuits. Their relevance to perceptual binding is further discussed.

Analysis of a statistical multiplexer with generalized periodic sources

We provide solution techniques for the analysis of multiplexers with periodic arrival streams, which accurately account for the effects of active and idle periods and of gradual arrival. In the models considered in this paper, it is assumed that each source alternates (periodically) between active and idle periods of fixed durations. Incoming packets are transmitted on the network link and excess information is stored in the multiplexing buffer when the aggregate input rate exceeds the capacity of the link. We are interested in the probability distribution of the buffer content for a given network link speed as a function of the number of sources and their characteristics, i.e., rate and duration of idle and active periods. We derive this distribution from two models: discrete time and continuous time systems. Discrete time systems operate in a slotted fashion, with a slot defining the base unit for data generation and transmission. In particular, in each slot the link is capable of transmitting one data unit and conversely an active source generates one data unit in that time. The continuous time model of the paper falls in the category of fluid models. Compared to previous works we allow a more general model for the periodic packet arrival process of each source. In discrete time, this means that the active period of a source can now extend over several consecutive slots instead of a single slot as in previous models. In continuous time, packet arrivals are not required to be instantaneous, but rather the data generation process can now take place over the entire duration of the active period. In both cases, these generalizations allow us to account for the progressive arrival of source data as a function of both the source speed and the amount of data it generates in an active period.

Eliminating periodic packet losses in the 4.3-Tahoe BSD TCP congestion control algorithm

The congestion control algorithm embedded in the 4.3-Tahoe BSD TCP implementation has dramatically improved congestion control over the Internet. However, several recent simulation studies on the dynamics of this algorithm has revealed that the algorithm exhibits clear oscillatory patterns in sending window size, round trip delay and bottleneck queue length. In this paper, we present a new congestion signal scheme and a dual traffic adjustment strategy. Simulation results show that our modifications can eliminate the periodic packet losses and substantially reduce the traffic oscillation.

Imminent access recognition: A protocol for integrated traffic on a CSMA/CD local area network

A technique is described for reducing the delay of speech packets caused by collisions with other packets on Ethernet. This improvement is achieved by incorporating a contentionfree mechanism during periodic packet transmission. The technique, ‘Imminent Access Recognition’, is easily implemented in hardware as an enhancement to an Ethernet node controller. Simulation models are used to estimate performance and suggest that the lmminent Access Recognition protocol has a significant advantage over CSMA/CD.

ANALYSIS OF SIMULTANEOUS RADIATIVE AND CONDUCTIVE HEAT TRANSFER IN FLUIDIZED BEDS

Abstract In the bubbling regime of operation for fluidized beds, the major mechanism for heat transfer is transient conduction to periodic packets of densely packed particles at the heat transfer surface. The well known Mickley and Fairbanks model, with various subsequently proposed modifications, adequately describes this transient conduction mechanism. However, no adequate theory exists for heat transfer in high-temperature fluidized beds where radiative contribution becomes significant. Analysis of the radiative contribution is complicated by the nonlinear interaction of radiation with conduction/convection. This paper describes a differential formulation of the combined radiative/ conductive heat transfer process. The discrete flux method used by Churchill et al. for radiative transport in heterogeneous media is applied here to the problem of transient heat transfer to packets in fluidized beds. Packets are modeled as radiatively participating media with absorption, scattering, and emission of radiati...