Hierarchically Clustered P2P Streaming System

Size: px
Start display at page:

Download "Hierarchically Clustered P2P Streaming System"

Transcription

1 Hieachically Clusteed P2P Steaming System Chao Liang, Yang Guo, and Yong Liu Polytechnic Univesity Thomson Lab Booklyn, NY Pinceton, NJ Abstact Pee-to-pee video steaming has been gaining populaity. Howeve, it is still challenging to manage a p2p system efficiently to suppot high video playback ate. In this pape, we popose HCPS - Hieachically Clusteed P2P Steaming system that can suppot the steaming ate appoaching the optimum uppe bound with shot delay, yet is simple enough to be implemented in pactice. In HCPS, the pees ae gouped into clustes and a hieachy is fomed among clustes. By actively balancing the uploading capacities among clustes, and executing the pefect scheduling algoithm [1] within each cluste, the system esouce can be fully utilized. The simulation expeiments diven by the taces collected fom a eal p2p steaming system demonstate the effectiveness of HCPS. I. INTRODUCTION Video-ove-IP applications, e.g. Youtube [2], have ecently attacted a lots of uses on the Intenet. In addition to the taditional seve-client based o content distibution netwoks based solutions, Pee-to-Pee (P2P) video steaming solution utilizes the uploading bandwidth of end uses to distibute video content at low distibution cost. Seveal P2P based systems have been deployed to povide on-demand o ealtime video steaming ove the Intenet [3], [4], [5], [6]. Ou ecent measuement study [7], [8] on one popula P2P steaming system ecoded ove 200, 000 simultaneous uses watching a live boadcast of an event duing 2006 Chinese New Yea Eve. We expect to see moe P2P steaming systems and applications on the Intenet in nea futue. Accoding to the ovelay stuctue, P2P steaming systems can be boadly classified into two categoies: teebased and mesh-based. The tee-based appoach, such as ESM [3], has well oganized ovelay stuctue and typically distibutes video by actively pushing data fom a pee to its childen pees. One majo dawback of tee-based steaming systems is thei vulneability to pee chuns. A pee depatue will inteupt video delivey to all its descendant pees in the subtee ooted at the depated pee. In a mesh-based P2P steaming system, pees ae not confined to a static topology. Instead, the peeing elationship ae established/ton down based on the content availability and bandwidth availability on pees. A pee dynamically connect to a subset of andom pees in the system. Pees peiodically exchange infomation about thei data availability. Video chunk is pulled by a pee fom its neighbo who has aleady obtained that chunk. Thee ae a lot of mesh-based P2P steaming systems deployed ecently, such as CoolSteaming [4], PPLive [7]. Those systems ae vey obust to pee chuns. Howeve, the dynamic peeing elationships make the video distibution efficiency unpedictable. Consequently, uses suffe fom the video playback quality degadation, such as low video bit ates, long statup delays and fequent feezing. Ou measuement esults on PPLive [8] show that most pogams have bit ates aound 400bps and the stat-up delays fo a channel ange fom a few second to a few minutes. This makes them unsuitable fo pogams with high video bit ate and ealtime playback equiement. In this pape, we popose a novel P2P steaming achitectue that combines the best of tee-based and mesh-based systems to suppot high data ate video steaming with shot delays. The key to achieve this goal is to oganize pees in clustes such that the uploading bandwidth of pees can be efficiently utilized to edistibute video quickly to all pees. Due to bandwidth and content bottleneck [9], it is quite challenging to sustain high steaming ate fo all pees in a P2P steaming system. As studied in [1], the maximum video steaming ate in a P2P steaming system is detemined by the video souce seve s capacity, the numbe of the pees in the system, and the aggegate uploading capacity of all pees. A pefect scheduling algoithm is poposed to achieve the maximum steaming ate. In such a scheduling algoithm, each pee uploads the video content obtained diectly fom the seve to all othe pees in the system. To guaantee 100% uploading capacity utilization on all pees, diffeent pees download diffeent content fom the seve and the ate at which a pee downloads content fom the seve is popotional to its uploading capacity.

2 Figue 1 shows an example how the diffeent potions of data ae scheduled among thee heteogenous nodes with pefect scheduling algoithm. Thee ae thee pees in the system. The seve has capacity of 6. The upload capacities of a, b and c ae 2, 4 and 6 espectively. Suppose pees all have enough downloading capacity, the maximum video ate can be suppoted in the system is 6. To achieve that ate, the seve divides video chunks into goups of 6. a is esponsible fo uploading 1 chunk out of each goup while b and c ae esponsible fo upload 2 and 3 chunks within each goup. This way, all pees can download video at the maximum ate of 6. To implement S c a b Fig. 1. Pefect Scheduling Example such a pefect scheduling algoithm, each pee needs to maintain a connection and exchange video content with all othe pees in the system. In addition, the seve needs to split the video steam into multiple sub-steams with diffeent ates, one fo each pee. A eal P2P steaming system can easily have a few thousand of pees. With cuent opeating systems, it is unealistic fo a egula pee to maintain thousands of concuent connections. It is also challenging fo a seve to patition a video steam into thousands of sub-steams in ealtime. To addess this poblem, we popose HCPS - Hieachically Clusteed P2P Steaming system, to appoach the theoetical maximum video steaming ate with ealistic pee setting. In the poposed system, at the base level, all pees ae oganized into bandwidth-balanced clustes. Each cluste consists of a cluste head and a small numbe e.g , nomal pees. Cluste heads of all clustes fom a supe cluste at uppe level and etieve video fom the video souce seve in P2P fashion. Afte obtaining video at the uppe level, a cluste head acts as a local video souce seve fo nomal pees in its cluste at lowe level. Nomal pees within the same cluste collaboate accoding to the pefect scheduling algoithm to etieve video fom thei cluste head. In this new achitectue, the seve only distibutes video to cluste heads; a nomal pee only maintain connections to its neighbos in the same cluste; and a cluste head connects to both othe cluste heads and nomal pees in its own cluste. Two-level HCPS aleady has the ability to suppot a lage numbe of pees with mild equiement on the numbe of connections on the seve, cluste heads and nomal pees. By employing pefect scheduling at both uppe and lowe level, HCPS can appoach the theoetical steaming ate uppe bound. HCPS has mesh topologies at both levels, which makes it obust to pee chuns. Cluste heads play impotant ole in bidging the two levels. The depatue of a cluste head only affect a small numbe of pees in its cluste. As will be shown in Section II-B, HCPS cluste management mechanism ensues the quick ecovey fom cluste head depatues. In addition, in a basic configuation of HCPS, video data can each any pee within two hops. This lagely impove the delay pefomance upon nomal mesh-based solutions. The pape is oganized as follows. In Section II, we fist fomulate the maximal suppotable steaming ate poblem and deduce the pee clusteing stategy. Then we descibe in details the system achitectue and the pee management scheme. In Section III simulation esults ae pesented. In Section IV the conclusion and the futue wok ae included. II. HCPS: HIERARCHICALLY CLUSTERED P2P STREAMING SYSTEM As descibed in Section I, pefect scheduling achieves the maximum steaming ate by constucting a fully connected mesh among all paticipating pees and the souce. A pee then elays the eceived content fom the

3 souce to othe pees in the system. The tansmitting ate fom the souce to pee is calculated using max = min{u s, u s + n i=1 u i } (1) n whee u s efes to the upload bandwidth of seve and u i efes to the bandwidth of the ith node of total n nodes. Fo a p2p steaming system with a lage numbe of pees, the out-degee of pees becomes pohibitively lage. Hence the pefect scheduling algoithm does not scale well as the numbe of pees inceases. In this pape, we popose the hieachically clusteed p2p steaming (HCPS) scheme to addess the scalability issue faced by pefect scheduling algoithm. Instead of foming a single lage mesh, HCPS goups the pees into clustes. The numbe of pees in a cluste is elatively small so that the pefect scheduling can be successfully applied at the cluste level. One pee in a cluste is elected as the cluste head and woks as the souce fo this cluste. The cluste heads eceive the steaming content by joining anothe cluste that is one level above its own cluste in the system hieachy. Clustes fom a tee-like topology with clustes as vetices. The cluste heads seve as the links that connect two clustes. The steaming content is pushed fom the uppe-level clustes to the lowe-level clustes. S Cluste Heade Head Mapping 1 2 a1 a3 b1 b3 a1 a2 b2 b3 a3 b1 3 a2 b Fig Hieachically Clusteed P2P Steaming System 7 Figue 2 illustates a simple example of two-level HCPS hieachy. At the base level, pees ae gouped into small size clustes. The pees ae fully connected within a cluste. One pee is elected as the cluste head. At the top level, all cluste heads and the video seve foms two clustes. Video seve (souce) distibutes the content to all cluste heads using the pefect scheduling algoithm at the top level. At the base level, each cluste head acts as the video seve in its cluste and distibutes the downloaded video to othe pees in the same cluste, again, using the pefect scheduling algoithm. The numbe of connections on each nomal pee is bounded by the size of its cluste. Cluste heads additionally maintain connections in the uppe level cluste, i.e., the numbe of connections fo cluste heads could be doubled. Suppose that the cluste size is bounded by N max, and the souce can suppot up to N s top laye clustes. The two-laye HCPS system, as shown in Fig. 2, can accommodate up to N s (N max ) 2 pees. Assume N s = 10 and N max = 20, HCPS suppots up to 4,000 pee. The maximum numbe of connections a pee needs to maintain is 40 fo cluste head and 20 fo nomal pee, which is quite manageable. Moe pees can be accommodated by adding moe levels into the hieachy. A. Pee clusteing stategy The key question in designing HCPS is how to cluste the pees so that the suppotable steaming ate can be maximized. The maximum steaming ate, max, fo a given set of pees and the souce is detemined by Equation (1), and can be achieved using the pefect scheduling algoithm using fully connected mesh. The mesh constucted in HCPS is not fully connected, which may educe the maximum suppotable steaming ate. In the following, we investigate the pee clusteing stategies that would allow HCPS to suppot the steaming ate close to max. We fist fomulate the maximum suppotable steaming ate fo a given HCPS mesh topology as an optimization

4 poblem. We then discuss heuistic pee clusteing stategy that allows HCPS to have good suppotable steaming ate. 1) Maximal suppotable steaming ate fo a given mesh topology: Assume thee ae C clustes, N pees, and one souce in the HCPS mesh. Cluste c has V c pees, c, c [1, C]. Denote by u i the pee i s upload capacity. A pee can paticipate in a HCPS cluste eithe as a nomal pee o as a cluste head. Fo instance, in Fig. 2, pee a1 joins two clustes. It is the cluste head in cluste 3 and a nomal pee in cluste 1. Denote by u ic the amount of upload capacity of pee i contibuted to cluste c as a nomal pee, and by h ic the amount of upload capacity of pee i contibuted to cluste c as a cluste head. Hence u (a1)3 = 0, h (a1)3 0 since pee a1 is the head in cluste 3; and u (a1)1 0, h (a1)1 = 0 since pee a1 is a nomal pee in cluste 1. u (a1)c = h (a1)c = 0 fo all othe clustes since pee a1 is not thei membe. Futhe denote by u s the total souce upload capacity; and by u s c the amount of souce capacity used fo cluste c. Similaly, u s c = 0 if the souce does not paticipate in cluste c. If c max epesents the maximum steaming ate fo cluste c using pefect scheduling algoithm, the maximum suppotable steaming ate fo a given cluste-based HCPS mesh, HCP S, can be fomulated as following optimization poblem. subject to: HCP S = max c { max min [ max c c = 1, 2,..., C] } (2) {u ic,h ic,u s c} = min { N i=1 (u ic + h ic ) V c, N i=1 h ic + u s } c C (u ic + h ic ) u i (4) c=1 C u s c u s (5) c=1 whee Eqn. (3) is tue fo all c [1, C] and Eqn. (4) is tue all fo i [1, N]. The value of u ic, h ic, and u s c is detemined by the HCPS topology. The maximum suppotable steaming ate fo a given mesh topology is the the steaming ate that can be suppoted by all clustes. Since the cluste head paticipates in both uppe laye and lowe laye clustes and the souce s upload capacity is used by seveal top laye clustes, the suppotable steaming ate fo HCPS can be maximized by adjusting the allocation of clustes upload capacity and souce s upload capacity. This explains the Equation (2). The fist tem in Equation (3) epesents the aveage upload capacity pe pee in cluste c; and the second tem epesents the cluste head s upload capacity (note that cluste head can be eithe the souce seve o a pee). Since the maximum value of steaming ate at cluste c, c max is govened by the theoetical uppe bound (see Equation (1)), this leads to the Equation (3). Futhe, the amount of bandwidth allocated fo the uppe laye and low laye clustes must not supass its total upload capacity, which explains Equation (4). Finally, fo the souce, the total allocated upload capacity fo all clustes must not supass the souce s total upload capacity, which explains Equation (5). 2) Pee clusteing heuistics: The goal is to have a HCPS mesh topology that suppots the steaming ate close to the optimal ate max. In the following, we use numeical analysis to examine the majo factos that affects the HCPS s pefomance. Assume thee ae 400 pees with one souce node. The cluste size is set to be 20. The pees ae gouped into 20 base laye clustes and one top laye cluste fo cluste heads. The maximum suppotable steaming ate fo HCPS is computed accoding to the optimization poblem as fomulate in Equation (2). The pees upload capacity obeys the distibution as descibed in Table I. The distibution in Table I is fom the measuement study conducted in [10]. We fist assign bandwidth andomly to the 20 nodes of one cluste accoding to the bandwidth distibution, then copy the allocation to othe clustes. Hence all clustes have the homogeneous bandwidth allocation and the node with lagest upload bandwidth is selected as the cluste head. We solve the linea pogamming poblem. (3)

5 TABLE I BANDWIDTH DISTRIBUTION Uplink(kbps) Faction of nodes The solution shows that the value of HCP S is vey close to the theoetical uppe bound (> 99%). In contast, when each node has i.i.d. upload bandwidth accoding to the pobability distibution shown in Table I, HCP S only achieves oughly 60% of the theoetical uppe bound. In the fist scenaio, the clustes ae pefectly balanced (they ae actually the same); while in the second scenaio, they ae not. Accoding to Equation (2), the maximum suppotable steaming ate, HCP S, takes the minimum cluste steaming ate among all clustes. The cluste steaming ate (Equation (3)) is the minimum of cluste aveage upload capacity and the cluste head s ate. Intuitively, the pees should be divided into clustes with equal (simila) aveage upload capacity to avoid wasting esouces. Based on the above discussion, we popose following heuistics: The discepancy of individual clustes aveage upload capacity pe pee should be minimized. The cluste head s upload capacity should be as lage as possible. The cluste head s capacity allocated fo the base laye capacity has to be lage than the aveage upload capacity to avoid being the bottleneck. Futhemoe, the cluste head also joins the uppe laye cluste. Ideally, the cluste head s ate should be 2 HCP S. The numbe of pees in a cluste should be bounded fom the above by a elatively small numbe. The numbe of pees in a cluste detemines the out-degee of pees. A lage size cluste pohibits a cluste fom pefoming popely using pefect scheduling. Due to the pee dynamics, i.e., the pees join and leave the system all the time, HCPS mesh needs to be dynamically adjusted to have consistent high suppotable steaming ate. Below we descibe the dynamic pee management in HCPS that implements the above heuistics. B. Dynamic pee management HCPS system has a bootstap node whose task is to manage HCPS topology to balance the esouces among clustes. Meanwhile, a cluste head manages the pees in the cluste it belongs to. Due to the pee chun, the aveage upload capacity pe pee at cluste-level vaies ove time. Hence the suppotable steaming ates of clustes change and the system s suppotable steaming ate, HCP S, may decease. The bootstap node s task is to handle the new pee aival and pee depatue in such a way that HCP S is maintained at the highest possible level. In addition, because of the unpedictability of pee chun, the clustes steaming ates may still dift apat thus lowe down the system steaming ate. The bootstap node peiodically moves the pees aound to achieve good esouce balancing. Below we descibe thee key opeations in HCPS that handles the pee dynamics: pee join, pee depatue, and cluste e-balancing. 1) Pee join: Depending on the new aival s upload capacity compaed to the cuent HCP S, the new pee is classified into thee categoies: HPee(esouce ich pee) if u HCP S + δ, MPee(esouce medium pee) if HCP S δ < u < HCP S + δ, and LPee(esouce poo pee) othewise, whee u is new pee s upload capacity, and δ is a configuation paamete. Denote by N max the maximum numbe of nodes allowed by a cluste. All clustes whose numbe of pees is less than N max ae eligible to accept the new pee. If the upload capacity of the new pee, u, is geate than some eligible cluste heads upload capacity by a magin β, the pee is assigned to the cluste with the smallest cluste head upload capacity. The new pee is to eplace the oiginal cluste head, and the oiginal head becomes the nomal pee and stay in the cluste. If the new pee does not eplace any cluste head, it is assigned to a cluste accoding to the pee type. Specifically, the pee is assigned to the cluste with the minimum aveage upload capacity if the pee is HPee; the pee is assigned

6 to the cluste with the smallest numbe of pees if it is MPee; and the pee is assigned to the cluste with maximum aveage upload capacity if it is LPee. The idea behind this is to balance the upload esouces among clustes.the new pee is ediected to the coesponding cluste head, and bootstap node asks the cluste head to admit the new pee. In case the cluste is full afte the new pee joins, the cluste will split itself into two clustes and then infom the bootstap node. 2) Pee depatue: The handling of pee depatue is staight fowad. If the pee is a nomal pee, it infoms the cluste head of its depatue. The cluste head takes the pee off its cluste membe list, and infoms othe pees its depatue. In case the depating pee is the cluste head, the pee infoms the bootstap node its depatue. The bootstap node selects one pee fom existing pees in the cluste as the new cluste head. The eplacement pocess follows the same pocedue as we descibed in the above section. If a pee is cashed, the handling is the same fo the nomal pee. If the cluste head cashes, a pee in the cluste can infom the bootstap node. The bootstap node selects the pee with lagest upload capacity as the new cluste head. The bootstap node behaves as the cluste head, and let the selected pee to eplace itself to become the new cluste head. 3) Cluste e-balancing: The clustes may lose balance in tems of the numbe of pees and the amount of esouces in a cluste as the esult of pee chun. In HCPS, the bootstap node peiodically attempts to e-balance the clustes. At the end of an epoch, the bootstap node fist attempts to balance the cluste sizes. The clustes ae soted in the descending ode of cluste size. If the gap between the clustes with the lagest and the smallest numbe of pees is geate than the theshold = max{α N max, β N}, whee N is the aveage cluste size, these two clustes will be meged and then splitted into two balanced clustes. The above pocess continues until no clustes violates the condition. In the second phase of cluste e-balancing, the bootstap node ties to balance the esouces. The clustes ae soted in the descending ode of aveage upload capacity pe pee. If the aveage upload capacity diffeence of the clustes with highest and lowest upload capacity is geate than the theshold of θ u, whee u is the system aveage upload capacity, these two clustes will be meged and then splitted into two balanced clustes. 4) Cluste mege: The bootstap node infoms the two cluste heads the decision of mege, and indicate which cluste is to be meged. The meged cluste head un-egistes itself fom uppe laye cluste, and sends the membe list to the new cluste head. The new cluste head infoms all pees in the new cluste the membe list. Connections ae set up among pees. The new cluste head also e-calculates the sending ate using pefect scheduling. 5) Cluste split: The goal of cluste split is to divide the pees into two clustes that have oughly the simila numbe of pees and simila aveage upload capacity pe pee. The cuent cluste head manages the entie pocess. The cluste head sots the pees in the descending ode of upload capacity. The pee with the lagest upload capacity become the cluste head of the emeging new cluste. In the following ounds, two pees at the top of soted list ae taken out of the list. The one with the lage upload capacity is assigned to the cluste with smalle aggegate upload capacity; and the pee with smalle upload capacity is assigned to the cluste with lage aggegate upload capacity. The pocess continues until all pees ae assigned. If thee is only one pee in the last ound, then the pee is assigned to the cluste with moe aggegate upload capacity. Once the cluste membeship is decided, the new cluste head ceates the new cluste. It egistes itself into the uppe laye cluste, and then boadcasts the membe list to all pees. The cluste head computes the sending ate to each pee using the pefect scheduling. The oiginal cluste head also infoms the pees emaining in the cluste of the new membe list, and computes the new sending ate. C. System Delay Analysis The above pee dynamic management gives the HCPS self-adaptive ability to balance membe clustes in ode to achieve high steaming ate. We want to emphasize anothe inspiing meit of HCPS which is supeio to othe P2P steaming systems. Assume one node i needs to spend t s tansmission delay to delive one specified chunk to othe pee, then by pefect scheduling algoithm the last pee in the top level to eceive the data chunk will wait t p + N max t s time, whee t p is the popagation delay fom the seve to node i. Given the designed small numbe

7 N max and conventional popagation and tansmission delay, the delay in one cluste can be quite small. To the two-level HCPS system, the data chunk will encounte at most two such above pocesses to each all the pees of the system, because the pocess fo the nomal pee in the base level fetching data fom cluste head is simila as that of head fetching data fom seve. III. SIMULATION RESULTS In this section we use the tace-diven simulation to demonstate that HCPS is able to achieve the high suppotable steaming ate. A flow level simulato is developed to evaluate the system pefomance. The simulato models the dynamic pee management and peiodic cluste e-balancing exactly. Howeve, the data taffic is simulated at flow level. Such abstaction speeds up the simulation and still gives us accuate steaming ate of system. The simulation is diven by the tace collected fom the measuement study of PPLive, a lage scale p2p live steaming system [7]. We extact the infomation of pees aival and thei life time fom a one-day PPLive channel tace dumped at Apil 3, The upload capacity of pees ae assigned andomly accoding to the distibution of Table I. We choose the peiod of cluste e-balancing to be one minute and the seve upload capacity is set to be 2Mbps. Fig. 3 depicts the numbe of pees in the system ove the peiod of 24 hous. Pee chun is evident in this figue. The Node Degee Fig Time(h) One-day Degee Evolution of Popula Channel numbe of pees peaks aound 9:00AM-12:00PM when the system has moe than 200 concuent uses. 1) HCPS suppotable steaming ate: Fig. 4 plots the HCPS steaming ate vs. theoetical uppe bound using pefect scheduling algoithm duing the peak hou (9:00AM-12:00PM) with two e-balancing theshold settings. Both cuves of HCPS steaming ate follow the tend of pefect scheduling steaming ate closely. The oscillation of pefect scheduling steaming ate is due to the upload capacity change caused by pee chun. The oscillation of HCPS steaming ate is due to the following two easons: (i) the upload capacity change due to pee chun; and (ii) the esouce imbalance among HCPS clustes due to chun. While the fist facto is caused by the natue of p2p system, the second facto can be mitigated by adjusting the e-balancing configuation paametes. Fo instance, in Fig. 4(a), the e-balancing paametes (see Section II-B.3) ae set to be α = 0.2, β = 0.4 and θ = 0.4; while the paametes used in Fig. 4(b) ae α = 0.15, β = 0.3 and θ = 0.3. Tighte theshold helps HCPS to impove the steaming ate a little bit as shown in Fig. 4(b). Howeve, the tighte theshold inceases the e-balancing ovehead. In pactice, the paametes have to be caefully toned to stike the ight balance between management ovehead and system pefomance. To bette undestand the pesistent pefomance, Table II epots the aveage steaming ate ove a fou hou peiod duing the day. In the table, the e-balancing theshold setting of HCPS is the same as in Fig. 4(a), so ae the following simulation esults. The HCPS steaming ate is within 90% of the theoetical uppe bound all the time. 2) Pefomance in suppoting fixed ate steaming: Constant-bit-ate (CBS) coding is typically used in on-line steaming sevice. In this section, we examine how HCPS pefoms in suppoting CBS content. Fig. 5 depicts the faction of uses that can watch the content at the given ate. Almost all uses in HCPS can steam video up to

8 HCPS Pefect HCPS Pefect Playback Rate Playback Rate Time(m) Time(m) (a) α = 0.2, β = 0.4 and θ = 0.4 (b) α = 0.15, β = 0.3 and θ = 0.3 Fig. 4. Rate Evolution in Peak Time TABLE II AVERAGE RATE PER TIME ZONE Time Zone(h) Pefect(kbps) HCPS(kbps) kbps. With the cuent encoding technology, the on-line content uses steaming ate in the ange of 400 kbps to 800 kpbs. Hence HCPS is able to suppot most uses in this ange. As the steaming ate inceases futhe, less uses can sustain full ate. The faction of full-ate uses deceases quickly as shown in Fig. 5. This is because HCPS maximizes the steaming ate by balancing the suppotable steaming ate of individual clustes. Once the fixed steaming ate supasses HCPS suppotable ate, most of clustes steaming ates become smalle than the fixed ate, which explains the apid decease of numbe of uses that can suppot full ate. Howeve, the faction of content eceived by uses deceases moe gadually as shown in Fig. 6. Table III futhe gives the esults of aveage faction of eceived content fo diffeent CBS steaming ate using HCPS and pefect scheduling. Again, HCPS appoaches the pefomance of pefect scheduling algoithm. Above esults suggest that TABLE III AVERAGE RATIO UNDER DIFFERENT PLAYBACK RATE Fixed Playate(kbps) Pefect(%) HCPS(%) using advanced coding technique, such as laye coding o multiple-desciption coding, may be a good way to combat the tempoay lack of upload esouces in system. IV. CONCLUSIONS In this pape, we poposed hieachically clusteed p2p steaming (HCPS) famewok that is capable of fully utilizeing uses upload capacity and appoaching the maximum steaming ate allowed by the system. The small size of cluste enables the use of pefect scheduling algoithm; and the hieachy makes the accommodation of lage numbe of uses possible. We fomulate the maximum steaming ate in HCPS as an optimization poblem, and popose distibuted potocol to appoximate the optimal solution. As the esults fom tace-diven simulation demonstate, the aveage suppotable steaming ate by HCPS is within 10% of the maximum steaming ate. As to futue wok, we ae inteested in implementing HCPS and testing it out in the eal wold. We ae also looking into ways to futhe simplify the implementation of pefect scheduling algoithm.

9 Pecentage of Suppotable Nodes Fig. 5. Time(m) Pecentage of suppotable nodes in HCPS 1 Faction of Received Content Fig Time(m) Aveage faction of eceived content in HCPS REFERENCES [1] R. Kuma, Y. Liu, and K. Ross, Stochastic Fluid Theoy fo P2P Steaming Systems, in Poceedings of IEEE INFOCOM, [2] Youtube, Youtube Homepage, [3] Y.-H. Chu, S. G.Rao, and H. Zhang, A case fo end system multicast, in Poceedings of ACM SIGMETRICS, [4] X. Zhang, J. Liu, B. Li, and T.-S. P. Yum, DONet/CoolSteaming: A data-diven ovelay netwok fo live media steaming, in Poceedings of IEEE INFOCOM, [5] PPLive, PPLive Homepage, [6] PPSteam, PPSteam Homepage, [7] X. Hei, C. Liang, J. Liang, Y. Liu, and K.W.Ross, Insights into PPLive: A measuement study of a lage-scale P2P IPTV system, in Poc. Wokshop on Intenet Potocol TV (IPTV) sevices ove Wold Wide Web in conjunction with WWW2006, [8] X. Hei, C. Liang, J. Liang, Y. Liu, and K. Ross, A Measuement Study of a Lage-Scale P2P IPTV System, IEEE Tansactions on Multimedia, [9] C. Gkantsidis and P. R. Rodiguez, Netwok Coding fo Lage Scale Content Distibution, in Poceedings of IEEE INFOCOM, [10] C. H. Ashwin R. Bhaambe and V. N. Padmanabhan, Analyzing and Impoving a BitToent Netwok Pefomance Mechanisms, in Poceedings of IEEE INFOCOM, 2006.

IP Network Design by Modified Branch Exchange Method

IP Network Design by Modified Branch Exchange Method Received: June 7, 207 98 IP Netwok Design by Modified Banch Method Kaiat Jaoenat Natchamol Sichumoenattana 2* Faculty of Engineeing at Kamphaeng Saen, Kasetsat Univesity, Thailand 2 Faculty of Management

More information

IP Multicast Simulation in OPNET

IP Multicast Simulation in OPNET IP Multicast Simulation in OPNET Xin Wang, Chien-Ming Yu, Henning Schulzinne Paul A. Stipe Columbia Univesity Reutes Depatment of Compute Science 88 Pakway Dive South New Yok, New Yok Hauppuage, New Yok

More information

SCALABLE ENERGY EFFICIENT AD-HOC ON DEMAND DISTANCE VECTOR (SEE-AODV) ROUTING PROTOCOL IN WIRELESS MESH NETWORKS

SCALABLE ENERGY EFFICIENT AD-HOC ON DEMAND DISTANCE VECTOR (SEE-AODV) ROUTING PROTOCOL IN WIRELESS MESH NETWORKS SCALABL NRGY FFICINT AD-HOC ON DMAND DISTANC VCTOR (S-AODV) ROUTING PROTOCOL IN WIRLSS MSH NTWORKS Sikande Singh Reseach Schola, Depatment of Compute Science & ngineeing, Punjab ngineeing College (PC),

More information

THE THETA BLOCKCHAIN

THE THETA BLOCKCHAIN THE THETA BLOCKCHAIN Theta is a decentalized video steaming netwok, poweed by a new blockchain and token. By Theta Labs, Inc. Last Updated: Nov 21, 2017 esion 1.0 1 OUTLINE Motivation Reputation Dependent

More information

Prioritized Traffic Recovery over GMPLS Networks

Prioritized Traffic Recovery over GMPLS Networks Pioitized Taffic Recovey ove GMPLS Netwoks 2005 IEEE. Pesonal use of this mateial is pemitted. Pemission fom IEEE mu be obtained fo all othe uses in any cuent o futue media including epinting/epublishing

More information

AN ANALYSIS OF COORDINATED AND NON-COORDINATED MEDIUM ACCESS CONTROL PROTOCOLS UNDER CHANNEL NOISE

AN ANALYSIS OF COORDINATED AND NON-COORDINATED MEDIUM ACCESS CONTROL PROTOCOLS UNDER CHANNEL NOISE AN ANALYSIS OF COORDINATED AND NON-COORDINATED MEDIUM ACCESS CONTROL PROTOCOLS UNDER CHANNEL NOISE Tolga Numanoglu, Bulent Tavli, and Wendi Heinzelman Depatment of Electical and Compute Engineeing Univesity

More information

Performance Optimization in Structured Wireless Sensor Networks

Performance Optimization in Structured Wireless Sensor Networks 5 The Intenational Aab Jounal of Infomation Technology, Vol. 6, o. 5, ovembe 9 Pefomance Optimization in Stuctued Wieless Senso etwoks Amine Moussa and Hoda Maalouf Compute Science Depatment, ote Dame

More information

On the Forwarding Area of Contention-Based Geographic Forwarding for Ad Hoc and Sensor Networks

On the Forwarding Area of Contention-Based Geographic Forwarding for Ad Hoc and Sensor Networks On the Fowading Aea of Contention-Based Geogaphic Fowading fo Ad Hoc and Senso Netwoks Dazhi Chen Depatment of EECS Syacuse Univesity Syacuse, NY dchen@sy.edu Jing Deng Depatment of CS Univesity of New

More information

Towards Adaptive Information Merging Using Selected XML Fragments

Towards Adaptive Information Merging Using Selected XML Fragments Towads Adaptive Infomation Meging Using Selected XML Fagments Ho-Lam Lau and Wilfed Ng Depatment of Compute Science and Engineeing, The Hong Kong Univesity of Science and Technology, Hong Kong {lauhl,

More information

Journal of World s Electrical Engineering and Technology J. World. Elect. Eng. Tech. 1(1): 12-16, 2012

Journal of World s Electrical Engineering and Technology J. World. Elect. Eng. Tech. 1(1): 12-16, 2012 2011, Scienceline Publication www.science-line.com Jounal of Wold s Electical Engineeing and Technology J. Wold. Elect. Eng. Tech. 1(1): 12-16, 2012 JWEET An Efficient Algoithm fo Lip Segmentation in Colo

More information

Scaling Location-based Services with Dynamically Composed Location Index

Scaling Location-based Services with Dynamically Composed Location Index Scaling Location-based Sevices with Dynamically Composed Location Index Bhuvan Bamba, Sangeetha Seshadi and Ling Liu Distibuted Data Intensive Systems Laboatoy (DiSL) College of Computing, Geogia Institute

More information

RANDOM IRREGULAR BLOCK-HIERARCHICAL NETWORKS: ALGORITHMS FOR COMPUTATION OF MAIN PROPERTIES

RANDOM IRREGULAR BLOCK-HIERARCHICAL NETWORKS: ALGORITHMS FOR COMPUTATION OF MAIN PROPERTIES RANDOM IRREGULAR BLOCK-HIERARCHICAL NETWORKS: ALGORITHMS FOR COMPUTATION OF MAIN PROPERTIES Svetlana Avetisyan Mikayel Samvelyan* Matun Kaapetyan Yeevan State Univesity Abstact In this pape, the class

More information

A Recommender System for Online Personalization in the WUM Applications

A Recommender System for Online Personalization in the WUM Applications A Recommende System fo Online Pesonalization in the WUM Applications Mehdad Jalali 1, Nowati Mustapha 2, Ali Mamat 2, Md. Nasi B Sulaiman 2 Abstact foeseeing of use futue movements and intentions based

More information

ANALYTIC PERFORMANCE MODELS FOR SINGLE CLASS AND MULTIPLE CLASS MULTITHREADED SOFTWARE SERVERS

ANALYTIC PERFORMANCE MODELS FOR SINGLE CLASS AND MULTIPLE CLASS MULTITHREADED SOFTWARE SERVERS ANALYTIC PERFORMANCE MODELS FOR SINGLE CLASS AND MULTIPLE CLASS MULTITHREADED SOFTWARE SERVERS Daniel A Menascé Mohamed N Bennani Dept of Compute Science Oacle, Inc Geoge Mason Univesity 1211 SW Fifth

More information

Detection and Recognition of Alert Traffic Signs

Detection and Recognition of Alert Traffic Signs Detection and Recognition of Alet Taffic Signs Chia-Hsiung Chen, Macus Chen, and Tianshi Gao 1 Stanfod Univesity Stanfod, CA 9305 {echchen, macuscc, tianshig}@stanfod.edu Abstact Taffic signs povide dives

More information

An Unsupervised Segmentation Framework For Texture Image Queries

An Unsupervised Segmentation Framework For Texture Image Queries An Unsupevised Segmentation Famewok Fo Textue Image Queies Shu-Ching Chen Distibuted Multimedia Infomation System Laboatoy School of Compute Science Floida Intenational Univesity Miami, FL 33199, USA chens@cs.fiu.edu

More information

An Improved Resource Reservation Protocol

An Improved Resource Reservation Protocol Jounal of Compute Science 3 (8: 658-665, 2007 SSN 549-3636 2007 Science Publications An mpoved Resouce Resevation Potocol Desie Oulai, Steven Chambeland and Samuel Piee Depatment of Compute Engineeing

More information

Dynamic Topology Control to Reduce Interference in MANETs

Dynamic Topology Control to Reduce Interference in MANETs Dynamic Topology Contol to Reduce Intefeence in MANETs Hwee Xian TAN 1,2 and Winston K. G. SEAH 2,1 {stuhxt, winston}@i2.a-sta.edu.sg 1 Depatment of Compute Science, School of Computing, National Univesity

More information

A modal estimation based multitype sensor placement method

A modal estimation based multitype sensor placement method A modal estimation based multitype senso placement method *Xue-Yang Pei 1), Ting-Hua Yi 2) and Hong-Nan Li 3) 1),)2),3) School of Civil Engineeing, Dalian Univesity of Technology, Dalian 116023, China;

More information

Adaptation of TDMA Parameters Based on Network Conditions

Adaptation of TDMA Parameters Based on Network Conditions Adaptation of TDMA Paametes Based on Netwok Conditions Boa Kaaoglu Dept. of Elect. and Compute Eng. Univesity of Rocheste Rocheste, NY 14627 Email: kaaoglu@ece.ocheste.edu Tolga Numanoglu Dept. of Elect.

More information

A Two-stage and Parameter-free Binarization Method for Degraded Document Images

A Two-stage and Parameter-free Binarization Method for Degraded Document Images A Two-stage and Paamete-fee Binaization Method fo Degaded Document Images Yung-Hsiang Chiu 1, Kuo-Liang Chung 1, Yong-Huai Huang 2, Wei-Ning Yang 3, Chi-Huang Liao 4 1 Depatment of Compute Science and

More information

Communication vs Distributed Computation: an alternative trade-off curve

Communication vs Distributed Computation: an alternative trade-off curve Communication vs Distibuted Computation: an altenative tade-off cuve Yahya H. Ezzeldin, Mohammed amoose, Chistina Fagouli Univesity of Califonia, Los Angeles, CA 90095, USA, Email: {yahya.ezzeldin, mkamoose,

More information

Combinatorial Mobile IP: A New Efficient Mobility Management Using Minimized Paging and Local Registration in Mobile IP Environments

Combinatorial Mobile IP: A New Efficient Mobility Management Using Minimized Paging and Local Registration in Mobile IP Environments Wieless Netwoks 0, 3 32, 200 200 Kluwe Academic Publishes. Manufactued in The Nethelands. Combinatoial Mobile IP: A New Efficient Mobility Management Using Minimized Paging and Local Registation in Mobile

More information

Module 6 STILL IMAGE COMPRESSION STANDARDS

Module 6 STILL IMAGE COMPRESSION STANDARDS Module 6 STILL IMAE COMPRESSION STANDARDS Lesson 17 JPE-2000 Achitectue and Featues Instuctional Objectives At the end of this lesson, the students should be able to: 1. State the shotcomings of JPE standad.

More information

A Memory Efficient Array Architecture for Real-Time Motion Estimation

A Memory Efficient Array Architecture for Real-Time Motion Estimation A Memoy Efficient Aay Achitectue fo Real-Time Motion Estimation Vasily G. Moshnyaga and Keikichi Tamau Depatment of Electonics & Communication, Kyoto Univesity Sakyo-ku, Yoshida-Honmachi, Kyoto 66-1, JAPAN

More information

EE 6900: Interconnection Networks for HPC Systems Fall 2016

EE 6900: Interconnection Networks for HPC Systems Fall 2016 EE 6900: Inteconnection Netwoks fo HPC Systems Fall 2016 Avinash Kaanth Kodi School of Electical Engineeing and Compute Science Ohio Univesity Athens, OH 45701 Email: kodi@ohio.edu 1 Acknowledgement: Inteconnection

More information

Event-based Location Dependent Data Services in Mobile WSNs

Event-based Location Dependent Data Services in Mobile WSNs Event-based Location Dependent Data Sevices in Mobile WSNs Liang Hong 1, Yafeng Wu, Sang H. Son, Yansheng Lu 3 1 College of Compute Science and Technology, Wuhan Univesity, China Depatment of Compute Science,

More information

Illumination methods for optical wear detection

Illumination methods for optical wear detection Illumination methods fo optical wea detection 1 J. Zhang, 2 P.P.L.Regtien 1 VIMEC Applied Vision Technology, Coy 43, 5653 LC Eindhoven, The Nethelands Email: jianbo.zhang@gmail.com 2 Faculty Electical

More information

On using circuit-switched networks for file transfers

On using circuit-switched networks for file transfers On using cicuit-switched netwoks fo file tansfes Xiuduan Fang, Malathi Veeaaghavan Univesity of Viginia Email: {xf4c, mv5g}@viginia.edu Abstact High-speed optical cicuit-switched netwoks ae being deployed

More information

MANET QoS support without reservations

MANET QoS support without reservations SECURITY AND COMMUNICATION NETWORKS Secuity Comm. Netwoks (2010) Published online in Wiley InteScience (www.intescience.wiley.com)..183 SPECIAL ISSUE PAPER MANET QoS suppot without esevations Soon Y. Oh,

More information

The Dual Round Robin Matching Switch with Exhaustive Service

The Dual Round Robin Matching Switch with Exhaustive Service The Dual Round Robin Matching Switch with Exhaustive Sevice Yihan Li, Shivenda S. Panwa, H. Jonathan Chao Abstact Vitual Output Queuing is widely used by fixed-length highspeed switches to ovecome head-of-line

More information

WIRELESS sensor networks (WSNs), which are capable

WIRELESS sensor networks (WSNs), which are capable IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, VOL. XX, NO. XX, XXX 214 1 Lifetime and Enegy Hole Evolution Analysis in Data-Gatheing Wieless Senso Netwoks Ju Ren, Student Membe, IEEE, Yaoxue Zhang, Kuan

More information

Controlled Information Maximization for SOM Knowledge Induced Learning

Controlled Information Maximization for SOM Knowledge Induced Learning 3 Int'l Conf. Atificial Intelligence ICAI'5 Contolled Infomation Maximization fo SOM Knowledge Induced Leaning Ryotao Kamimua IT Education Cente and Gaduate School of Science and Technology, Tokai Univeisity

More information

The Internet Ecosystem and Evolution

The Internet Ecosystem and Evolution The Intenet Ecosystem and Evolution Contents Netwok outing: basics distibuted/centalized, static/dynamic, linkstate/path-vecto inta-domain/inte-domain outing Mapping the sevice model to AS-AS paths valley-fee

More information

(1) W tcp = (3) N. Assuming 1 P r 1. = W r (4) a 1/(k+1) W 2/(k+1)

(1) W tcp = (3) N. Assuming 1 P r 1. = W r (4) a 1/(k+1) W 2/(k+1) 1 Multi Path PERT Ankit Singh and A. L. Naasimha Reddy Electical and Compute Engineeing Depatment, Texas A&M Univesity; email: eddy@ece.tamu.edu. Abstact This pape pesents a new multipath delay based algoithm,

More information

Point-Biserial Correlation Analysis of Fuzzy Attributes

Point-Biserial Correlation Analysis of Fuzzy Attributes Appl Math Inf Sci 6 No S pp 439S-444S (0 Applied Mathematics & Infomation Sciences An Intenational Jounal @ 0 NSP Natual Sciences Publishing o Point-iseial oelation Analysis of Fuzzy Attibutes Hao-En hueh

More information

Slotted Random Access Protocol with Dynamic Transmission Probability Control in CDMA System

Slotted Random Access Protocol with Dynamic Transmission Probability Control in CDMA System Slotted Random Access Potocol with Dynamic Tansmission Pobability Contol in CDMA System Intaek Lim 1 1 Depatment of Embedded Softwae, Busan Univesity of Foeign Studies, itlim@bufs.ac.k Abstact In packet

More information

Achievable Catalog Size in Peer-to-Peer Video-on-Demand Systems

Achievable Catalog Size in Peer-to-Peer Video-on-Demand Systems Achievable Catalog Size in Pee-to-Pee Video-on-Demand Systems Yacine Boufkhad, Fabien Mathieu, Fabien De Montgolfie, Diego Peino, Lauent Viennot To cite this vesion: Yacine Boufkhad, Fabien Mathieu, Fabien

More information

Dynamic Processor Scheduling with Client Resources for Fast Multi-resolution WWW Image Browsing

Dynamic Processor Scheduling with Client Resources for Fast Multi-resolution WWW Image Browsing Dynamic Pocesso Scheduling with Resouces fo Fast Multi-esolution WWW Image Bowsing Daniel Andesen, Tao Yang, David Watson, and Athanassios Poulakidas Depatment of Compute Science Univesity of Califonia

More information

Optical Flow for Large Motion Using Gradient Technique

Optical Flow for Large Motion Using Gradient Technique SERBIAN JOURNAL OF ELECTRICAL ENGINEERING Vol. 3, No. 1, June 2006, 103-113 Optical Flow fo Lage Motion Using Gadient Technique Md. Moshaof Hossain Sake 1, Kamal Bechkoum 2, K.K. Islam 1 Abstact: In this

More information

Interference-Aware Multicast for Wireless Multihop Networks

Interference-Aware Multicast for Wireless Multihop Networks Intefeence-Awae Multicast fo Wieless Multihop Netwoks Daniel Letpatchya School of Electical and Compute Engineeing Geogia Institute of Technology Atlanta, Geogia 30332 0250 Douglas M. Blough School of

More information

arxiv: v2 [physics.soc-ph] 30 Nov 2016

arxiv: v2 [physics.soc-ph] 30 Nov 2016 Tanspotation dynamics on coupled netwoks with limited bandwidth Ming Li 1,*, Mao-Bin Hu 1, and Bing-Hong Wang 2, axiv:1607.05382v2 [physics.soc-ph] 30 Nov 2016 1 School of Engineeing Science, Univesity

More information

HISTOGRAMS are an important statistic reflecting the

HISTOGRAMS are an important statistic reflecting the JOURNAL OF L A T E X CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 1 D 2 HistoSketch: Disciminative and Dynamic Similaity-Peseving Sketching of Steaming Histogams Dingqi Yang, Bin Li, Laua Rettig, and Philippe

More information

An Efficient Handover Mechanism Using the General Switch Management Protocol on a Multi-Protocol Label Switching Network

An Efficient Handover Mechanism Using the General Switch Management Protocol on a Multi-Protocol Label Switching Network An Efficient andove Mechanism Using the Geneal Switch Management Potocol on a Multi-Potocol abel Switching Netwok Seong Gon hoi, yun Joo Kang, and Jun Kyun hoi Using the geneal switch management potocol

More information

Lifetime and Energy Hole Evolution Analysis in Data-Gathering Wireless Sensor Networks

Lifetime and Energy Hole Evolution Analysis in Data-Gathering Wireless Sensor Networks 788 IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, VOL. 12, NO. 2, APRIL 2016 Lifetime and Enegy Hole Evolution Analysis in Data-Gatheing Wieless Senso Netwoks Ju Ren, Student Membe, IEEE, Yaoxue Zhang,

More information

Effective Missing Data Prediction for Collaborative Filtering

Effective Missing Data Prediction for Collaborative Filtering Effective Missing Data Pediction fo Collaboative Filteing Hao Ma, Iwin King and Michael R. Lyu Dept. of Compute Science and Engineeing The Chinese Univesity of Hong Kong Shatin, N.T., Hong Kong { hma,

More information

Analysis of Wired Short Cuts in Wireless Sensor Networks

Analysis of Wired Short Cuts in Wireless Sensor Networks Analysis of Wied Shot Cuts in Wieless Senso Netwos ohan Chitaduga Depatment of Electical Engineeing, Univesity of Southen Califonia, Los Angeles 90089, USA Email: chitadu@usc.edu Ahmed Helmy Depatment

More information

Topological Characteristic of Wireless Network

Topological Characteristic of Wireless Network Topological Chaacteistic of Wieless Netwok Its Application to Node Placement Algoithm Husnu Sane Naman 1 Outline Backgound Motivation Papes and Contibutions Fist Pape Second Pape Thid Pape Futue Woks Refeences

More information

Modelling, simulation, and performance analysis of a CAN FD system with SAE benchmark based message set

Modelling, simulation, and performance analysis of a CAN FD system with SAE benchmark based message set Modelling, simulation, and pefomance analysis of a CAN FD system with SAE benchmak based message set Mahmut Tenuh, Panagiotis Oikonomidis, Peiklis Chachalakis, Elias Stipidis Mugla S. K. Univesity, TR;

More information

A New Finite Word-length Optimization Method Design for LDPC Decoder

A New Finite Word-length Optimization Method Design for LDPC Decoder A New Finite Wod-length Optimization Method Design fo LDPC Decode Jinlei Chen, Yan Zhang and Xu Wang Key Laboatoy of Netwok Oiented Intelligent Computation Shenzhen Gaduate School, Habin Institute of Technology

More information

APPLICATION OF STRUCTURED QUEUING NETWORKS IN QOS ESTIMITION OF TELECOMMUNICATION SERVICE

APPLICATION OF STRUCTURED QUEUING NETWORKS IN QOS ESTIMITION OF TELECOMMUNICATION SERVICE APPLICATION OF STRUCTURED QUEUING NETWORKS IN QOS ESTIMITION OF TELECOMMUNICATION SERVICE 1 YAROSLAVTSEV A.F., 2 Al-THUNEIBAT S.A., 3 AL TAWALBEH N.A 1 Depatment of Netwoking, SSUTI, Novosibisk, Russia

More information

Separability and Topology Control of Quasi Unit Disk Graphs

Separability and Topology Control of Quasi Unit Disk Graphs Sepaability and Topology Contol of Quasi Unit Disk Gaphs Jiane Chen, Anxiao(Andew) Jiang, Iyad A. Kanj, Ge Xia, and Fenghui Zhang Dept. of Compute Science, Texas A&M Univ. College Station, TX 7784. {chen,

More information

Erasure-Coding Based Routing for Opportunistic Networks

Erasure-Coding Based Routing for Opportunistic Networks Easue-Coding Based Routing fo Oppotunistic Netwoks Yong Wang, Sushant Jain, Magaet Matonosi, Kevin Fall Pinceton Univesity, Univesity of Washington, Intel Reseach Bekeley ABSTRACT Routing in Delay Toleant

More information

A Caching Algorithm using Evolutionary Game Theory in a File-Sharing System

A Caching Algorithm using Evolutionary Game Theory in a File-Sharing System A Caching Algoithm using Evolutionay Game Theoy in a File-Shaing System Masahio Sasabe Cybemedia Cente Osaka Univesity 1-32 Machikaneyama-cho, Toyonaka-shi Osaka 56-43, Japan m-sasabe@cmc.osaka-u.ac.jp

More information

Tier-Based Underwater Acoustic Routing for Applications with Reliability and Delay Constraints

Tier-Based Underwater Acoustic Routing for Applications with Reliability and Delay Constraints Tie-Based Undewate Acoustic Routing fo Applications with Reliability and Delay Constaints Li-Chung Kuo Depatment of Electical Engineeing State Univesity of New Yok at Buffalo Buffalo, New Yok 14260 Email:

More information

An Optimised Density Based Clustering Algorithm

An Optimised Density Based Clustering Algorithm Intenational Jounal of Compute Applications (0975 8887) Volume 6 No.9, Septembe 010 An Optimised Density Based Clusteing Algoithm J. Hencil Pete Depatment of Compute Science St. Xavie s College, Palayamkottai,

More information

A Novel Automatic White Balance Method For Digital Still Cameras

A Novel Automatic White Balance Method For Digital Still Cameras A Novel Automatic White Balance Method Fo Digital Still Cameas Ching-Chih Weng 1, Home Chen 1,2, and Chiou-Shann Fuh 3 Depatment of Electical Engineeing, 2 3 Gaduate Institute of Communication Engineeing

More information

The concept of PARPS - Packet And Resource Plan Scheduling

The concept of PARPS - Packet And Resource Plan Scheduling The concept of PARPS - Packet And Resouce Plan Scheduling Magnus Eiksson 1 and Håkan Sätebeg 2 1) Dept. of Signals, Sensos and Systems, Royal Inst. of Technology, Sweden. E-mail: magnus.eiksson@ite.mh.se.

More information

Segmentation of Casting Defects in X-Ray Images Based on Fractal Dimension

Segmentation of Casting Defects in X-Ray Images Based on Fractal Dimension 17th Wold Confeence on Nondestuctive Testing, 25-28 Oct 2008, Shanghai, China Segmentation of Casting Defects in X-Ray Images Based on Factal Dimension Jue WANG 1, Xiaoqin HOU 2, Yufang CAI 3 ICT Reseach

More information

Efficient protection of many-to-one. communications

Efficient protection of many-to-one. communications Efficient potection of many-to-one communications Miklós Molná, Alexande Guitton, Benad Cousin, and Raymond Maie Iisa, Campus de Beaulieu, 35 042 Rennes Cedex, Fance Abstact. The dependability of a netwok

More information

Root Cause Analysis for Long-Lived TCP Connections

Root Cause Analysis for Long-Lived TCP Connections Root Cause Analysis fo Long-Lived TCP Connections M. Siekkinen, G. Uvoy-Kelle, E. W. Biesack, T. En-Najjay Institut Euécom BP 93, 694 Sophia-Antipolis Cedex, Fance Email: {siekkine,uvoy,ebi,ennajja}@euecom.f

More information

ADDING REALISM TO SOURCE CHARACTERIZATION USING A GENETIC ALGORITHM

ADDING REALISM TO SOURCE CHARACTERIZATION USING A GENETIC ALGORITHM ADDING REALISM TO SOURCE CHARACTERIZATION USING A GENETIC ALGORITHM Luna M. Rodiguez*, Sue Ellen Haupt, and Geoge S. Young Depatment of Meteoology and Applied Reseach Laboatoy The Pennsylvania State Univesity,

More information

Decentralized Trust Management for Ad-Hoc Peer-to-Peer Networks

Decentralized Trust Management for Ad-Hoc Peer-to-Peer Networks Decentalized Tust Management fo Ad-Hoc Pee-to-Pee Netwoks Thomas Repantis Vana Kalogeaki Depatment of Compute Science & Engineeing Univesity of Califonia, Riveside Riveside, CA 92521 {tep,vana}@cs.uc.edu

More information

INFORMATION DISSEMINATION DELAY IN VEHICLE-TO-VEHICLE COMMUNICATION NETWORKS IN A TRAFFIC STREAM

INFORMATION DISSEMINATION DELAY IN VEHICLE-TO-VEHICLE COMMUNICATION NETWORKS IN A TRAFFIC STREAM INFORMATION DISSEMINATION DELAY IN VEHICLE-TO-VEHICLE COMMUNICATION NETWORKS IN A TRAFFIC STREAM LiLi Du Depatment of Civil, Achitectual, and Envionmental Engineeing Illinois Institute of Technology 3300

More information

Lecture # 04. Image Enhancement in Spatial Domain

Lecture # 04. Image Enhancement in Spatial Domain Digital Image Pocessing CP-7008 Lectue # 04 Image Enhancement in Spatial Domain Fall 2011 2 domains Spatial Domain : (image plane) Techniques ae based on diect manipulation of pixels in an image Fequency

More information

Hierarchical Peer-to-peer Systems

Hierarchical Peer-to-peer Systems Hieachical Pee-to-pee ystems L. Gacés-Eice 1, E.W. Biesack 1, P.A. Felbe 1, K.W. Ross 2, and G. Uvoy-Kelle 1 1 Institut EURECOM 06904 ophia Antipolis, Fance {gaces ebi felbe uvoy}@euecom.f 2 Polytechnic

More information

Fifth Wheel Modelling and Testing

Fifth Wheel Modelling and Testing Fifth heel Modelling and Testing en Masoy Mechanical Engineeing Depatment Floida Atlantic Univesity Boca aton, FL 4 Lois Malaptias IFMA Institut Fancais De Mechanique Advancee ampus De lemont Feand Les

More information

Minimizing Application-Level Delay of Multi-Path TCP in Wireless networks: A Receiver-Centric Approach

Minimizing Application-Level Delay of Multi-Path TCP in Wireless networks: A Receiver-Centric Approach Minimizing Application-Level Delay of Multi-Path TCP in Wieless netwoks: A Receive-Centic Appoach Se-Yong Pak, Changhee Joo, Yongseok Pak, and Saewoong Bahk Depament of ECE and INMC, Seoul National Univesity,

More information

An Energy-Efficient Approach for Provenance Transmission in Wireless Sensor Networks

An Energy-Efficient Approach for Provenance Transmission in Wireless Sensor Networks An Enegy-Efficient Appoach fo Povenance Tansmission in Wieless Senso Netwoks S. M. Iftekhaul Alam Pudue Univesity alams@pudue.edu Sonia Fahmy Pudue Univesity fahmy@cs.pudue.edu Abstact Assessing the tustwothiness

More information

An Extension to the Local Binary Patterns for Image Retrieval

An Extension to the Local Binary Patterns for Image Retrieval , pp.81-85 http://x.oi.og/10.14257/astl.2014.45.16 An Extension to the Local Binay Pattens fo Image Retieval Zhize Wu, Yu Xia, Shouhong Wan School of Compute Science an Technology, Univesity of Science

More information

RT-WLAN: A Soft Real-Time Extension to the ORiNOCO Linux Device Driver

RT-WLAN: A Soft Real-Time Extension to the ORiNOCO Linux Device Driver 1 RT-WLAN: A Soft Real-Time Extension to the ORiNOCO Linux Device Dive Amit Jain Daji Qiao Kang G. Shin The Univesity of Michigan Ann Abo, MI 4819, USA {amitj,dqiao,kgshin@eecs.umich.edu Abstact The cuent

More information

High performance CUDA based CNN image processor

High performance CUDA based CNN image processor High pefomance UDA based NN image pocesso GEORGE VALENTIN STOIA, RADU DOGARU, ELENA RISTINA STOIA Depatment of Applied Electonics and Infomation Engineeing Univesity Politehnica of Buchaest -3, Iuliu Maniu

More information

Fast quality-guided flood-fill phase unwrapping algorithm for three-dimensional fringe pattern profilometry

Fast quality-guided flood-fill phase unwrapping algorithm for three-dimensional fringe pattern profilometry Univesity of Wollongong Reseach Online Faculty of Infomatics - Papes (Achive) Faculty of Engineeing and Infomation Sciences 2010 Fast quality-guided flood-fill phase unwapping algoithm fo thee-dimensional

More information

Fault-Tolerant Routing Schemes in RDT(2,2,1)/α-Based Interconnection Network for Networks-on-Chip Designs

Fault-Tolerant Routing Schemes in RDT(2,2,1)/α-Based Interconnection Network for Networks-on-Chip Designs Fault-Toleant Routing Schemes in RDT(,,)/α-Based Inteconnection Netwok fo Netwoks-on-Chip Designs Mei Yang, Tao Li, Yingtao Jiang, and Yulu Yang Dept. of Electical & Compute Engineeing Univesity of Nevada,

More information

GTOC 9, Multiple Space Debris Rendezvous Trajectory Design in the J2 environment

GTOC 9, Multiple Space Debris Rendezvous Trajectory Design in the J2 environment GTOC 9, Multiple Space Debis Rendezvous Tajectoy Design in the J envionment Macus Hallmann, Makus Schlottee, Ansga Heidecke, Maco Sagliano Fedeico Fumenti, Volke Maiwald, René Schwaz Institute of Space

More information

Configuring RSVP-ATM QoS Interworking

Configuring RSVP-ATM QoS Interworking Configuing RSVP-ATM QoS Intewoking Last Updated: Januay 15, 2013 This chapte descibes the tasks fo configuing the RSVP-ATM QoS Intewoking featue, which povides suppot fo Contolled Load Sevice using RSVP

More information

Number of Paths and Neighbours Effect on Multipath Routing in Mobile Ad Hoc Networks

Number of Paths and Neighbours Effect on Multipath Routing in Mobile Ad Hoc Networks Numbe of Paths and Neighbous Effect on Multipath Routing in Mobile Ad Hoc Netwoks Oday Jeew School of Infomation Systems and Accounting Univesity of Canbea Canbea ACT 2617, Austalia oday.jeew@canbea.edu.au

More information

Embeddings into Crossed Cubes

Embeddings into Crossed Cubes Embeddings into Cossed Cubes Emad Abuelub *, Membe, IAENG Abstact- The hypecube paallel achitectue is one of the most popula inteconnection netwoks due to many of its attactive popeties and its suitability

More information

The EigenRumor Algorithm for Ranking Blogs

The EigenRumor Algorithm for Ranking Blogs he EigenRumo Algoithm fo Ranking Blogs Ko Fujimua N Cybe Solutions Laboatoies N Copoation akafumi Inoue N Cybe Solutions Laboatoies N Copoation Masayuki Sugisaki N Resonant Inc. ABSRAC he advent of easy

More information

IP MULTICAST. Adiseshu Hari, T. V. Lakshman and Gordon Wilfong Nokia Bell Labs

IP MULTICAST. Adiseshu Hari, T. V. Lakshman and Gordon Wilfong Nokia Bell Labs IP MULTICAST Adiseshu Hai, T. V. Lakshman and Godon Wilfong Nokia DIMACS Wokshop on Algoithms fo Data Cente Netwoks Rutges Univesity, NJ 1 Why is IP Multicast not deployed in public netwoks? Denial-of-Sevice

More information

UCB CS61C : Machine Structures

UCB CS61C : Machine Structures inst.eecs.bekeley.edu/~cs61c UCB CS61C : Machine Stuctues Lectue SOE Dan Gacia Lectue 28 CPU Design : Pipelining to Impove Pefomance 2010-04-05 Stanfod Reseaches have invented a monitoing technique called

More information

Spiral Recognition Methodology and Its Application for Recognition of Chinese Bank Checks

Spiral Recognition Methodology and Its Application for Recognition of Chinese Bank Checks Spial Recognition Methodology and Its Application fo Recognition of Chinese Bank Checks Hanshen Tang 1, Emmanuel Augustin 2, Ching Y. Suen 1, Olivie Baet 2, Mohamed Cheiet 3 1 Cente fo Patten Recognition

More information

Quality Aware Privacy Protection for Location-based Services

Quality Aware Privacy Protection for Location-based Services In Poceedings of the th Intenational Confeence on Database Systems fo Advanced Applications (DASFAA 007), Bangkok, Thailand, Apil 9-, 007. Quality Awae Pivacy Potection fo Location-based Sevices Zhen Xiao,,

More information

Input Layer f = 2 f = 0 f = f = 3 1,16 1,1 1,2 1,3 2, ,2 3,3 3,16. f = 1. f = Output Layer

Input Layer f = 2 f = 0 f = f = 3 1,16 1,1 1,2 1,3 2, ,2 3,3 3,16. f = 1. f = Output Layer Using the Gow-And-Pune Netwok to Solve Poblems of Lage Dimensionality B.J. Biedis and T.D. Gedeon School of Compute Science & Engineeing The Univesity of New South Wales Sydney NSW 2052 AUSTRALIA bbiedis@cse.unsw.edu.au

More information

CRESTBOT: A New Family of Resilient Botnets

CRESTBOT: A New Family of Resilient Botnets CRESTBOT: A New Family of Resilient Botnets Duc T. Ha, Hung Q. Ngo, Madhusudhanan Chandasekaan Depatment of Compute Science and Engineeing State Univesity of New Yok at Buffalo Amhest, NY 1426, USA Email:

More information

Assessment of Track Sequence Optimization based on Recorded Field Operations

Assessment of Track Sequence Optimization based on Recorded Field Operations Assessment of Tack Sequence Optimization based on Recoded Field Opeations Matin A. F. Jensen 1,2,*, Claus G. Søensen 1, Dionysis Bochtis 1 1 Aahus Univesity, Faculty of Science and Technology, Depatment

More information

Signal integrity analysis and physically based circuit extraction of a mounted

Signal integrity analysis and physically based circuit extraction of a mounted emc design & softwae Signal integity analysis and physically based cicuit extaction of a mounted SMA connecto A poposed geneal appoach is given fo the definition of an equivalent cicuit with SMAs mounted

More information

SYSTEM LEVEL REUSE METRICS FOR OBJECT ORIENTED SOFTWARE : AN ALTERNATIVE APPROACH

SYSTEM LEVEL REUSE METRICS FOR OBJECT ORIENTED SOFTWARE : AN ALTERNATIVE APPROACH I J C A 7(), 202 pp. 49-53 SYSTEM LEVEL REUSE METRICS FOR OBJECT ORIENTED SOFTWARE : AN ALTERNATIVE APPROACH Sushil Goel and 2 Rajesh Vema Associate Pofesso, Depatment of Compute Science, Dyal Singh College,

More information

Simulation and Performance Evaluation of Network on Chip Architectures and Algorithms using CINSIM

Simulation and Performance Evaluation of Network on Chip Architectures and Algorithms using CINSIM J. Basic. Appl. Sci. Res., 1(10)1594-1602, 2011 2011, TextRoad Publication ISSN 2090-424X Jounal of Basic and Applied Scientific Reseach www.textoad.com Simulation and Pefomance Evaluation of Netwok on

More information

DUe to the recent developments of gigantic social networks

DUe to the recent developments of gigantic social networks Exploing Communities in Lage Pofiled Gaphs Yankai Chen, Yixiang Fang, Reynold Cheng Membe, IEEE, Yun Li, Xiaojun Chen, Jie Zhang 1 Abstact Given a gaph G and a vetex q G, the community seach (CS) poblem

More information

Characterizing Data Deliverability of Greedy Routing in Wireless Sensor Networks

Characterizing Data Deliverability of Greedy Routing in Wireless Sensor Networks Chaacteizing Data Deliveability of Geedy Routing in Wieless Senso Netwoks Jinwei Liu, Lei Yu, Haiying Shen, Yangyang He and Jason Hallstom Depatment of Electical and Compute Engineeing, Clemson Univesity,

More information

Wormhole Detection and Prevention in MANETs

Wormhole Detection and Prevention in MANETs Womhole Detection and Pevention in MANETs Lija Joy Compute Science and Engineeing KMEA Engineeing College Enakulum, Keala, India lijavj@gmail.com Sheena Kuian K Compute Science and Engineeing KMEA Engineeing

More information

A Cross-Layer Framework of QoS Routing and Distributed Scheduling for Mesh Networks

A Cross-Layer Framework of QoS Routing and Distributed Scheduling for Mesh Networks A Coss-Laye Famewok of QoS Routing and Distibuted Scheduling fo Mesh Netwoks Chi Haold Liu, Athanasios Gkelias, and Kin K. Leung Depatment of Electical and Electonic Engineeing, Impeial College London

More information

MapReduce Optimizations and Algorithms 2015 Professor Sasu Tarkoma

MapReduce Optimizations and Algorithms 2015 Professor Sasu Tarkoma apreduce Optimizations and Algoithms 2015 Pofesso Sasu Takoma www.cs.helsinki.fi Optimizations Reduce tasks cannot stat befoe the whole map phase is complete Thus single slow machine can slow down the

More information

A ROI Focusing Mechanism for Digital Cameras

A ROI Focusing Mechanism for Digital Cameras A ROI Focusing Mechanism fo Digital Cameas Chu-Hui Lee, Meng-Feng Lin, Chun-Ming Huang, and Chun-Wei Hsu Abstact With the development and application of digital technologies, the digital camea is moe popula

More information

On Adaptive Bandwidth Sharing with Rate Guarantees

On Adaptive Bandwidth Sharing with Rate Guarantees On Adaptive Bandwidth Shaing with Rate Guaantees N.G. Duffield y T. V. Lakshman D. Stiliadis y AT&T Laboatoies Bell Labs Rm A175, 180 Pak Avenue Lucent Technologies Floham Pak, 101 Cawfods Cone Road NJ

More information

Frequency Domain Approach for Face Recognition Using Optical Vanderlugt Filters

Frequency Domain Approach for Face Recognition Using Optical Vanderlugt Filters Optics and Photonics Jounal, 016, 6, 94-100 Published Online August 016 in SciRes. http://www.scip.og/jounal/opj http://dx.doi.og/10.436/opj.016.68b016 Fequency Domain Appoach fo Face Recognition Using

More information

Ethernet PON (epon): Design and Analysis of an Optical Access Network.

Ethernet PON (epon): Design and Analysis of an Optical Access Network. Ethenet PON epon: Design and Analysis of an Optical Access Netwo. Glen Kame Depatment of Compute Science Univesity of Califonia, Davis, CA 9566, USA Tel:.530.297.527; Fax:.530.297.530 E-mail: ame@cs.ucdavis.edu

More information

And Ph.D. Candidate of Computer Science, University of Putra Malaysia 2 Faculty of Computer Science and Information Technology,

And Ph.D. Candidate of Computer Science, University of Putra Malaysia 2 Faculty of Computer Science and Information Technology, (IJCSIS) Intenational Jounal of Compute Science and Infomation Secuity, Efficient Candidacy Reduction Fo Fequent Patten Mining M.H Nadimi-Shahaki 1, Nowati Mustapha 2, Md Nasi B Sulaiman 2, Ali B Mamat

More information

DISTRIBUTION MIXTURES

DISTRIBUTION MIXTURES Application Example 7 DISTRIBUTION MIXTURES One fequently deals with andom vaiables the distibution of which depends on vaious factos. One example is the distibution of atmospheic paametes such as wind

More information