Dynamic Node Join Algorithm with Rate-Distortion for P2P Live Multipath Networks
|
|
- Jane Carson
- 5 years ago
- Views:
Transcription
1 Dynamc Node Jon Algorthm wth Rate-Dstorton for P2P Lve Multpath Networs Shyh-Chang Lu 1,Tsung-Hung Chen 1, Tsang-Hung Wu 1, Peter P.S. Wang 2 1 Dept. of Informaton Eng, 2 Dept. of Computer Scence and Informaton Eng 1 I-Shou Unversty, 2 Natonal Central Unversty 1 Koushung, 2 Taoyuan Tawan lusc@su.edu.tw, jhchenysl@yahoo.com.tw, peterw8865@hotmal.com, pdespd@gmal.com Abstract: - The delvery of multmeda that effcently maxmzes ts qualty n changng networ condtons s one of the most challengng tass n the desgn of lve streamng systems. Ths study attempts to mprove current P2P (peer-to-peer) lve streamng systems by allowng users to enjoy hgh-qualty servce under the lmtatons of networ resources. The proposed mprovement method nvolves summng up and analyzng the consderaton factors and restrcton factors nvolvng lve stream qualty durng system operatons. The proposed R-D (Rate-Dstorton) optmzed dynamc nodes jon algorthm s based on multpath streamng concept and recever-drven approach. Ths dstrbuted algorthm enables the system to evaluate the current networ statu n order to optmze the end-to-end dstorton of P2P networs. Results of ths study demonstrate the effectveness of the proposed approach. Key-Words: - P2P lve streamng, Rate-Dstorton, multpath streamng, recever-drven, end-to-end dstorton 1 Introducton The use of lve streamng systems s growng dramatcally, especally for sharng real-tme multmeda programs. The structures of multmeda streamng systems are classfed as CDN (Content Delver Networ) systems and P2P lve streamng systems. A CDN system has a clent-server archtecture (e.g., Aama, Lmelght Networs) [1][2]. One advantage of CDN systems are that users can effcently obtan hgh-qualty streamng servce from adjacent proxy servers. Ther drawbacs are hgh server costs and bandwdth rental fees for ICP (Internet Content Provder). P2P lve streamng systems ntegrate the dspersed resource dstrbute system loadng and have low ICP operaton cost. Achevng hgh streamng qualty n end-to-end streamng transmsson and broadcastng poses a major challenge for systems wth varous equpment and bandwdth [3]. The followng ssues need to be addressed : (1) Related Scalablty Analyss: the selecton of the best nodes for future end-to-end connectons by a networ probe or recever-drven approach[4][5]; (2) Related R-D Optmzaton Method: how optmzed R-D algorthm mprove pacet schedulng or flow transmsson qualty[6][7]; (3) Related Meshed- Multpath approach: ncreasng the qualty of system servce through Meshed-Multpath archtectures [3][8]. Whle meda schedulng processes and system archtectures have been extensvely studed, the lmtatons of P2P lve meda transmsson n enhancng streamng qualty have seldom been addressed [9][10]. However, users of P2P lve streamng systems usually do not use them for very long perods. Snce the stream paths change frequently, the mang too many adjustments to optmze the transmssons for clents affects the vewng stablty for users. Therefore, ths paper presents a novel approach that s optmzed for P2P lve transmsson, based on prevous approache that acheves vewng stablty. The rest of ths paper s organzed as follows. Secton 2 summarzes the ssues nfluencng meda streamng qualty, and explans the mproved method n these ssues. Secton 3 then dscusses recever-drven approach wth rate-dstorton theory, and apples t to the relevant proposed approach. Secton 4 descrbes the smulaton test performed wth dfferent algorthms to measure the effcency of the proposed algorthm. Conclusons are fnally drawn n Secton 5. ISSN: Issue 6, Volume 6, June 2009
2 2 Improvement for Lve Stream Qualty of P2P Networ Wthout loss of generalty, ths chapter dscusses the ssues nfluencng the qualty of P2P lve streamng. An approprate networ structure and operatonal process are then presented to acheve our research objectve. 2.1 Qualty Issues of P2P Lve Stream Networ Indvduals access the Internet frequently. In partcular, P2P lve streamng systems are ncreasngly popular due to advances n computer equpment, applcaton functons and operatng envronments[11][12]. Streamng servces have a hgh bandwdth requrement, as well as requre both serve and play streamed content smultaneously. Ths operaton pattern not only maes a heavy load on the networ nterface, but also ncreases the utlzaton of CPU and Memory, mang ts operaton unrelable. Varous studes have ndcated that the stream flow qualty of the clents networs s affected by endto-end dstorton (ncludng the end-to-end streamng rate and dstorton rate)[13][14], Addtonally, lmtatons n computer resources and bandwdth nfluence the networ capablty of a node when the user clent transmts meda streamng to others (reference Table 1)[15]. We can thus nfer that the followng factors nfluence the stream qualty of the P2P lve streamng system: Table 1 Optmal degree out number of a P2P clent under dfferent bandwdths Downln Bandwdth Upln Bandwdth Degree out 512 bps 256 bps 0 3 Mbps 384 bps Mbps 896 bps 3 20 Mbps 2 Mbps 6 20 Mbps 5 Mbps 17 The test streamng rate n the experments s 220 bps [Issue 1: Maxmum Connectons Capacty] The number of upln (degreeout) and downln (degreen) connectons n a node should be restrcted accordng to computer performance and the networ envronment. [Issue 2: End-to-end Delay Lmtaton] To ensure that the P2P lve stream networ s operatng n real tme, the end-to-end delay must be no larger than a reasonable lmt value (δ). [Issue 3: End-to-end Dstorton Effect] Although the stream qualty of the P2P lve stream networ s affected by the end-to-end dstorton, the value of the end-to-end dstorton contans the end-to-end streamng rate and the dstorton rate. Based on these three factor the lve stream qualty of P2P networ s assessed accordng to the end-to-end delay, streamng rate and dstorton rate. 2.2 P2P Networ Structure The frst ssue [Issue 1] s a fundamental restrcton. Therefore, based on the other two ssue ths study chooses an approprate networ structure and handlng mechansm, then develops an algorthm to mprove the lve streamng qualty of P2P networ. The networ structure n ths wor s based on Mesh-Overlay, and has the followng characterstcs. Frst, every front-end may have parent nodes and chld nodes (as found n other Mesh-Overlay P2P systems such as CoopNet and Splt Stream) [16][17][18]. Second, any networ nner-node can ndependently obtan the requred resources from adjacent nodes va a pull method [19]. Ths approach mproves transmsson performance, snce the Mesh-Overlay structure suffcently maes good use of the upln bandwdth of each node. However, ths approach s unsutable for the lve streamng servce, because the characterstcs of the Mesh- Overlay structure are loosely controllng and nconsstent undrectonal transferrng [20]. To elmnate overcome ths shortcomng, two mechansms are added, namely multpath streamng concept and a recever-drven approach. Frst, the multpath streamng concept s added to the Mesh- Overlay structure so that each node can mantan the consstent stream flow drecton [21]. The resultng networ s called a P2P multpath networ. In recever-drven networng, the recever node controls the download schedule durng meda transmsson based on the estmaton of the current networ status. One control method s the "Parent selecton" [20]; Therefore, parent selecton s addressed n order to mprove the end-to-end delay, the streamng rate and dstorton rate of lve streamng n the P2P multpath networ. 2.3 Operaton Scenaro and Processes Wthout loss of generalty, the operaton scenaro of the proposed system has three aspects (Fg.1). Frst, the ssues relatng to the P2P multpath networ structure are consdered. Second, the proposed system adopts RTP streamng strateges wth nether feedbac nor pacet retransmsson. Thrd, snce the streamng rate does not affect the dstorton rate, the proposed approach does not consder the nfluence of channel codng such as FEC (forward error correcton) protecton [22]. ISSN: Issue 6, Volume 6, June 2009
3 Ths wor focuses on two types of afflated nodes: Node A as the new node. Node B as the node loses ts parents and loos for new parents. Under such condton Note A selects the sutable parents to establsh connectons usng optmzed dstorton method, through the recever-drven approach when t adds tself to a networ. The process flow and related algorthm of an apprasal process are descrbed below (Table 2). RTP Stream RTP Stream Fg.1 Operaton Scenaro of P2P Multpath Networ The algorthm can be broen down nto four parts. Frst, f Node A wshes to jon the system, t transmts the request appont parent to the server, and obtans Canddate_lst_1 from the server. After that, to begn wth through Algorthm_1 flterng out those canddate node Node A obtans the end-toend delay nformaton of all canddate nodes from Canddate_lst_1 through ICMP. Based on the prncple of [Issue 2], Node A elmnates the nodes wth RTT>δ, and generates Canddate_lst_2. Second, Node A tests each connecton on Canddate_lst_2 one by one usng Algorthm_2 (from the server passng through the canddate nodes to Node A), then gets ASR (the average streamng rate) and ADR (average dstorton rate) whch are generated durng streamng transmsson between all canddate nodes and Node_A n a perod of tmes. Thrd, Algorthm_3 s responsble for streamng test n a unt perod tme and provdes nformaton needed by Algorthm_2. It computes RTP dstorton rate wth followng nformaton: the accumulatve RTP pacet lost from canddate node, and extended hghest sequence number of the current RTP pacet. Fnally, Algorthm_Opt proceeds the most mportant wor-selects the best parent nodes. Durng the selectng proces Node A transmts these end-to-end streamng and dstorton rates to Algorthm_Opt to calculate the most approprate canddate node and establshes the networ connectons. The followng secton descrbes the mathematcal model and detals of the proposed method. Unle Node A, Node B must transmt ts degree to ts orgnal grandfather node after beng dsconnected from ts parent. Node B selects other canddate nodes accordng to the above procedure once the degree out values of all grandfather nodes have been dstrbuted. Snce ths process nvolves mult-stage dstrbuted handlng, t can be appled to large scale networ and therefore s consstent wth the man am of ths study. Table2. Parent Selecton Algorthm Algorthm_1 Input Canddate_lst_1 1.For each canddate_nodes n Canddate_lst_1 [1] Node_A gets the RTT value from canddate nodes through ICMP protocol. 2.End for 3.Flter n [1] the canddate nodes wth RTT value> δ away, and the remanng nodes wll output as Canddate_lst_2 4.Output(Canddate_lst_2) Algorthm_2 Input Canddate_lst_2 1. Use all avalable downln connecton of Node_A, proceed: 2.For each canddate_nodes n Canddate_lst_2 [1] Call Algorthm-3 to calculate ASR and ADR of connecton provded by these nodes [2]Save the nformaton whch Algorthm-3 sends bac nto canddate_lst_3 [3] If the dstorton rate of ths connecton s the lowest on Node_A, eep ths connecton, otherwse Node_A stops acceptng that stream flow connecton 1. 3.End for 4.Output(Canddate_lst_3) 1 :f avalable downln connecton of Node_A s only one, drectly test next canddate node. Algorthm_3 Input canddate node 1. Requre that canddate_node sends the stream comng from the server to Node_A, establsh connecton 2. Record the followng nformaton from RR pacet 1 : A.cumulatve_number_of_pacet lost(cpl) : accumulatve RTP pacet lost from canddate node. B.extended_hghest_sequence_number_receved (EHS)(32bt) : low 16 bts record the maxmal sequence number of the current RTP pacet, and Hgh 16 bts are sequence cycles. 3. Compute RTP dstorton rate (DR): DR= ( RR2[ CPL] RR1[ CPL]) ( RR [ EHS] RR [ EHS]) 2 1 ISSN: Issue 6, Volume 6, June 2009
4 base on two consecutve RR pacets (RR 1 and RR 2 ) and the travelng tme from RR 1 to RR Compute statstcs for average streamng rate (SR). 5. Output (SR, DR). 1: RR(Receve Report) s conveyed n exstng RTCP pacets. Algorthm_Opt 1 Input Canddate_lst_3 1.For =1 to m 2 [1] Get th path nformaton form server,(nclude N, P, p, mn(b ) n P ) [2] Compute the streamng btrate r and dstorton rate p of th path 3. 2 End for. 3.Intal J m (π)=[j(π 1 ),J(π 2 ),.., J(π m )], π 1 =[P 1 ], π 2 =[P 1, P 2 ],..,π m =[P 1,..,P m ], arrange the networ paths s ascendng order of ther p 4.For = 1 to m do [1] If [Frst Termnated Condton] s hold then brea; [2] Compute D(π ),R(π ) [3] If [Second Termnated Condton] s hold then brea; [4] Compute J(π ) 5.End for 6.Comput J*(π)= arg mn π J m (π) 7.Output the optmzed parent node set:j*(π) 1 : the tme complexty of Algorthm_Opt s O(N) 2 : m equal to the number of Canddate_lst_3 3 : th path s the dsjont path from server to Node_A through canddate node N 3 Optmzed Parents Selecton wth R-D Algorthm An optmzed parent selecton algorthm s descrbed to ensure that Node A selects the most approprate canddate nodes (such that stream paths have mnmal end-to-end dstorton scenaros based on a recever-drven P2P multpath networ. The related mathematcal model and method of the algorthm s descrbed below. Step 1: Identfy the method of quantzng end-to-end dstorton, n terms of [Issue 3]. Step 2: Establsh the basc flow model for P2P multpath networng n order to analyze the relevant flow constrants. Step 3: Acheve the multpath streamng characterstcs based on the stream features (streamng rate and dstorton rate) and the analyss results n Step 2. Step 4: Ln the nformaton provded by Step 3 wth the quantzaton process of Step 1, and buld the optmzed parent selecton algorthm. 3.1 Dstorton Model and Related Constrants Frst, the end-to-end dstorton of stream flow n the system s measured usng ths well-nown formula [23]: J=J R +J D.. (1). Here, J denotes the end-to-end dstorton, and comprses the expected rate J R and expected dstorton J D. The J R depends on the average streamng rate R (also called encodng rate), and J D s drven by p(rtt > τ) (where RTT s roundtrp tme, and p(rtt > τ) denotes the average dstorton rate that a pacet transmtted at sender tme T s s not receved at recever by T s +τ, whether lost or smply delayed, and τ < δ). A smple P2P multpath networ M(N,L) s used to explan the requred relevant parameter (see Fg. 2). N denotes a peer node, and N N. L s the networ ln between two adjacent nodes N and N j, and L L. p represents the dstorton rate of the stream flow F passng through L, and p [0,1). b (o ) ndcates the upln bandwdth of N ; b () s the downln bandwdth of N. P,j denotes the stream path from N to N j (such as P 1,5 ). Fnally, r,j represents the streamng rate of P,j. The relevant constrants n the P2P multpath networ are descrbed below. [Constrant 1: End-to-End Flow Constrants] The streamng rate r,j of drect connecton between two adjacent nodes N and N j s not greater then the mnmum value between the upln bandwdth of N and the downln bandwdth of N j, such that r, j mn( b ( o), b j( ) ). [Constrant 2: Downln Flow Constrants] If node N has many flow resources wth stream paths between N and the resource then the sum of the streamng rate passng through all the paths s no larger than the downln bandwdth of the node N, such that P M r b. j,, j j, ( ) [Constrant 3: Upln Flow Constrants] If a node N has many flow recevers wth stream path then the sum of the streamng rate passng through all the paths s no greater than the upln bandwdth of N, such that P M r b., j, j, j ( o) [Constrant 4: Flow Conservaton Constrants] If flow F passes through node N, then the streamng ISSN: Issue 6, Volume 6, June 2009
5 rate r ( o ) r ( ) of F nto N s equal to the streamng rate of F leavng N : b 1 ( O ) b 2 ( O ) ( ) = r ( o, and ) = r ) b 3 ( I ) b 3 ( O ) r F ( r F ( o). b 4 ( I ) b 5 ( I ) Fg.2 Dagram of a Basc P2P Multpath Networ 3.2 Optmzed Method and Related Lemmas The system adopts the parameters of end-to-end dstorton n P2P multpath networ derved n the thrd step for canddate selecton [13][14]. Thu f the system can calculate the J R and J D of stream flows from the server that pass through canddate nodes to Note A, then t can evaluate the total endto-end dstorton J through formula(1). Every the Scenaro of Fg. 1 can be transformed nto the P2P multpath networ dagram of Fg. 3, the notaton of Fg. 3 s explaned as follows. Fg.3 Operaton Scenaro of a P2P Multpath Networ 1. N s denotes the server, and A denotes Node A. 2. The degree table on the rght hand sde of Fg. 3, based on [Constrant 3], shows the restrctons of the upln and the downln numbers of all user nodes. 3. The b of L between adjacent nodes N and N j replaces mn( b ( o), b j( ) ). Defne P as the stream path from N s to N, and defne p as the dstorton rate of P. Defne r as the streamng rate of P (note that f more than one dsjont stream path s found from N s to N, these dsjont paths are ndcated as P (), by analogy wth p () and r () ) 4. All stream paths on the P2P multpath networ are mared. Ln L 1 s shared by paths P S,4(1) and P S,5(1), and L 2 s shared by P S,4(2) and P S,5(2), These shared lns on dfferent paths are called bottlenec lns. The canddate nodes (N 3, N 4, N 5 ) for Node A are also mared, and the possble lns are gven by { La, Lb, Lc} (These lns are adopted to estmate the end-to-end dstorton J ). The quantzaton procedures for the end-to-end dstorton are followed to obtan all r and p, and thus obtan the followng deducton based on the multpath characterstc and the above constrants. [Lemma 1: Server Flow Budget] The total streamng bandwdths of drect-ln from N s to adjacent nodes N s no larger than the upln bandwdths of N s such that b s( o) r. [Proof] Snce the server s regarded as a specfc node wth no downln bandwdth, ICP can estmate the upln bandwdth of N s n the networ center by [Constrant 3]. Ths bandwdth s no less than the streamng rate obtaned by clents lned drectly to servers. [Lemma 2: Server-to-end Flow Rate] In a P2P Multpath networ, each streamng rate of the dsjont path from N s to N s no greater than the mnmal networ bandwdth of all lns n the path, such that: r s mn { b }., L P [Proof] Accordng to [Constrant 1] and [Constrant 4], any flow F passes through adjacent nodes N and N j wth r, j = r ( ) = r j ( o ) mn( b ( o), b j( ) ). Snce P s conssts of the lns connectng contnuous adjacent nodes from N s to N, the mnmal b of all lns can be obtaned by multple comparson. The streamng rate r s computed from ths mnmal b. [Lemma 3: Server-to-End Dstorton Rate] The dstorton rate p of path P n a P2P multpath networ s wrtten as. p s = 1 (1 p ), L P s, [Proof] The dstorton rate n the prevous operaton scenaro s not affected by the streamng rate, and s derved ndependently for each ln. Therefore, p of each L s not nfluenced by other lns. Hence, get the total dstorton rate p s calculated by dervng the value (1- p ) of each ln sesson. The computaton formula of r and p can be confrmed by applyng [Lemma 2] and [Lemma 3] [22][24][25]. The Langrangan formulaton and polcy π s adopted to quantze (1), as: J m (π)=( J(π 1 ), J(π 2 ),, J(π m )).. (2), and J(π )=λ 1 [R(π )] η + D(π ). (3) p j r j Amog R(π ) = r s, j, D(π )= λ 2 (4) r j = 1 j= 1 j ISSN: Issue 6, Volume 6, June 2009
6 In ths wor, polcy π conssts of a set of vectors {π 1,.., π m } as π 1 ={P 1 },,π m ={P 1,.., P m }, and m s the number of all nodes n canddate_lst_2. π s the paths set from N s to Node A through the canddate nodes. Thu λ 1 and λ 2 are postve Langrangan multplers and η [ 1,0]. The system generates the dstorton metrc J m (π) based on two factor namely the restrctons mposed by the operaton scenaro, and the most sutable polcy π *. The system generates the dstorton metrc J m (π) by runnng Algorthm_Opt to calculate nodes n Canddate_lst_2 (accordng to formula(2)). The close-to-optmal dstorton J * s thus obtaned from the mnmum argument of J m (π): J * =arg mn J ( π ) = arg mn (J(π 1 ),, J(π m ))..(5) π m π Formulae (3) and (4) reveal that the orderng approach of polcy π nfluences the result of formula (5). Therefore, experments were performed to fnd an orderng method of polcy π * that yelds J *. Experment A: A node was connected to two parent nodes (denoted as N 3, N 1 and N 2 of Fg 2), where N 1 and N 2 separately transmt stream flows F 1 and F 2 to N 3, p 1 and p 2 are fxed and p 1 < p 2. Based on [Constrant 2], b3 = 1M and r 1,3 + r2,3 = b3 were set so that r2,3 = b3 r 1,3. The value of r 1,3 was rased gradually from 0 bps to 1 Mbps to observe the dstorton MSE (mean-square-error) of N 3. Fgure 4 shows the expermental results. The transmsson dstorton gradually decreased as r 1,3 ncreased (and r 2,3 decreased), and was lowest when r1,3 = b3, r2, 3 = 0. Thu f a node has more than one source node then t has mnmal dstorton when all of ts downln bandwdths are assgned to the recever node wth the lowest dstorton rate. Fg. 4 Experment for Measurng End-to-End Dstorton Experment B: A node was connected to two chld nodes (see N 3, N 4 and N 5 of Fgure. 2; N 4 and N 5 are chld nodes of N 3 ); N 3 transmtted stream flows F 3 and F 4 to N 4 and N 5 ; p 3 and p 4 were fxed, and p 3 < p 4. Accordng to [Constrant 3], b4 = 1M and b 4 = r3,4 + r3,5 were set such that r 3, 5 = b4 r3,4. The expermental results produced the same concluson as those of experment A. Thu f a node has more than one recevng node then ts dstorton s optmal when t assgns ts upln bandwdths to the recever node wth the lowest dstorton rate. The path arrangement order for polcy π can be determned from these expermental results as ascendng order for the dstorton rate of all paths. Stll, three ssues need to be solved to obtan the optmal J * n the proposed algorthm. 1. Gven that the sequence of P n polcy π s n ascendng order of the dstorton rate, then the streamng rates of P formng soluton J * need to be determned. 2. The streamng rate of the bottlenec ln to all paths P allocated by the system needs to be consdered. 3. The termnaton condtons of the algorthm must be dentfed. The frst ssue s deduced n terms of the nfluence of J ( π ) on the varance of r. Thu formula (4) s transformed nto (3). The partal dfferental form of J ( π ) s then taen, resultng n the followng Formula (6): J (π ) 1 r, (,, ) j s j ps p η s j = λ1η ( r, ) r s λ.(6) ( r, ) s Accordng to ths formula, J ( π ) decreases when the value of formula (6) s negatve, whle J ( π ) s ncreased when the value of the formula (6) s postve. Addtonally, the parametrc propertes of formulae (3) and (4) demonstrate that formula (6) s always negatve f =1. Thu dsregardng the streamng rate r s (0, mn { b }) based on [Lemma, 1 L Ps,1 2], the path P 1 wth the mnmal dstorton rate s always part of J * (whch s why Algorthm_2[3] must eep a stream path wth the mnmal dstorton rate). To obtan the value of the streamng rate for other paths that may form part of J *, formula (6) = 0 * s set n order to obtan the extreme r of r : * λ2 1+ η r = [ ( p p j ) r j j ] r j j (7) λ1η Accordng to formula (7), streamng rate r = 0 for P ( 2 m ). The dstorton of J ( π ) ncreases * * when r < rs,, and decreases when rs, > rs,. Therefore, a possble component of J * s 1 ISSN: Issue 6, Volume 6, June 2009
7 * r s = mn { b } when r s < mn { b }. Consequently,, L P, L Ps, for all J ( π ) ( 2 m), the soluton of J * may nclude r = 0 and r = mn { b }. s, L P Fgure 2 llustrates the second problem. A bottlenec ln can be defned as a ln where the end nodes of the path have more than one resource node or recevng node. Hence, based on [Constrant 2], [Constrant 3] and [Constrant 4], the paths sharng the same bottlenec ln are separated nto fully ndependent paths. The bandwdth of the bottlenec ln s allocated to these dsjont path accordng to the concluson of the frst problem. Therefore, the dstrbuted prncple can be nferred as follows: [Lemma 4: Bottlenec-ln Bandwdth Sharng] In a P2P multpath networ, f one bottlenec ln s shared by many stream path then t allocates ts bandwdth n a greedy way, begnnng wth the path P wth the lowest dstorton rate, and sets the streamng rate accordng to mn { b (the last one L P s, P j, the allocated streamng rate may be less than mn { b } ). L Ps, j The termnaton condton of the algorthm s now consdered. Frst, accordng to [Issue 1] and [Lemma 4], Node A can choose from numbers n=mn(m,degree n (1)) n canddate_lst_3, ndcatng the Frst Termnaton Condton (Judgment reference for Algorthm_Opt 4[1]). The dstorton resultng from N afflatng wth a new end-to-end stream path may rse f r <, accordng to * rs, formula (4) and answers from the frst and second problems. In summary, not all canddate nodes necessarly be chosen, and the frst termnaton condton s as follows (combnng Eqs. (5) and (6)): J m ( π ) 0, r 1= mn { b }, r = mn { b } rs, L Ps, 1 L Ps, J m ( π ) 0, r 1= mn { b }, rs, = 0 rs, L Ps, 1 Therefore, the second termnated condton s as follows (Judgment reference for Algorthm_Opt 4[3]): λ1η η (Condton-2A): D( π ) [ R( π )] + p s,, λ2 = mn { b }, r = mn { } r s, 1 L P 1 } L P λ1η η (Condton-2B): D( π ) [ R ( π )] + p s,, λ2 r, 1= mn { b }, r, = 0 s L P 1 s b Thu n a P2P lve stream networ, the calculaton of the end-to-end dstorton for J m (π) termnates the computaton when ether the Frst Termnaton Condton or the Second Termnaton Condton occurs. In summary, Algorthm_Opt based on the lemmas n ths chapter s shown after suspendng the calculaton of the end-to-end dstorton. The optmzed soluton set J * s derved by computng the optmal answer of J m (π). 4 Smulaton Results Ths chapter descrbes the smulaton results obtaned wth the descrbed operaton scenaro and algorthms. The P2P stream networ was constructed on NS2 and MyEvalvd (relevant parameters n Table 3) [26][27]. The networ comprsed 20 nodes (wth the frst node labeled N s ). Node nformaton was recorded n a matrx M as M(degree n (N ), degree out (N ), p,j, b,j ) contanng the degree n, degree out of N, and b,j, p,j of the drect-ln value wth N j. The followng three networ scenaros were smulated. Scenaro-1) Wred networ pattern: lns between nodes have hgh bandwdth and low dstorton rate. Scenaro-2) Wreless networ pattern: lns between nodes have low bandwdth and hgh dstorton rate. Scenaro-3) Hybrd networ pattern: a hybrd of above two pattern where N s s connected to a wred networ, and clents can access streams va a wred or wreless ln. Table 3: Parameters for Smulaton Scenaros p -max p -mn b -max b -mn Smulaton M bps 1 Mbps Smulaton bps 128 pbs Smulaton M bps 128 pbs degree n/out =[1,4], δ=200mτ=100mθ=[2.00,4.00]sec η=-0.87,λ 1 =1.8x10 5,λ 2 =1624, The dstorton smulatons were performed accordng to these three patterns (total 18 smulaton networs). Wth the b,j, p,j values of each of the twenty nodes n M assgned from parameter s range n Table 3 accordng to the followng algorthms. The values of N of matrx M gradually afflate system n tme nterval θ (wth α as the path parameter, and S=mn(3,degree n (N ))). 1.A DR : a sngle-path selecton algorthm wth dstorton rate gven by α [1-Π (1-p )]. ISSN: Issue 6, Volume 6, June 2009
8 2.A DR3 : an S-path selecton algorthm wth dstorton rate measured as α [1-Π (1-p )], usng the maxmum avalable value of S. 3.A GP : a sngle-best-path selecton algorthm wth good-put value gven by α b p. 4.A GP3 : an S-path selecton algorthm wth good-put value measured as α b p. 5.A MP : a maxmum-paths algorthm, usng the maxmum avalable degree n number of N. In the NS2 networ smulaton scenaro, the networ maxmum transmsson unt (MTU) was set to 1000 bytes worth of vdeo data. The foreman_cf sequence was joned wth tself to form 2700-frame vdeo stream encoded at 30 frames per second. Each GOP (Group of Pctures) conssted of 9 frames (pctures). Each newly afflated node passes through the paths chosen by those sx algorthm and gans the pacets from N s followng an FCFS schedulng. The three networ scenaros were each smulated 100 tmes. The PSNR (Pea Sgnal-to- Nose Rato) between the orgnal and decoded sequences was calculated, and the average total endto-end dstorton was then computed. Fgures 5, 6 and 7 present the expermental results. The proposed algorthm had the hghest average effcency among the sx algorthms tested. The expermental results are summarzed as follows: 1. In the Scenaro-1 smulaton, calculated dstorton of A MF was closest to A OPT. Clearly, a new node n an envronment wth a lower dstorton rate and hgher streamng rate can obtan a better stream qualty by usng more alternatve paths to transmt stream flow thus reducng the end-to-end dstorton. 2. In the Scenaro-2 smulaton, the dstorton calculaton result of A DR compared to others s closer to A OPT, n contrast to Scenaro-1. A new node n an envronment wth a hgher dstorton rate and lower streamng rate t cannot decrease the dstorton effect by usng many transmsson paths. In ths case, the closest to optmal streamng qualty s obtaned by streamng meda only through the path wth the lowest dstorton rate. 3. The smulaton Scenaro-3, whch combnes propertes of both Scenaro-1 and Scenaro-2, confrmed the smulaton results of both those scenaros. Algorthm A DR acheved less total dstorton than A GP algorthm n choosng canddate nodes. To compare smulaton results wth actual executon, experments were performed wth A OPT, A MF and A DR (smulaton 1, Node A pattern). Fgure 8 shows the expermental result. The result sequences of the smulated and actual dstorton for each algorthm were smlar. Next, 18 smulaton networs were constructed by sx algorthms randomly removes 5 nodes excludng N s. The chld nodes were requred to loo for other parent and generate connected paths under each of the sx algorthms. Each smulaton was performed 50 tme and the average dstorton value was obtaned. Fgures 9, 10 and 11 llustrate the expermental results. The results of these smulatons were smlar to those of the orgnal smulatons. These smulaton results demonstrate that the user node can reduce overall networ dstorton by selectng paths wth hgh streamng rates and low dstorton rates for meda transmsson under varous P2P multpath networ scenaros AO P T ADR 1 ADR 3 AG P 1 AG P 3 AMF Fg.5 Dstorton Comparson for Scenaro-1 (Type: NodeA) AO P T ADR 1 ADR3 AG P 1 AG P 3 AMF Fg.6 Dstorton Comparson for Scenaro-2 (Type: NodeA) AO P T ADR 1 ADR3 AG P 1 AG P 3 AMF Fg.7 Dstorton Comparson for Scenaro-3(Type: NodeA) ISSN: Issue 6, Volume 6, June 2009
9 Smulaton Experment AOPT ADR1 AMF wor proposes a recever-drven dynamc node jon algorthm for a P2P multpath networ. Smulaton results ndcate that the proposed algorthm performs better than others under varous networ crcumstances. The proposed dstrbuted structure and algorthm can decrease the computatonal loadng of servers and mprove the overall streamng qualty. Fg8. Dstorton Comparson for Smulaton wth Experment AO P T ADR1 ADR3 AG P 1 AG P 3 AMF Fg.9 Dstorton Comparson for Scenaro-1 (Type: Node B) AO P T ADR1 ADR3 AG P 1 AG P 3 AMF Fg.10 Dstorton Comparson for Scenaro-2 (Type: Node B) PSNR(DB) AO P T ADR1 ADR3 AG P 1 AG P 3 AMF Fg.11 Dstorton Comparson for Scenaro-3 (Type: Node B) 5 Concluson Owng to the shared stream flows between front end P2P lve streamng s affected by bandwdth lmtaton and the clent computer envronment. In addton to summarzng stream qualty-related ssues and analyzng n depth the end-to-end dstorton constrants n the operaton scenaro, ths References: [1] Aama, [2] Lmelght Networ rs.com. [3] Hucheng Ch, Qan Zhang, Juncheng Ja and Xuemn (Sherman) Shen, "Effcent Search and Schedulng n P2P-based Meda-on-Demand Streamng Servce," IEEE Journal on Selected Areas n Communcaton Vol. 25, No. 1, Jan [4] We Cheng, We Tsang Oo,"Recever-Drven Vew-Dependent Streamng of Progressve Mesh,"n proceedng of NOSSDAV 2008, Braunschweg, Germany, [5] Jn-feng Zhang, Rong-gang Wang, Jan-we Nu, Yuan Dong, Ha-la Wang,"Rate-dstorton optmzed btstream swtchng for peer-to-peer lve streamng," Journal of Zhejang Unversty Scence, Vol. 9, pp , [6] Prashant Ramanathan, Mar Kalman, Bernd Grod, "Rate-Dstorton Optmzed Interactve Lght Feld Streamng," IEEE Transactons on Multmeda 9(4): pp , [7] Dan Jurca, Pascal Frossard,"Dstrbuted Meda Rate Allocaton n Multpath Networ" Image Communcaton, Vol. 23, pp , Nov [8] Feng Wang,Yongqang Xong,Jangchuan Lu,"mTreebone: A Hybrd Tree/Mesh Overlay for Applcaton-Layer Lve Vdeo Multcast," Internatonal Conference on Dstrbuted Computng System ICDCS 07, pp ,, Burnaby,Canada, [9] Bo Lu,Yansheng Lu,Y Cu,Yuan Xue,"A measurement study on AS-aware P2P streamng stratege" ChnaCom2008, pp ,aug [10] Xaojun He, Yong Lu, Keth W. Ros"Inferrng Networ-Wde Qualty n P2P Lve Streamng System" IEEE Journal on Selected Areas n Communcaton" 25(9): pp , 2007 [11] PPLve, [12] PPStream, ISSN: Issue 6, Volume 6, June 2009
10 [13] E. Setton, J. Noh and B. Grod, "Rate- Dstorton Optmzed Vdeo Peer-to-Peer Multcast Streamng", n ACM P2PMMS, [14] P. Mundur, R. Smon, and A. Sood, "End-toend Analyss of Dstrbuted Vdeo-on-Demand System" IEEE Transactons on Multmeda, Volume 6, Number 1, Feb [15] J. Noh, S. Deshpande, "A Method for P2P Streamng System Jon Latency Reducton Usng Preferred Peer Lst," Proc. IEEE 5th Consumer Communcatons and Networng Conference (CCNC 2008), USA, Jan [16] Nazann Maghare, Reza Rejae, "Understandng Mesh-based Peer-to-Peer Streamng," Proceedngs of the Internatonal Worshop on Networ and Operatng Systems Support for Dgtal Audo and Vdeo, pp , Newport, Rhode Island, May [17] Nazann Maghare, Reza Rejae, Yang Guo,"Mesh or Multple-Tree: A Comparatve Study of Lve P2P Streamng Approache" Proceedngs of IEEE INFOCOM, pp , Alasa, USA, May [18] Damano Carra, Renato Lo Cgno, Ernst W. Bersac: Graph Based Analyss of Mesh Overlay Streamng Systems. IEEE Journal on Selected Areas n Communcatons 25(9): pp , [19] Xaojun He, Yong Lu and Keth Ros"IPTV over P2P Stream networs: the Mesh-pull Approach", IEEE Communcatons Magazne, Feb [20] N Maghare, R Rejae, "PRIME: Peer-to-Peer Recever-drIven MEsh-Based Streamng," 26th IEEE Internatonal Conference on Computer Communcaton pp , [21] Irfan Sherff, Elzabeth M. Beldng-Royer, "Multpath Selecton n Mult-rado Mesh Networ " Broadband Communcaton Networs and System pp.1-11, [22] L. Golubch, J. Lu, T. Tung, A. Chow, and W. Lee, Mult-path contnuous meda streamng: What are the benefts? ACM Journal of Performance Evaluaton, vol. 49, no. 1-4, pp , Sept [23] P. A. Chou and Z. Mao, "Rate-dstorton optmzed sender-drven streamng over besteffort networ" IEEE Worshop on Multmeda Sgnal Processng, pp , France, Oct [24] A. C. Begen, Y. Altunbasa, O. Ergun, and M. H. Ammar, "Multpath selecton for multple descrpton vdeo streamng over overlay networ" Sgnal Processng: Image Communcaton, vol. 20, pp , [25] Jacob Chaare Sangeun Han, Bernd Grod, "Layered codng vs. multple descrptons for vdeo streamng over multple path Multmeda System,10(4), pp , 2005 [26] J. Klaue, B. Rathe, and A. Wolsz, "EvalVd - A Framewor for Vdeo Transmsson and Qualty Evaluaton", In Proc. of the 13th Internatonal Conference on Modellng Technques and Tools for Computer Performance Evaluaton, pp , Urbana, Illno USA, Sept [27] ISSN: Issue 6, Volume 6, June 2009
Parallelism for Nested Loops with Non-uniform and Flow Dependences
Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr
More informationARTICLE IN PRESS. Signal Processing: Image Communication
Sgnal Processng: Image Communcaton 23 (2008) 754 768 Contents lsts avalable at ScenceDrect Sgnal Processng: Image Communcaton journal homepage: www.elsever.com/locate/mage Dstrbuted meda rate allocaton
More informationLoad-Balanced Anycast Routing
Load-Balanced Anycast Routng Chng-Yu Ln, Jung-Hua Lo, and Sy-Yen Kuo Department of Electrcal Engneerng atonal Tawan Unversty, Tape, Tawan sykuo@cc.ee.ntu.edu.tw Abstract For fault-tolerance and load-balance
More informationSimulation Based Analysis of FAST TCP using OMNET++
Smulaton Based Analyss of FAST TCP usng OMNET++ Umar ul Hassan 04030038@lums.edu.pk Md Term Report CS678 Topcs n Internet Research Sprng, 2006 Introducton Internet traffc s doublng roughly every 3 months
More informationLoad Balancing for Hex-Cell Interconnection Network
Int. J. Communcatons, Network and System Scences,,, - Publshed Onlne Aprl n ScRes. http://www.scrp.org/journal/jcns http://dx.do.org/./jcns.. Load Balancng for Hex-Cell Interconnecton Network Saher Manaseer,
More informationVideo Proxy System for a Large-scale VOD System (DINA)
Vdeo Proxy System for a Large-scale VOD System (DINA) KWUN-CHUNG CHAN #, KWOK-WAI CHEUNG *# #Department of Informaton Engneerng *Centre of Innovaton and Technology The Chnese Unversty of Hong Kong SHATIN,
More informationEnd-to-end Distortion Estimation for RD-based Robust Delivery of Pre-compressed Video
End-to-end Dstorton Estmaton for RD-based Robust Delvery of Pre-compressed Vdeo Ru Zhang, Shankar L. Regunathan and Kenneth Rose Department of Electrcal and Computer Engneerng Unversty of Calforna, Santa
More informationEfficient Distributed File System (EDFS)
Effcent Dstrbuted Fle System (EDFS) (Sem-Centralzed) Debessay(Debsh) Fesehaye, Rahul Malk & Klara Naherstedt Unversty of Illnos-Urbana Champagn Contents Problem Statement, Related Work, EDFS Desgn Rate
More informationSRB: Shared Running Buffers in Proxy to Exploit Memory Locality of Multiple Streaming Media Sessions
SRB: Shared Runnng Buffers n Proxy to Explot Memory Localty of Multple Streamng Meda Sessons Songqng Chen,BoShen, Yong Yan, Sujoy Basu, and Xaodong Zhang Department of Computer Scence Moble and Meda System
More informationNUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS
ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana
More informationA mathematical programming approach to the analysis, design and scheduling of offshore oilfields
17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 A mathematcal programmng approach to the analyss, desgn and
More informationCompiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz
Compler Desgn Sprng 2014 Regster Allocaton Sample Exercses and Solutons Prof. Pedro C. Dnz USC / Informaton Scences Insttute 4676 Admralty Way, Sute 1001 Marna del Rey, Calforna 90292 pedro@s.edu Regster
More informationImprovement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration
Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,
More informationVirtual Machine Migration based on Trust Measurement of Computer Node
Appled Mechancs and Materals Onlne: 2014-04-04 ISSN: 1662-7482, Vols. 536-537, pp 678-682 do:10.4028/www.scentfc.net/amm.536-537.678 2014 Trans Tech Publcatons, Swtzerland Vrtual Machne Mgraton based on
More informationThe Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique
//00 :0 AM Outlne and Readng The Greedy Method The Greedy Method Technque (secton.) Fractonal Knapsack Problem (secton..) Task Schedulng (secton..) Mnmum Spannng Trees (secton.) Change Money Problem Greedy
More informationVirtual Memory. Background. No. 10. Virtual Memory: concept. Logical Memory Space (review) Demand Paging(1) Virtual Memory
Background EECS. Operatng System Fundamentals No. Vrtual Memory Prof. Hu Jang Department of Electrcal Engneerng and Computer Scence, York Unversty Memory-management methods normally requres the entre process
More informationEvaluation of an Enhanced Scheme for High-level Nested Network Mobility
IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.15 No.10, October 2015 1 Evaluaton of an Enhanced Scheme for Hgh-level Nested Network Moblty Mohammed Babker Al Mohammed, Asha Hassan.
More informationHermite Splines in Lie Groups as Products of Geodesics
Hermte Splnes n Le Groups as Products of Geodescs Ethan Eade Updated May 28, 2017 1 Introducton 1.1 Goal Ths document defnes a curve n the Le group G parametrzed by tme and by structural parameters n the
More informationCluster Analysis of Electrical Behavior
Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School
More informationAn Optimal Algorithm for Prufer Codes *
J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,
More informationShared Running Buffer Based Proxy Caching of Streaming Sessions
Shared Runnng Buffer Based Proxy Cachng of Streamng Sessons Songqng Chen, Bo Shen, Yong Yan, Sujoy Basu Moble and Meda Systems Laboratory HP Laboratores Palo Alto HPL-23-47 March th, 23* E-mal: sqchen@cs.wm.edu,
More informationDetermining the Optimal Bandwidth Based on Multi-criterion Fusion
Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn
More informationSmoothing Spline ANOVA for variable screening
Smoothng Splne ANOVA for varable screenng a useful tool for metamodels tranng and mult-objectve optmzaton L. Rcco, E. Rgon, A. Turco Outlne RSM Introducton Possble couplng Test case MOO MOO wth Game Theory
More informationDelay Variation Optimized Traffic Allocation Based on Network Calculus for Multi-path Routing in Wireless Mesh Networks
Appl. Math. Inf. Sc. 7, No. 2L, 467-474 2013) 467 Appled Mathematcs & Informaton Scences An Internatonal Journal http://dx.do.org/10.12785/ams/072l13 Delay Varaton Optmzed Traffc Allocaton Based on Network
More informationCircuit Analysis I (ENGR 2405) Chapter 3 Method of Analysis Nodal(KCL) and Mesh(KVL)
Crcut Analyss I (ENG 405) Chapter Method of Analyss Nodal(KCL) and Mesh(KVL) Nodal Analyss If nstead of focusng on the oltages of the crcut elements, one looks at the oltages at the nodes of the crcut,
More informationAADL : about scheduling analysis
AADL : about schedulng analyss Schedulng analyss, what s t? Embedded real-tme crtcal systems have temporal constrants to meet (e.g. deadlne). Many systems are bult wth operatng systems provdng multtaskng
More informationProblem Definitions and Evaluation Criteria for Computational Expensive Optimization
Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty
More informationLearning the Kernel Parameters in Kernel Minimum Distance Classifier
Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department
More informationA Binarization Algorithm specialized on Document Images and Photos
A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a
More informationSecurity Enhanced Dynamic ID based Remote User Authentication Scheme for Multi-Server Environments
Internatonal Journal of u- and e- ervce, cence and Technology Vol8, o 7 0), pp7-6 http://dxdoorg/07/unesst087 ecurty Enhanced Dynamc ID based Remote ser Authentcaton cheme for ult-erver Envronments Jun-ub
More informationTerm Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task
Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto
More informationNetwork Coding as a Dynamical System
Network Codng as a Dynamcal System Narayan B. Mandayam IEEE Dstngushed Lecture (jont work wth Dan Zhang and a Su) Department of Electrcal and Computer Engneerng Rutgers Unversty Outlne. Introducton 2.
More informationConstructing Minimum Connected Dominating Set: Algorithmic approach
Constructng Mnmum Connected Domnatng Set: Algorthmc approach G.N. Puroht and Usha Sharma Centre for Mathematcal Scences, Banasthal Unversty, Rajasthan 304022 usha.sharma94@yahoo.com Abstract: Connected
More informationAn efficient iterative source routing algorithm
An effcent teratve source routng algorthm Gang Cheng Ye Tan Nrwan Ansar Advanced Networng Lab Department of Electrcal Computer Engneerng New Jersey Insttute of Technology Newar NJ 7 {gc yt Ansar}@ntedu
More information3. CR parameters and Multi-Objective Fitness Function
3 CR parameters and Mult-objectve Ftness Functon 41 3. CR parameters and Mult-Objectve Ftness Functon 3.1. Introducton Cogntve rados dynamcally confgure the wreless communcaton system, whch takes beneft
More informationPriority-Based Scheduling Algorithm for Downlink Traffics in IEEE Networks
Prorty-Based Schedulng Algorthm for Downlnk Traffcs n IEEE 80.6 Networks Ja-Mng Lang, Jen-Jee Chen, You-Chun Wang, Yu-Chee Tseng, and Bao-Shuh P. Ln Department of Computer Scence Natonal Chao-Tung Unversty,
More informationQoS-aware routing for heterogeneous layered unicast transmissions in wireless mesh networks with cooperative network coding
Tarno et al. EURASIP Journal on Wreless Communcatons and Networkng 214, 214:81 http://wcn.euraspournals.com/content/214/1/81 RESEARCH Open Access QoS-aware routng for heterogeneous layered uncast transmssons
More informationA Fast Content-Based Multimedia Retrieval Technique Using Compressed Data
A Fast Content-Based Multmeda Retreval Technque Usng Compressed Data Borko Furht and Pornvt Saksobhavvat NSF Multmeda Laboratory Florda Atlantc Unversty, Boca Raton, Florda 3343 ABSTRACT In ths paper,
More informationDynamic Bandwidth Provisioning with Fairness and Revenue Considerations for Broadband Wireless Communication
Ths full text paper was peer revewed at the drecton of IEEE Communcatons Socety subject matter experts for publcaton n the ICC 008 proceedngs. Dynamc Bandwdth Provsonng wth Farness and Revenue Consderatons
More informationDESIGNING TRANSMISSION SCHEDULES FOR WIRELESS AD HOC NETWORKS TO MAXIMIZE NETWORK THROUGHPUT
DESIGNING TRANSMISSION SCHEDULES FOR WIRELESS AD HOC NETWORKS TO MAXIMIZE NETWORK THROUGHPUT Bran J. Wolf, Joseph L. Hammond, and Harlan B. Russell Dept. of Electrcal and Computer Engneerng, Clemson Unversty,
More informationInternet Traffic Managers
Internet Traffc Managers Ibrahm Matta matta@cs.bu.edu www.cs.bu.edu/faculty/matta Computer Scence Department Boston Unversty Boston, MA 225 Jont work wth members of the WING group: Azer Bestavros, John
More informationLobachevsky State University of Nizhni Novgorod. Polyhedron. Quick Start Guide
Lobachevsky State Unversty of Nzhn Novgorod Polyhedron Quck Start Gude Nzhn Novgorod 2016 Contents Specfcaton of Polyhedron software... 3 Theoretcal background... 4 1. Interface of Polyhedron... 6 1.1.
More informationEfficient Content Distribution in Wireless P2P Networks
Effcent Content Dstrbuton n Wreless P2P Networs Qong Sun, Vctor O. K. L, and Ka-Cheong Leung Department of Electrcal and Electronc Engneerng The Unversty of Hong Kong Pofulam Road, Hong Kong, Chna {oansun,
More informationSolving two-person zero-sum game by Matlab
Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by
More informationReal-time interactive applications
Real-tme nteractve applcatons PC-2-PC phone PC-2-phone Dalpad Net2phone vdeoconference Webcams Now we look at a PC-2-PC Internet phone example n detal Internet phone over best-effort (1) Best effort packet
More informationComparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments
Comparson of Heurstcs for Schedulng Independent Tasks on Heterogeneous Dstrbuted Envronments Hesam Izakan¹, Ath Abraham², Senor Member, IEEE, Václav Snášel³ ¹ Islamc Azad Unversty, Ramsar Branch, Ramsar,
More informationAn Efficient Genetic Algorithm with Fuzzy c-means Clustering for Traveling Salesman Problem
An Effcent Genetc Algorthm wth Fuzzy c-means Clusterng for Travelng Salesman Problem Jong-Won Yoon and Sung-Bae Cho Dept. of Computer Scence Yonse Unversty Seoul, Korea jwyoon@sclab.yonse.ac.r, sbcho@cs.yonse.ac.r
More informationFast Retransmission of Real-Time Traffic in HIPERLAN/2 Systems
Fast Retransmsson of Real-Tme Traffc n HIPERLAN/ Systems José A Afonso and Joaqum E Neves Department of Industral Electroncs Unversty of Mnho, Campus de Azurém 4800-058 Gumarães, Portugal {joseafonso,
More informationAnalysis of Collaborative Distributed Admission Control in x Networks
1 Analyss of Collaboratve Dstrbuted Admsson Control n 82.11x Networks Thnh Nguyen, Member, IEEE, Ken Nguyen, Member, IEEE, Lnha He, Member, IEEE, Abstract Wth the recent surge of wreless home networks,
More informationChannel 0. Channel 1 Channel 2. Channel 3 Channel 4. Channel 5 Channel 6 Channel 7
Optmzed Regonal Cachng for On-Demand Data Delvery Derek L. Eager Mchael C. Ferrs Mary K. Vernon Unversty of Saskatchewan Unversty of Wsconsn Madson Saskatoon, SK Canada S7N 5A9 Madson, WI 5376 eager@cs.usask.ca
More informationMaintaining temporal validity of real-time data on non-continuously executing resources
Mantanng temporal valdty of real-tme data on non-contnuously executng resources Tan Ba, Hong Lu and Juan Yang Hunan Insttute of Scence and Technology, College of Computer Scence, 44, Yueyang, Chna Wuhan
More informationSupport Strong Consistency for Mobile Dynamic Contents Delivery Network
Nnth IEEE Internatonal Symposum on Multmeda 27 Support Strong Consstency for Moble Dynamc Contents Delvery Networ Zhou Su, Jro Katto and Yasuho Yasuda Graduate School of Scence and Engneerng, Waseda Unversty,
More informationHybrid Non-Blind Color Image Watermarking
Hybrd Non-Blnd Color Image Watermarkng Ms C.N.Sujatha 1, Dr. P. Satyanarayana 2 1 Assocate Professor, Dept. of ECE, SNIST, Yamnampet, Ghatkesar Hyderabad-501301, Telangana 2 Professor, Dept. of ECE, AITS,
More informationFor instance, ; the five basic number-sets are increasingly more n A B & B A A = B (1)
Secton 1.2 Subsets and the Boolean operatons on sets If every element of the set A s an element of the set B, we say that A s a subset of B, or that A s contaned n B, or that B contans A, and we wrte A
More informationReducing Frame Rate for Object Tracking
Reducng Frame Rate for Object Trackng Pavel Korshunov 1 and We Tsang Oo 2 1 Natonal Unversty of Sngapore, Sngapore 11977, pavelkor@comp.nus.edu.sg 2 Natonal Unversty of Sngapore, Sngapore 11977, oowt@comp.nus.edu.sg
More informationVRT012 User s guide V0.1. Address: Žirmūnų g. 27, Vilnius LT-09105, Phone: (370-5) , Fax: (370-5) ,
VRT012 User s gude V0.1 Thank you for purchasng our product. We hope ths user-frendly devce wll be helpful n realsng your deas and brngng comfort to your lfe. Please take few mnutes to read ths manual
More informationHIERARCHICAL SCHEDULING WITH ADAPTIVE WEIGHTS FOR W-ATM *
Copyrght Notce c 1999 IEEE. Personal use of ths materal s permtted. However, permsson to reprnt/republsh ths materal for advertsng or promotonal purposes or for creatng new collectve wors for resale or
More informationSum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints
Australan Journal of Basc and Appled Scences, 2(4): 1204-1208, 2008 ISSN 1991-8178 Sum of Lnear and Fractonal Multobjectve Programmng Problem under Fuzzy Rules Constrants 1 2 Sanjay Jan and Kalash Lachhwan
More informationNeeded Information to do Allocation
Complexty n the Database Allocaton Desgn Must tae relatonshp between fragments nto account Cost of ntegrty enforcements Constrants on response-tme, storage, and processng capablty Needed Informaton to
More informationThe Codesign Challenge
ECE 4530 Codesgn Challenge Fall 2007 Hardware/Software Codesgn The Codesgn Challenge Objectves In the codesgn challenge, your task s to accelerate a gven software reference mplementaton as fast as possble.
More informationEfficient QoS Provisioning at the MAC Layer in Heterogeneous Wireless Sensor Networks
Effcent QoS Provsonng at the MAC Layer n Heterogeneous Wreless Sensor Networks M.Soul a,, A.Bouabdallah a, A.E.Kamal b a UMR CNRS 7253 HeuDaSyC, Unversté de Technologe de Compègne, Compègne Cedex F-625,
More informationAn Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation
17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 An Iteratve Soluton Approach to Process Plant Layout usng Mxed
More informationPositive Semi-definite Programming Localization in Wireless Sensor Networks
Postve Sem-defnte Programmng Localzaton n Wreless Sensor etworks Shengdong Xe 1,, Jn Wang, Aqun Hu 1, Yunl Gu, Jang Xu, 1 School of Informaton Scence and Engneerng, Southeast Unversty, 10096, anjng Computer
More informationCourse Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms
Course Introducton Course Topcs Exams, abs, Proects A quc loo at a few algorthms 1 Advanced Data Structures and Algorthms Descrpton: We are gong to dscuss algorthm complexty analyss, algorthm desgn technques
More informationVISUAL SELECTION OF SURFACE FEATURES DURING THEIR GEOMETRIC SIMULATION WITH THE HELP OF COMPUTER TECHNOLOGIES
UbCC 2011, Volume 6, 5002981-x manuscrpts OPEN ACCES UbCC Journal ISSN 1992-8424 www.ubcc.org VISUAL SELECTION OF SURFACE FEATURES DURING THEIR GEOMETRIC SIMULATION WITH THE HELP OF COMPUTER TECHNOLOGIES
More informationContent Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers
IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth
More informationA Saturation Binary Neural Network for Crossbar Switching Problem
A Saturaton Bnary Neural Network for Crossbar Swtchng Problem Cu Zhang 1, L-Qng Zhao 2, and Rong-Long Wang 2 1 Department of Autocontrol, Laonng Insttute of Scence and Technology, Benx, Chna bxlkyzhangcu@163.com
More informationS1 Note. Basis functions.
S1 Note. Bass functons. Contents Types of bass functons...1 The Fourer bass...2 B-splne bass...3 Power and type I error rates wth dfferent numbers of bass functons...4 Table S1. Smulaton results of type
More informationScheduling Remote Access to Scientific Instruments in Cyberinfrastructure for Education and Research
Schedulng Remote Access to Scentfc Instruments n Cybernfrastructure for Educaton and Research Je Yn 1, Junwe Cao 2,3,*, Yuexuan Wang 4, Lanchen Lu 1,3 and Cheng Wu 1,3 1 Natonal CIMS Engneerng and Research
More informationA Load-balancing and Energy-aware Clustering Algorithm in Wireless Ad-hoc Networks
A Load-balancng and Energy-aware Clusterng Algorthm n Wreless Ad-hoc Networks Wang Jn, Shu Le, Jnsung Cho, Young-Koo Lee, Sungyoung Lee, Yonl Zhong Department of Computer Engneerng Kyung Hee Unversty,
More informationVideo streaming over the Internet is becoming very popular and
Streamng MPEG-4 AudoVsual Objects Usng TCP-Frendly Rate Control and Unequal Error Protecton Toufk Ahmed 1, Ahmed Mehaoua 1 and Vncent Lecure 2 1 2 CNRS-PRSM LabUnversty of Versalles CRAN lab CNRS UMR 739
More informationTHere are increasing interests and use of mobile ad hoc
1 Adaptve Schedulng n MIMO-based Heterogeneous Ad hoc Networks Shan Chu, Xn Wang Member, IEEE, and Yuanyuan Yang Fellow, IEEE. Abstract The demands for data rate and transmsson relablty constantly ncrease
More informationMinimum Cost Optimization of Multicast Wireless Networks with Network Coding
Mnmum Cost Optmzaton of Multcast Wreless Networks wth Network Codng Chengyu Xong and Xaohua L Department of ECE, State Unversty of New York at Bnghamton, Bnghamton, NY 13902 Emal: {cxong1, xl}@bnghamton.edu
More informationPricing Network Resources for Adaptive Applications in a Differentiated Services Network
IEEE INFOCOM Prcng Network Resources for Adaptve Applcatons n a Dfferentated Servces Network Xn Wang and Hennng Schulzrnne Columba Unversty Emal: {xnwang, schulzrnne}@cs.columba.edu Abstract The Dfferentated
More informationScalable Video Streaming over P2P Networks: A Matter of Harmony?
211 IEEE 16th Internatonal Workshop on Computer Aded Modelng and Desgn of Communcaton Lnks and Networks (CAMAD) Scalable Vdeo Streamng over P2P Networks: A Matter of Harmony? Samr Medah 1, Toufk Ahmed
More informationConvolutional interleaver for unequal error protection of turbo codes
Convolutonal nterleaver for unequal error protecton of turbo codes Sna Vaf, Tadeusz Wysock, Ian Burnett Unversty of Wollongong, SW 2522, Australa E-mal:{sv39,wysock,an_burnett}@uow.edu.au Abstract: Ths
More informationConfiguration Management in Multi-Context Reconfigurable Systems for Simultaneous Performance and Power Optimizations*
Confguraton Management n Mult-Context Reconfgurable Systems for Smultaneous Performance and Power Optmzatons* Rafael Maestre, Mlagros Fernandez Departamento de Arqutectura de Computadores y Automátca Unversdad
More informationImproved Resource Allocation Algorithms for Practical Image Encoding in a Ubiquitous Computing Environment
JOURNAL OF COMPUTERS, VOL. 4, NO. 9, SEPTEMBER 2009 873 Improved Resource Allocaton Algorthms for Practcal Image Encodng n a Ubqutous Computng Envronment Manxong Dong, Long Zheng, Kaoru Ota, Song Guo School
More informationResearch Article. ISSN (Print) s k and. d k rate of k -th flow, source node and
Scholars Journal of Engneerng and Technology (SJET) Sch. J. Eng. Tech., 2015; 3(4A):343-350 Scholars Academc and Scentfc Publsher (An Internatonal Publsher for Academc and Scentfc Resources) www.saspublsher.com
More informationControl strategies for network efficiency and resilience with route choice
Control strateges for networ effcency and reslence wth route choce Andy Chow Ru Sha Centre for Transport Studes Unversty College London, UK Centralsed strateges UK 1 Centralsed strateges Some effectve
More informationNetwork-Driven Layered Multicast with IPv6
Network-Drven Layered Multcast wth IPv6 Ho-pong Sze and Soung C. Lew Department of Informaton Engneerng, The Chnese Unversty of Hong Kong, Shatn, N.T., Hong Kong {hpsze8, soung}@e.cuhk.edu.hk Abstract.
More informationA New Approach For the Ranking of Fuzzy Sets With Different Heights
New pproach For the ankng of Fuzzy Sets Wth Dfferent Heghts Pushpnder Sngh School of Mathematcs Computer pplcatons Thapar Unversty, Patala-7 00 Inda pushpndersnl@gmalcom STCT ankng of fuzzy sets plays
More informationK-means and Hierarchical Clustering
Note to other teachers and users of these sldes. Andrew would be delghted f you found ths source materal useful n gvng your own lectures. Feel free to use these sldes verbatm, or to modfy them to ft your
More informationSLAM Summer School 2006 Practical 2: SLAM using Monocular Vision
SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,
More informationAnalysis on the Workspace of Six-degrees-of-freedom Industrial Robot Based on AutoCAD
Analyss on the Workspace of Sx-degrees-of-freedom Industral Robot Based on AutoCAD Jn-quan L 1, Ru Zhang 1,a, Fang Cu 1, Q Guan 1 and Yang Zhang 1 1 School of Automaton, Bejng Unversty of Posts and Telecommuncatons,
More informationProper Choice of Data Used for the Estimation of Datum Transformation Parameters
Proper Choce of Data Used for the Estmaton of Datum Transformaton Parameters Hakan S. KUTOGLU, Turkey Key words: Coordnate systems; transformaton; estmaton, relablty. SUMMARY Advances n technologes and
More informationWavelet-Based Image Compression System with Linear Distortion Control
Je-Hung Lu, Kng-Chu Hung Wavelet-Based Image Compresson System wth Lnear Dstorton Control Je-Hung Lu, Kng-Chu Hung Insttute of Engneerng Scence and Technology Natonal Kaohsung Frst Unversty of Scence and
More informationAvailable online at Available online at Advanced in Control Engineering and Information Science
Avalable onlne at wwwscencedrectcom Avalable onlne at wwwscencedrectcom Proceda Proceda Engneerng Engneerng 00 (2011) 15000 000 (2011) 1642 1646 Proceda Engneerng wwwelsevercom/locate/proceda Advanced
More informationEnhanced AMBTC for Image Compression using Block Classification and Interpolation
Internatonal Journal of Computer Applcatons (0975 8887) Volume 5 No.0, August 0 Enhanced AMBTC for Image Compresson usng Block Classfcaton and Interpolaton S. Vmala Dept. of Comp. Scence Mother Teresa
More informationNAG Fortran Library Chapter Introduction. G10 Smoothing in Statistics
Introducton G10 NAG Fortran Lbrary Chapter Introducton G10 Smoothng n Statstcs Contents 1 Scope of the Chapter... 2 2 Background to the Problems... 2 2.1 Smoothng Methods... 2 2.2 Smoothng Splnes and Regresson
More informationMobileGrid: Capacity-aware Topology Control in Mobile Ad Hoc Networks
MobleGrd: Capacty-aware Topology Control n Moble Ad Hoc Networks Jle Lu, Baochun L Department of Electrcal and Computer Engneerng Unversty of Toronto {jenne,bl}@eecg.toronto.edu Abstract Snce wreless moble
More informationHalmstad University Post-Print
Halmstad Unversty Post-Prnt Admsson Control for Swtched Realtme Ethernet Schedulng Analyss versus etwor Calculus Xng Fan and Magnus Jonsson.B.: When ctng ths wor cte the orgnal artcle. Orgnal Publcaton:
More informationAchievable Bandwidth Estimation for Stations in Multi-Rate IEEE WLAN Cells
Achevable Bandwdth Estmaton for Statons n Mult-Rate IEEE 802. WLAN Cells Eduard Garca, Davd Vamonte, Rafael Vdal and Josep Paradells Wreless Networks Group - echncal Unversty of Catalona (UPC) {eduardg,
More informationIP Camera Configuration Software Instruction Manual
IP Camera 9483 - Confguraton Software Instructon Manual VBD 612-4 (10.14) Dear Customer, Wth your purchase of ths IP Camera, you have chosen a qualty product manufactured by RADEMACHER. Thank you for the
More informationFAHP and Modified GRA Based Network Selection in Heterogeneous Wireless Networks
2017 2nd Internatonal Semnar on Appled Physcs, Optoelectroncs and Photoncs (APOP 2017) ISBN: 978-1-60595-522-3 FAHP and Modfed GRA Based Network Selecton n Heterogeneous Wreless Networks Xaohan DU, Zhqng
More informationA New Feedback Control Mechanism for Error Correction in Packet-Switched Networks
Proceedngs of the th IEEE Conference on Decson and Control, and the European Control Conference 005 Sevlle, Span, December -15, 005 MoA1. A New Control Mechansm for Error Correcton n Packet-Swtched Networks
More informationResearch of Dynamic Access to Cloud Database Based on Improved Pheromone Algorithm
, pp.197-202 http://dx.do.org/10.14257/dta.2016.9.5.20 Research of Dynamc Access to Cloud Database Based on Improved Pheromone Algorthm Yongqang L 1 and Jn Pan 2 1 (Software Technology Vocatonal College,
More informationPrivate Information Retrieval (PIR)
2 Levente Buttyán Problem formulaton Alce wants to obtan nformaton from a database, but she does not want the database to learn whch nformaton she wanted e.g., Alce s an nvestor queryng a stock-market
More informationSteps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices
Steps for Computng the Dssmlarty, Entropy, Herfndahl-Hrschman and Accessblty (Gravty wth Competton) Indces I. Dssmlarty Index Measurement: The followng formula can be used to measure the evenness between
More informationCHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION
24 CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION The present chapter proposes an IPSO approach for multprocessor task schedulng problem wth two classfcatons, namely, statc ndependent tasks and
More information