Efficient Translation of Network Performance Parameters for Transport of IP Packets over Cell-Switched Subnetworks

Size: px
Start display at page:

Download "Efficient Translation of Network Performance Parameters for Transport of IP Packets over Cell-Switched Subnetworks"

Transcription

1 Efficient Tanlation of Netwok Pefomance Paamete fo Tanpot of IP Packet ove Cell-Switched Subnetwok Jen Schmitt 1, Matin Katen 1, and Ralf Steinmetz 1,2 1 Indutial Poce and Sytem Communication, Damtadt Univeity of Technology, Gemany 2 Geman National Reeach Cente fo Infomation Technology, GMD IPSI, Damtadt, Gemany {Jen.Schmitt,Matin.Katen,Ralf.Steinmetz}@KOM.tu-damtadt.de Abtact -- In thi pape we deal with mapping pefomance-oiented IP evice uch a Integated Sevice (IntSev) o Diffeentiated Sevice (DiffSev) onto netwok pefomance paamete fo cell-witched tanpot netwok, a fo example ATM. The impact of tanlating IP pefomance paamete into ATM netwok evice i analyzed and the detimental effect of caele mapping ae illutated. Then, appoache to cicumvent thee detimental effect ae peented. Keywod: Netwok QoS, IntSev, DiffSev, ATM, AAL. A. Motivation I. INTRODUCTION Both the Intenet tandadization oganization, IETF, and the telecommunication tandadization committee have developed Quality of Sevice (QoS) achitectue. While the telecommunication people took a athe evolutionay tep with the Aynchonou Tanfe Mode (ATM) a the candidate fo the next-geneation integated evice netwok, the Intenet community tie to follow an evolutionay path by integating QoSenabling component into the exiting IP technology. Stemg fom two vey diffeent, though today omewhat conveging eeach and tandadization communitie thi led to vey diffeent QoS achitectue. The gand plan of the telecommunication community with a global ATM netwok at the heat of a homogeneou integated evice netwok nowaday eem to fade away. Yet, while not being ued a end-to-end olution the gowth of ATM netwok in the backbone of lage-cale intenetwok, a, e.g. the Intenet, i a eality. In geneal, it i ageed that heteogeneity i a fact fo today lage-cale intenetwok, with the global Intenet a the mot poent example. Theefoe it i alo a fact fo QoS achitectue. Anyway, competition, diffeent tength and evolution ae agument fo heteogeneity with egad to QoS a well. QoS achitectue can be viewed a a combination of QoS pocedue, a, e.g. ignalling potocol, and QoS model, which captue the declaative pat of the achitectue, a, e.g. the available evice clae. We will concentate on the mapping of QoS model between IP and ATM netwok, and hee in paticula on the tanlation of the netwok pefomance paamete. With egad to the pocedual apect of mapping, ee fo example [1], o [2], yet thee ae many moe. B. Outline Fit we give a bief oveview of the diffeent QoS model of IP and ATM, then ome geneal thought on the mapping between the IP and ATM QoS model ae peented, befoe we go into the detail of one geneic poblem of any mapping: the tanlation of the pefomance paamete fom packet o byte unit into cell unit. We identify two majo poblem of a taightfowad tanlation and then peent appoache to olve o at leat alleviate each of them. At the end, we take a look at elated wok and daw ome concluion fom ou invetigation. II. QUALITY OF SERVICE MODELS A. ATM (Foum) Model The ATM evice model i baed on the taditional call paadigm with the following evice clae a emantic intepetation famewok fo the netwok pefomance paamete of an ATM netwok [3]: Contant Bit Rate(CBR): offe a contant bit ate evice uited fo eal-time application with tingent equiement on delay and bandwidth. eal-time Vaiable Bit Rate(t-VBR): offe a imila evice to CBR but allow fo ome contolled butine of the data team. non-eal-time Vaiable Bit Rate (nt-vbr): a non-ealtime evice with a deteitic bound on lo a long a taffic adhee to it pecified hape. Unpecified Bit Rate(UBR): plain bet-effot evice without any guaantee. Available Bit Rate (ABR): feedback-baed evice that allow fo a imum ate to be pecified and enue fai haing within thi cla of taffic. Guaanteed Fame Rate(GFR): a fame-awae evice that allow fo a imum ate to be pecified, and, e.g. take AAL5 fame boundaie into account when making cell dicad o tagging deciion. Mot of thee equie a taffic pecification which i baed on the Geneic Cell Rate Algoithm (GCRA). The unit of the paamete ae cell ep. cell/, even fo the GFR evice. Fo the exact definition of the paamete and thei applicability to the evice categoie, ee [3].

2 B. IETF Model Much wok inide the IETF ha been devoted to the development of QoS model fo the Intenet. The outcome ae two diffeent model fo achieving QoS: IntSev and DiffSev. Thee howeve deal with diffeent need and can alo be een a complementay and mutually aiting [4], and not neceaily competing. 1) IntSev Thi model i moe in the tadition of telecommunication buine model, whee an end-to-end evice i offeed to the cutome at the end-ytem. Theefoe, the evice offeed ae pecified at the flow-level, i.e. vey fine-gained. Two evice have advanced to popoed tandad: Guaanteed Sevice (GS) [5] and Contolled Load Sevice (CLS) [6]. GS offe deteitic guaantee on the imum endto-end delay and the available bandwith a well a a zeo lo auance. It equie a taffic pecification, called TSpec, which i eentially a double token bucket, with a the token ate of the fit bucket and b a it bucket depth, and fo the econd bucket the peak ate p and the imum packet ize M a the bucket depth. The evice ate R a pecified by the eceive() detee the expeienced queuing delay and thu eve a contol paamete to adjut the imum delay toleable fo a GS ue. CLS ha a much looe pecification which i uppoed to offe a evice that i compaable to bet-effot evice in a lightly loaded netwok. It alo equie the pecification of a TSpec and enue that unde any load condition of the netwok a CLS ue will at leat have a thoughput of the token ate. Fo both evice the unit in which paamete ae pecified ae byte ep. byte/. 2) DiffSev Thi model [7] i a moe pagmatic/le ambitiou appoach motivated by the eality of today Intenet evice povide (ISP), which would like to offe highe value evice to thei cutome, who ae end-ue a well a othe ISP. Hence, the evice offeed will be baed on taffic aggegate and will thu be athe coae-gained. The appoach taken fo DiffSev i not to pecify the evice - thee hall be pat of bilateal Sevice Level Ageement (SLA) between povide o cutome - but to pecify the behaviou of the fowading element in o called Pe- Hop Behaviou (PHB). Two PHB have been advanced to popoed tandad: Expedited Fowading (EF) [8] Aued Fowading (AF) [9] Both of them equie the configuation of a cetain evice ate to atify thei pecified behaviou. Thi ate will be given in bit/ o byte/, which ae of coue equivalent (fo ou pupoe). C. Selecting the ATM Sevice Categoie An impotant deciion fo the mapping of the QoS model i the aignment of IP-elated evice to ATM evice categoie. Although thi i not the focu of ou pape, we biefly want to dicu the iue fo that fundamental election hee, fo both IntSev and DiffSev. 1) IntSev A mapping of the IntSev clae, GS and CLS, onto ATM evice categoie hould ty to peeve thei epective emantic, while at the ame time tying to imize the eouce uage inide the ATM ubnetwok. Fo GS, the mot taightfowad candidate i t-vbr, although CBR i anothe, but peumably moe cotly altenative. The othe evice categoie do not eem to be uited ince they ae fo non-ealtime application. Fo CLS, eithe nt-vbr, ABR, o GFR ae good candidate, wheea CBR o t-vbr ae pincipally poible, but too cotly a they offe moe than i needed to atify the CLS pecification. 2) DiffSev The fit quetion to be anweed fo DiffSev i whethe ATM VC coepond to SLA o PHB. While the ATM Foum take the fome poition, the IETF favou the latte. Fo EF-baed SLA o fo the EF PHB the hottet candidate ae CBR and t-vbr, wheea fo AF, ABR o GFR eem the mot eaonable choice. The above conideation ae led by technical and economical ationale and imila, yet much moe detailed teatment of thi can be found in ([10] fo IntSev, and [11] fo Diff- Sev). It i ou futue goal and ou tong belief that it i neceay to veify and poibly modify thee by tial and meauement fo eal taffic. Howeve, in thi pape we want to concentate on a common iue of any mapping. That i the conveion of the paamete fom byte to cell, which eult fom the fundamentally diffeent chaacteitic of vaiable v. fixed tanpot unit ize. No matte whethe it i IntSev, DiffSev o any othe IP-pefomance-oiented evice, they all have to deal with thi iue of tanlating the packet-baed natue of IP-pefomance elated metic into ATM cell-baed countepat - uppoing an ATM ubnetwok i coed. III. TRANSLATING THE PERFORMANCE PARAMETERS A. Staightfowad Tanlation Conide a flow of packet, fo which an IP netwok evice pefomance commitment exit, with each packet in iolation and aume that no moe than one packet fit into a ingle cell (often moe cell ae equied). Note hee that an IP heade aleady conume 20 byte, and a UDP heade anothe 8 byte o that fo example an application uing UDP/IP neve poduce packet of which moe than one would fit into a ingle ATM cell, epecially if poible futhe AAL-elated encapulation ovehead i taken into account.

3 Let u look at that in a moe fomal and geneal way. Fit we define ome tem: Cell Ovehead: o c [in byte]. Packet Ovehead: o p [in byte]. Packet ize: p [in byte] with p [ mm, ], i.e. m i the imum and M the imum packet ize fo the flow. Cell ize: c [in byte]. Numbe of cell pe packet: n c [in cell/packet], whee n p + o c p. c o c Since p may vay (while the othe paamete ae fixed, at leat pe flow), the numbe of cell pe packet may alo be egaded a a function of the packet ize: n c ( p ). Given that p [ mm, ], the following bound on the numbe of cell pe packet ae given: m+ o n c n c ( m) p M + o n p c n c ( M) n c c o c c o c. Given a cetain IP pefomance-elated ate [in byte/], we get a packet ate p with p ---- [in packet/], p which again allow to compute the equied cell ate c with c p n c [in cell/]. Again the only vaiable paamete i p and we theefoe ealize that the cell ate i a function of the packet ize(), c ( p ), a well a the packet ate, p ( p ). Both n c and p vay with p. While p weakly deceae with p, n c weakly inceae with p. Noticing that c i a weakly deceaing function in p, i.e., c how ome pontaneou hot-cale inceae due to wellfitting packet ize, but how long-cale deceae due to the haing of packet ovehead, we obtain the following bound on c : c p + o p p c o c ( )n p p c ( p ) c p ( p )n c ( p ) p + o p p c o c c whee p ag c ( p ) p [ mm, ] ag c ( p ) p { m, m + o p c o c ( c o c ) o p + 1} p ag c ( p ) p [ mm, ] ag c ( p ) p { M, M + o p c o c ( c o c ) o p } Of coue fo ATM c 48 (1), and fo diffeent AAL the eulting numbe and fomulae ae given in Table 1, whee it i aumed that LLC/SNAP encapulation a defined in [12] i ued in all cae. If intead of that VC-baed multiplexing wa ued then all o p value would need to be dihed by 8. TABLE 1 Application of the Mathematical Famewok. AAL Type o c o p n c c AAL 1 AAL 2 AAL 3/4 AAL Thi table i a lightly aguable ince fo AAL1 and AAL2 thee ae no tandad o popoal a to how to encapulate IP packet. To ae how much the choice of the packet ize affect the cell ate that i to be allocated, take a look at the cell ate fo diffeent packet ize a depicted in Figue 1. Hee we aumed the ue of AAL5 and LLC/SNAP encapulation and an IP pefomance-elated ate of byte/. Cell Rate ( c ) Depending on the packet ize we have to allocate cell ate diffeing by a facto of almot thee. Futhemoe, we notice that even fo packet ize cloely togethe the diffeence in thei coeponding cell ate may be huge. Let u look at that moe igoouly. B. Pefomance Analyi p p p p p p p p p + 16 p p + 16 p 48 Packet Size ( p ) Figue 1: Cell Rate fo diffeent packet ize. In thi ection we fit define and motivate ome metic, which then eve a citeia fo dicuing diffeent cheme fo tanlation of the packet-baed pefomance paamete into thei cell-baed countepat.

4 1) Metic Let u fit define a metic called Cell Utilization Efficiency (CUE) a follow: CUE [ , ] [,] 0 1 (2) c c c c c c The CUE i a meaue of how well utilized allocated eouce of the cell-witched netwok ae if the expected packet ize matche the actual packet ize. It may howeve be the cae that the expected packet ize when the allocation i made i not the packet ize actually een in the data flow. Theefoe let u define a futhe metic to meaue the cell utilization efficiency fo thi cae. Aume c i choen a cell ate baed on an expected packet ize p, yet p tun out to be the actual packet ize. Then let u define the ealized CUE (CUE) a function of p : CUE( p ) p < p c c c c p p c c c Cetainly, the wot cae with egad to efficiency i that the actual packet ize i the packet ize that imize the cell ate, i.e. p. We captue thi cae in a metic called wot-cae CUE (wcue), which i defined a: c wcue CUE( M) c (4) c c c In any cae that mean that it i favouable to bae the cell ate on a lage a poible packet ize. But cell utilization i jut one ide of the toy, the othe i how badly we may oveload the cell ate allocation by ovely optimitic packet ize expectation. That i captued in the following metic. The Cell Lo Rate (CLR) i defined a a function of p : CLR( p ) c p < p c 0 p p Of coue the highet ate of cell loe i incued if the actual packet ize imize the cell ate, i.e. it i p. Thu we define the wot-cae Cell Lo Rate (wclr) a: wclr CLR( m) c (6) c The wclr meaue how badly oveloaded the cellwitched netwok may be due to an undeized cell ate allocation a the eult of oveetimating packet ize. 2) Dicuion Let u now take a look at how the taightfowad tanlation of IP pefomance paamete onto cell-witched netwok paamete behave with egad to the intoduced metic. (3) (5) In Figue 2 the wcue i depicted, again fo the cae whee AAL5 with LLC/SNAP encapulation i ued and the IPelated ate i byte/. wcue Packet Size ( p ) Figue 2: Wot-Cae Cell Utilization Efficiency. Thee ae two baic and othogonal poblem that lead to inefficient ue of cell-ate eouce which ae illutated in the above gaph: 1. Ove-eevation due to the uncetainty about packet ize, and theefoe about the numbe of packet pe unit of time, ince thi influence the ovehead haing of fag packet fo tanpot ove the cell-witched netwok. The weakening of thi effect a the imum packet ize i appoached i epeented by the long-tem inceae of the wcue cuve. 2. Ove-eevation due to unued capacity in patially filled cell eulting fom inconvenient packet ize. Thi effect i epeented by the pontaneou hot-tem deceae of the wcue cuve, wheneve a cell bounday i exceeded by the packet ize on which the cell ate allocation i baed. Obviouly, fo efficiency eaon it would be advantageou to aume lage packet ize and to caefully chooe the packet ize (on one of the peak if poible). Yet, in Figue 3 the wclr i depicted fo diffeent packet ize. wcue Packet Size ( p ) Figue 3: Wot-Cae Cell Lo Rate. Of coue, the wclr ie a the packet ize inceae, on which the cell ate allocation ae baed. Futhemoe, the

5 packet ize that wee convenient with epect to wcue ae vey bad fo the wclr a they coepond to pontaneou peak of it. Obviouly, the wcue and wclr ae competing metic becaue when tying to impove the cell utilization efficiency by loweing the cell ate, the ik i to incu a highe cell lo ate. Theefoe a compomie fo the aumed packet ize of the IP data team mut be found accoding to it evice emantic. A tict evice a, e.g. IntSev GS will not toleate any cell lo, o that p mut be aumed a packet ize fo the calculation of the cell ate coeponding to the evice ate R. Fo evice that do not povide fo uch tict guaantee a tadeoff between the ik of incuing cell lo and an impoved efficiency i poible. All of the above aume that the packet ize i not a contolled vaiable. Of coue, one may ague that application could geneate IP packet of well-uited ize that fit exactly into an integal numbe of cell and ae a lage a poible. Yet, in geneal thi eem to be not feaible o at leat not convenient due to the following poblem: application hould not need to know about a (poibly fa away ) cell-witched ubnetwok, ATM i jut one link, othe link might have diffeent need with egad to packet ize, application would need link laye knowledge, which contitute a go layeing violation. Conequently, edge device, mediating between IP and ATM, have to cope with uncetainty about packet ize and with unluckily ized packet that do not fit the cell team well. While olution appoache to the fome poblem will be dealt with in Section V, we will at fit adde the latte poblem by a cheme we called cell-aligned fag. IP Packet AAL Fame ATM Cell Figue 4: Cell-Aligned Fag. if the wot cae of a team ending but at ized packet i actually occuing, becaue on thi cae the ate calculation have to be baed (at leat fo had guaantee a, e.g. fo IntSev GS). At thi tage, one may ague that imum packet ize may be lage enough to make the ovehead incued by patially filled cell negligible, yet that i not the cae fo many eal-time application whee packetization delay till play a cetain ole, and futhemoe not fo IP taffic aggegate a they have to be dealt with when uing DiffSev. Hee packet ize may vay highly (alo to the lowe end) and may be not known befoehand o that mall packet ize mut be aumed to be on the afe ide. To give a feeling fo cuent IP taffic packet ize ditibution, ee Figue 5, which wa poduced by [CMT98] fom a 24 hou taffic tace at an OC3 link of the MCI netwok backbone. p Taile IV. EFFICIENT TRANSLATION BASED ON CELL-ALIGNED FRAMING A. Idea The taightfowad ate tanlation cheme peented and analyed in the peviou ection egaded each packet of an IP data team in iolation and encapulated it into a epaate AAL fame. That lead to the poblem of patially filled cell that have to be padded with byte containing no infomation. The idea of cell-aligned fag i to fill AAL fame uch that they fit exactly into the cell team iepective of the packet boundaie. Theefoe a ingle AAL fame may contain two (patial) packet. Howeve only the lat cell of a fame hould contain data fom both packet: the end of the fit packet and the beginning of the next packet. Thi cheme i illutated in Figue 4. Thi cheme equie that thee i a way to mak the tat of a new packet inide an AAL fame, which may eult in ome additional potocol ovehead, which howeve a we will ee in Section IV.D hould not be inhibitive. Futhemoe, note hee that it i not neceaily equied to cicumvent padded cell but to ue cell-alignment only in cae it i neceay, i.e., Figue 5: Typical Packet Size Ditibution. Thi clealy how that mall packet ize ae till pedoant at leat fo today IP taffic at the aggegate level. One hould howeve be awae that new evice a intoduced by IntSev and DiffSev will cetainly change taffic chaacteitic, a, e.g. the packet ize ditibution. B. Analyi and Compaion Uing the notation and definition of Section III let u analye the appoach of cell-aligned fag and compae it with the taighfowad appoach: Ovehead fo cell-alignment: o align. In thi cae the cell ate coeponding to a byte ate i:

6 c ( p ) ---- p + o p + o align --- p c o c whee we have the following bound on c M + o c ---- p + o align ---- M c c o c (7) In Figue 6 the wcue fo the cae of a taightfowad tanlation and the appoach baed on cell-aligned fag ae compaed. wcue --- m m o p + o align c o c We ued the ame etting a in the example befoe and aumed no ovehead fo the cell-alignment, which a will be hown in Section IV.D i poible fo AAL5. It i obviou that cell-aligned fag can achieve quite a ubtantial efficiency gain, epecially fo vey mall packet ize. Let u now take a look at the wclr fo both cae a it i depicted in Figue 7. wclr c Again it can be een that cell-aligned fag i a conideable impovement ove the taightfowad appoach whee packet ae teated in iolation. Thi i due to the fact that the pace of poible cell ate, i.e. [ c, c ], i conideably compeed and thu the ik of aug lage packet ize fo the cell ate allocation tanlate into much lowe cell lo Packet Size ( p ) Figue 6: Wot-Cae Cell Utilization Efficiency Staighfowad Fag Cell-Aligned Fag Staighfowad Fag Cell-Aligned Fag Packet Size ( p ) Figue 7: Wot-Cae Cell Lo Rate. (8) ate if the actual packet ize i le. C. Potential Dawback Afte having hown the benefit of cell-aligned fag ove the taightfowad ate tanlation, let u now look at ome potential counte-agument that may be aied againt it: One quetion cetainly i how expenive the egeneation of packet boundaie i. A mentioned above, a making technique i needed, which may conume ome PCI (Potocol Contol Infomation), and we have ome moe computational effot in ode to keep tack of the fagmented packet. We will ee below that thi ovehead can be kept eaonably mall. When uing cell-aligned fag not all the cell ae equally impotant any moe, becaue one lot cell may kill two packet, if it i the haed cell of two conecutive packet. Howeve, it can be agued that eithe the packet ae mall and then thee i not o much lot o they ae lage and then thi hould be an infequent event. Fame may have to wait to be filled up. Yet, hee the olution i to neve wait fo following packet to fill up the cell team, but only fill it up if thee ae aleady packet waiting in the queue. The ationale hee i that the ate computation ae baed on cetain wot-cae cenaio in which the appoach would actually be applied, wheea if the ate i not fully ued then the watage of cell pace i not uch a big iue. The main point i that the ate tanlation which ae baed on thi wot-cae cenaio can be kept low. D. Implementation Uing AAL5 Afte having hown the benefit and potential dawback of cell-aligned fag, we now peent a vey imple way of how the cheme could be implemented when AAL5 i ued a adaptation laye fo the tanpot of IP taffic ove an ATM ubnetwok. In the ATM teology thi could alo be called a SSCS (Sevice-Specific Convegence Sublaye) of AAL5 fo IP-pefomance oiented evice uch a IntSev o DiffSev. The tak of that SSCS i to mak whee a new packet tat within an AAL5 fame in ode to be able to eaemble packet at the eceiving ide. The AAL5 CPCS-PDU (Common Pat Convegence Sublaye) i tuctued a depicted in Figue 8. Payload Taile Length CRC Fotunately, it poee an unued field called UU (Ueto-Ue Indication). The idea i now to ue that field a a CPI CPCS-UU(1octet) Figue 8: CPCS-PDU fomat fo AAL5.

7 pointe to the beginning of the next IP packet in an AAL5 fame. Thu, the emantic of the UU field i the numbe of byte fom the end of an AAL5 fame to the location whee a new IP packet tat. Thi can of coue be at mot 255 byte apat, yet it i ufficient if only the lat cell i alway with filled with the beginning of the next packet, a wa popoed above. Note that UU0 mean that the encapulated IP packet plu ovehead fitted exactly an integal numbe of ATM cell. In Figue 9 the equied potocol poceing fo cellaligned fag i illutated in peudocode fo both, ending and eceiving ide. At the ending ide, it ha to be computed whethe padding of the payload i neceay and if o, how many byte of padding. If anothe packet i aleady waiting, then intead of padding the AAL5 fame, it i filled up with the fit byte of the waiting packet and the UU pointe i et to the beginning of that packet. At the eceiving ide, the packet ae poibly eaembled by uing the infomation deliveed in the UU field of the AAL fame. Uing thee algoithm eult in no PCI ovehead fo cellaligned fag, i.e. o team 0, but intoduce a highe potocol poceing cot due to the moe complicated buffe management, which howeve fom ou pepective hould be jutified due to the conideable efficiency impovement peented above. V. SOLUTION APPROACHES TO THE UNKNOWN NUMBER OF PACKETS PROBLEM While cell-aligned fag avoid the egmentation ovehead due to patially filled cell, a olution to the poblem of the vaiability of packet ize would ave ovehead that i accounted pe packet, i.e. o p. Thi ovehead i popotional to o p / p, and can of coue not be totally cicumvented but loweed by uing ome (heuitic) knowledge about the packet ize ditibution. Thi could be baed upon tatitic o pat expeience in geneal which might be available. The appoach i mainly aimed at evice that only povide fo oft guaantee, a fo example IntSev CLS o DiffSev AF. The idea i to be able to make a quantitative tatement Sende-Algoithm foeve { wait fo packet to be ent; compute #byte equied fo padding; if (padding! 0 && anothe packet waiting { take padding byte fom waiting packet; fill it togethe with packet in AAL5 fame F; et F.cpc-uu padding; } ele { end padded AAL5 fame; et F.cpc-uu 0; } } about cetain metic given a cetain packet ize ditibution. A an example, it hould be poible to povide an auance like: if packet ize ae unifomly ditibuted ove [m,m], then at a pobability of 95% we obtain a CLR of 0. Let u look at that in a moe fomal manne. Recall that p i a andom vaiable which mut be etimated well in ode to be able to make ate allocation with favouable cell utilization and toleable lo chaacteitic. Poent example cae ae: 1. p i unifomly ditibuted ove [m, M], i.e. it p.d.f. i f ( p ) M m p i tapezoidally ditibuted ove [m, M] (with the lope a of the tapezoid epeenting the optimim/peimim of the aumption on the packet ize), i.e. it p.d.f. i: f a ( p ) a p a M + m M m (10) 2 2 a [ , ] ( M m) 2 ( M m) 2 At fit we define quantilized cell ate c,α a p( CLR 0 c, α ) > 1 α (11) which mean the pobability to incu cell lo if we allocate c,α i le than α. Let u look at the geneal cae, whee we aume that p ha the ditibution function F( p ). Yet, intead of the packet ize ditibution we intoduce a tanfom of it, the packet ate ditibution, whee the packet ate i defined a: p ---- (12) p Fom thi the quantilized cell ate can be computed moe eaily (if cell-aligned fag i aumed), ince the cell ate fo the cae of uing cell-alignment can be ewitten a: + p ( o p + o align ) c Receive-Algoithm (9). (13) NetwokBuffe outtanding_packet empty; foeve { eceive AAL5 fame F; append F.payload[0, F.length-F.cpc_uu-1] to outtanding_packet; end outtanding_packet to uppe laye; buffe F.payload[F.length-F.cpc_uu, F.length] in outtanding_packet; } Figue 9: Cell-Aligned Fag Algoithm at Sende and Receive. c o c

8 Since the packet ate ha the mioed ditibution of the packet ize (ince p i a homomophim of p ), aumption about packet ize tanlate eadily in the ditibution of the packet ate. To calculate thoe quantilized cell ate, note that 1 α < p( CLR 0 c, α ) p ( c < c, α ) + p p ( o p + o align ) p, α ( o p + o align ) < c o c p + p( o p + o align ) + p, α ( o p + o align ) < < c o c c o c (14) p p c o < p, α c o p + o align G c o p, α c o p + o align 1 F c o p, α c o p + o align Hee p,α i the packet ate coeponding to c,α and G i the ditibution of p. Due to the integality containt on cell ate it i not poible to calculate them exactly fo evey α, but only a (tight) lowe bound can be computed, which give a cell ate at which the CLR 0 with a pobability of at leat 1-α (aug a cetain packet ize ditibution and theefoe packet ate ditibution). To compute the cell ate fom (14), note that fom (13) it follow that c, α ( c o c ) p, α (15) o p + o align Uing that elation and afte ome algeba we obtain the elation: o c, α p + o > (16) c o align + 1 c F 1 ( α) which allow u to compute the quantilized cell ate a c, α (17) c o 1 o p + o align + 1 c F 1 ( α) In Table 2, ae given ome example value of quantilized cell ate fo the ample packet ize ditibution (9) and (10). We ued the ame paamete etting a fo the example in peceding ection (in paticula we ued m33 and M500), and fo the paamete a of the tapezoidal ditibution we ued the exteme value ± 2 ( M m) 2, which epeent vey optimitic epectively peimitic aumption on the packet ize ditibution. TABLE 2 Quantilized Cell Rate. c,α α0.01 α0.05 α0.1 α0.2 unifom c o c optimitic tapezoidal peimitic tapezoidal TABLE 2 Quantilized Cell Rate. c,α α0.01 α0.05 α0.1 α Altenatively and imilaly, the CUE could be taken a a metic to define quantilized cell ate, o the CLR could be choen le o equal to ome β>0. Yet, one mut be awae that the latte would intoduce anothe paamete that might be difficult to pecify - paimoniou model ae geneally pefeable. VI. RELATED WORK The iue of ovelaying IP QoS evice onto ATM ubnetwok, ha been and till i dealt with extenively, fo an oveview of that lage field of elated wok, ee [2]. Yet, diectly elated to the iue of mapping the QoS model, the mot impotant wok ha cetainly been done in the IETF and ATM Foum. Hee, fo IntSev it exit a popoed IETF tandad [10], that give vey detailed teatment on how to chooe the ATM evice categoie fo the GS and CLS clae. Similaly, thee i wok in poge [11] in the IETF and ATM Foum [13] on the mapping of PHB ep. SLA to ATM evice categoie. Non-tandadization wok concened with thoe iue can be found in [14], whee it i hown that the IntSev to ATM mapping popoed in [10] ae at leat dubiou, a they ae hown to lead to exceive cell lo in imulation. The autho of [15] ae epecially concened with how to map CLS to ATM and give ome imulation eult on thei pecific mapping cheme. Howeve, all of thee do not conide the tanlation of the diffeent paamete unit in the detailed and igoou manne we did in thi pape. Futhemoe, they etict thei invetigation towad a cetain IP QoS model o even only pat of it, wheea ou wok i geneally applicable to pefomanceoiented IP netwok evice, of which IntSev and DiffSev ae jut example. Futhemoe, mot of ou eult ae alo geneic fo abitay cell-witched netwok and not jut fo ATM. So, we ee the majo contibution of ou wok in the geneality of the eult on how to tanlate efficiently between IP and cell-witching netwok pefomance paamete. VII. CONCLUSIONS Afte thooughly analying peviouly popoed taighfowad appoache we identified the two main obtacle to an efficient tanlation of IP to cell-witching pefomance paamete a egmentation ovehead and vaiability of packet ize. We intoduced and analyed the appoach of cell-aligned fag in ode to olve the iue of only pa-

9 tially filled cell due to the egmentation ovehead. Futhemoe, we peented a imple and efficient way to implement that appoach fo the cae of AAL5 fag of IP packet. Baed on cell-aligned fag we popoed a cheme to adde the poblem of vaiable packet ize and theefoe unknown ovehead accounted pe packet. The cheme i baed on aumption on packet ize and allow fo nondeteitic evice guaantee to tadeoff eouce allocation efficiency againt cell lo pobabilitie. VIII. FUTURE WORK One obviou iue of futue wok i the actual implementation of the cheme peented and analyed above in ode to veify expeimentally the efficiency gain of cell-aligned fag and quantilized cell ate. Since we have implemented a vey flexible IP/ATM adaptation module fo ue at an edge device (decibed in [16]), which ue taightfowad tanlation of the paamete, it hould be eay to extend it to allow fo the moe ophiticated technique popoed in thi pape. Futhe wok item fo the futue could alo be to find new definition fo quantilized cell ate and to bae them on othe packet ize ditibution, which might be deived by obevation fo eal taffic. Yet, note hee that cuent IP taffic tace ae only of limited value to pedict tatitic fo IP-pefomance oiented taffic baed on IntSev and Diffev, ince thoe ae not commonly ued in poduction netwok at the time of witing. Aued Fowading PHB Goup, June RFC [10] M. Gaett and M. Boden. Inteopeation of Contolled Load and Guaanteed Sevice with ATM, Augut RFC [11] S. Ayandeh, A. Kihnamuthy, and A. Mali. Mapping to ATM Clae of Sevice fo Diffeentiated Sevice Achitectue, Novembe Intenet Daft, wok in poge. [12] J. Heinanen. Multipotocol Encapulation ove ATM Adaptation Laye 5, July RFC [13] ATM Foum Technical Commitee: Addendum to Taffic Management (TM) Specification Enhancement to Suppot IP Diffeentiated Sevice (Daft), Decembe wok in poge. [14] P. Fanci-Cobley and N. Davie. Pefomance Implication of QoS Mapping in Heteogeneou Netwok Involving ATM. In Poc. of IEEE Confeence on ATM 98 (ICATM 98). IEEE, June [15] P. Giacomazzi and L. Muumeci. Tanpot of IP Contolled-Load Sevice ove ATM Netwok. IEEE Netwok, 13(1), Januay [16] J. Schmitt. A Flexible, QoS-Awae IP/ATM Adaptation Module. Technical Repot TR-KOM , Damtadt Univeity of Technology, Decembe IX. REFERENCES [1] L. Bege, E. Cawley, S. Beon, F. Bake, M. Boden, and J. Kawczyk. A Famewok fo Integated Sevice with RSVP ove ATM, Augut RFC [2] J. Schmitt and J. Antich. Iue in Ovelaying RSVP and IP Multicat on ATM Netwok. Technical Repot TR- KOM , Univeity of Technology Damtadt, Augut [3] ATM Foum Technical Commitee: Taffic Management (TM) Specification 4.1, Mach [4] Y. Benet, R. Yavatka, P. Fod, F. Bake, L. Zhang, M. Spee, R. Baden, B. Davie, J. Woclawki, and E. Feltaine. A Famewok fo Ue of RSVP with Diff- Sev Netwok, Septembe Intenet Daft, wok in poge. [5] S. Shenke, C. Patidge, and R. Guein. Specification of Guaanteed Quality of Sevice, Septembe RFC [6] J. Woczlawki. Specification of the Contolled-Load Netwok Element Sevice, Septembe RFC [7] D. Black, S. Blake, M. Calon, E. Davie, Z. Wang, and W. Wei. An Achitectue fo Diffeentiated Sevice, Decembe RFC [8] V. Jacobon, K. Nichol, and K. Podui. An Expedited Fowading PHB, June RFC [9] J. Heinanen, F. Bake, W. Wei, and J. Woclawki.

IEEE C802.16m-07/060. IEEE Broadband Wireless Access Working Group <http://ieee802.org/16>

IEEE C802.16m-07/060. IEEE Broadband Wireless Access Working Group <http://ieee802.org/16> 007-3-5 IEEE C80.6m-07/060 Poject Title Date Submitted IEEE 80.6 Boadband Wiele Acce Woking Goup Coelation Model fo Shadow Fading Simulation 007-3-5 Souce() I-Kang Fu, Chi-Fang Li,

More information

Dynamic Virtual Network Embedding based on Auction-Game Theory NingNing Liang 1, JuLong Lan 1, YunFu Long 2, GuanYing Liu 2

Dynamic Virtual Network Embedding based on Auction-Game Theory NingNing Liang 1, JuLong Lan 1, YunFu Long 2, GuanYing Liu 2 Intenational Confeence on Advance in Mechanical Engineeing and Indutial Infomatic (AMEII 2015) Dynamic Vitual etwok Embedding baed on Auction-Game Theoy inging Liang 1, JuLong Lan 1, YunFu Long 2, GuanYing

More information

A Fast Texture Feature Extraction Method for Region-based Image Segmentation

A Fast Texture Feature Extraction Method for Region-based Image Segmentation A Fat extue Featue Extaction Method fo Region-baed Image Segmentation Hui Zhang *, Jaon E. Fitt, Sally A. Goldman Dept. of Compute Science and Engineeing, Wahington Univeity, One Booking Dive. St. Loui,

More information

Outage Probability in Arbitrarily-Shaped Finite Wireless Networks

Outage Probability in Arbitrarily-Shaped Finite Wireless Networks 1 Outage Pobability in Abitaily-Shaped Finite Wiele Netwok Jing Guo, Student Membe, IEEE, Salman Duani, Senio Membe, IEEE, and Xiangyun Zhou, Membe, IEEE axiv:134.6172v4 [c.it] 2 Dec 213 Abtact Thi pape

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

Improved Graph-Based Image Segmentation Based on Mean Shift 1

Improved Graph-Based Image Segmentation Based on Mean Shift 1 03 8th Intenational Confeence on Communication and Netwoking in China (CHINACOM) Impoved Gaph-Baed Image Segmentation Baed on Mean Shift Jianwen Mo, Chaoxuan Wang,Tong Zhang, Hua uan School of Infomation

More information

Source-independent Hessian-free Gauss-Newton Full-waveform Inversion

Source-independent Hessian-free Gauss-Newton Full-waveform Inversion Souce-independent Heian-fee Gau-Newton Full-wavefom Inveion Wenyong Pan and Ki Innanen CREWES Poject Depatment of Geocience Univeity of Calgay Summay The etimation of ouce wavelet i vey impotant fo implementing

More information

Computer Vision - Histogram Processing. Dr. S. Das IIT Madras, CHENNAI - 36

Computer Vision - Histogram Processing. Dr. S. Das IIT Madras, CHENNAI - 36 Comute Viion - Hitogam Poceing D. S. Da IIT Mada, CHENNAI - 36 HISTOGRAM In a gay level image the obabilitie aigned to each gay level can be given by the elation: n, N Inut image,,2...l - - The nomalized

More information

A Design of Overlay Anonymous Multicast Protocol

A Design of Overlay Anonymous Multicast Protocol A Deign of Ovelay Anonymou Multicat Potocol Li Xiao, Xiaomei Liu, Wenun Gu, Dong Xuan, Yunhao Liu 3 Dept. of Compute Science and Engineeing, Michigan State Univeity, Eat Laning, MI 4884, USA Dept. of Compute

More information

Validation Plan for SPDM Task Verification Facility

Validation Plan for SPDM Task Verification Facility Poceeding of the 7 th Intenational Sympoium on Atificial Intelligence, Robotic and Automation in Space: i-sairas 2003, NARA, Japan, May 19-23, 2003 Validation Plan fo Tak Veification Facility Ou Ma New

More information

K-Nearest Neighbors Algorithm: Prediction and Classification

K-Nearest Neighbors Algorithm: Prediction and Classification K-Neaet Neighbo Algoithm: Peiction an Claification Pof. Thoma B. Fomb Depatment of Economic Southen Methoit Univeit Dalla, TX 7575 Febua 8 A NONPARAMETRIC METHOD The K-Neaet Neighbo (K-NN algoithm i a

More information

Chinese Traffic Police Gesture Recognition in Complex Scene

Chinese Traffic Police Gesture Recognition in Complex Scene 0 Intenational oint Confeence of IEEE TutCom-/IEEE ICESS-/FCST- Chinee Taffic Police Getue Recognition in Complex Scene Fan Guo, Zixing Cai, in Tang School of Infomation Science and Engineeing Cental South

More information

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

P284 A Full-wave Equation Based Seismic Illumination Analysis Method

P284 A Full-wave Equation Based Seismic Illumination Analysis Method P284 A Full-wave Equation Baed Seimic Illumination Analyi Method X.-B. Xie* (Univeity of Califonia) & H. Yang (Univeity of Califonia) SUMMARY We popoe a full-wave equation baed method fo eimic illumination

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

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

COMPUTER SCIENCE AND ENGINEERING VI SEM CSE

COMPUTER SCIENCE AND ENGINEERING VI SEM CSE COMPUTER SCIENCE AND ENGINEERING VI SEM CSE Pinciple of Compile Deign Quetion and anwe ----------------------------------------------------------------------------------------------------- 1) What i a

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

Intelligent Knowledge Discovery in Peer-to-Peer File Sharing

Intelligent Knowledge Discovery in Peer-to-Peer File Sharing Intelligent Knowledge Dicovey in Pee-to-Pee File Shaing Yugyung Lee School of Intediciplinay Computing and Engineeing Univeity of Mioui Kana City leeyu@umkc.edu Changgyu Oh School of Intediciplinay Computing

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

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

Representing Texture Images Using Asymmetric Gray Level Aura Matrices

Representing Texture Images Using Asymmetric Gray Level Aura Matrices Repeenting Textue Image Uing Aymmetic Gay Level Aua Matice Xueie Qin Yee-Hong Yang {xuq, yang}@c.ualbeta.ca Depatment of Computing Science Univeity of Albeta Abtact In thi pape, we peent a new mathematical

More information

Routing Definition 4.1

Routing Definition 4.1 4 Routing So far, we have only looked at network without dealing with the iue of how to end information in them from one node to another The problem of ending information in a network i known a routing

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

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

Visual Servoing from Spheres using a Spherical Projection Model

Visual Servoing from Spheres using a Spherical Projection Model Viual Sevoing fom Sphee uing a Spheical Pojection Model omeo Tatambon Fomena and Fançoi Chaumette Abtact In thi pape, we invetigate the ue of a pheical pojection model to each fo optimal viual featue fo

More information

Lecture 14: Minimum Spanning Tree I

Lecture 14: Minimum Spanning Tree I COMPSCI 0: Deign and Analyi of Algorithm October 4, 07 Lecture 4: Minimum Spanning Tree I Lecturer: Rong Ge Scribe: Fred Zhang Overview Thi lecture we finih our dicuion of the hortet path problem and introduce

More information

MAC Protocol for Supporting QoS in All-IP HiperLAN2

MAC Protocol for Supporting QoS in All-IP HiperLAN2 MAC Potocol fo Suppoting QoS in All-IP HipeLAN Chistos E. Politis, Rahim Tafazolli Univesity of Suey Cente fo Communication Systems Reseach Guildfod, GU 7XH, Suey, UK Tel.: +44 483 683463, FAX: +44 483

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

Regular Expressions - a Graphical User Interface

Regular Expressions - a Graphical User Interface Regula Expeion - a Gaphical Ue Inteface Stefan Kah Univeity of Kent at Cantebuy Computing Laboatoy email: mk@ukc.ac.uk Septembe 13, 1999 Abtact Regula expeion ae a tandad tool fo language deciption, in

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

Analysis of Circuit Switching for the Torus Interconnect Networks with Hot-Spot Traffic *

Analysis of Circuit Switching for the Torus Interconnect Networks with Hot-Spot Traffic * Analyi of Cicuit Switching fo the Tou Inteconnect Netwok with Hot-Spot Taffic * F. Safaei,3, A. Khonai,, M. Fathy 3, M. Ould-Khaoua 4 IPM School of Compute Science, Tehan, Ian Dept. of Electical and Compute

More information

Applications. Sumit Basu*, Kentaro Toyamay, and Alex Pentland* *MIT Media Laboratory and ymicrosoft Research.

Applications. Sumit Basu*, Kentaro Toyamay, and Alex Pentland* *MIT Media Laboratory and ymicrosoft Research. M.I.T Media Laboatoy Peceptual Computing Section Technical Repot No. 486 ubmitted to ICCV'99 A Conitent Method fo Function Appoximation in Meh-baed Application Sumit Bau*, Kentao Toyamay, and Alex Pentland*

More information

Angle domain illumination analysis along a target horizon Jian Mao, Sang Suh and Bin Wang, TGS

Angle domain illumination analysis along a target horizon Jian Mao, Sang Suh and Bin Wang, TGS Downloaded /9/3 to.9.79.3. Reditibution ubject to SEG licene o copyiht; ee Tem of Ue at http://libay.e.o/ Anle domain illumination analyi alon a taet hoizon Jian Mao, San Suh and Bin Wan, TGS Summay We

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

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

A REGION-BASED TECHNIQUE FOR FUSION OF HIGH RESOLUTION IMAGES USING MEAN SHIFT SEGMENTATION

A REGION-BASED TECHNIQUE FOR FUSION OF HIGH RESOLUTION IMAGES USING MEAN SHIFT SEGMENTATION REGION-BSED TECHNIQUE FOR FUSION OF HIGH RESOLUTION IMGES USING MEN SHIFT SEGMENTTION Li Shuang a, Li Zhilin a, b a LIESMRS, Wuhan Univeit, P.R. China - lihuang9@gmail.com b Dept. of Land Suveing and Geo-Infomatic,

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

A Window Retrieval Algorithm for Spatial Databases Using Quadtrees

A Window Retrieval Algorithm for Spatial Databases Using Quadtrees A Window Retieval Algoithm fo Spatial Databae Uing Quadtee Walid G. Aef Matuhita Infomation Technology Laboatoy Panaonic Technologie, Inc. Reeach Way, Pinceton, NJ 00 Phone: 60-3-3, Fax: 60-- E-mail: aef@mitl.eeach.panaonic.com

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

Translation of Specification Units Between IP and ATM Quality of Service Declarations

Translation of Specification Units Between IP and ATM Quality of Service Declarations Translation of Specification Units Between IP and ATM Quality of Service Declarations Jens Schmitt Multimedia Communications Lab (KOM) Department of Electrical Engineering and Information Technology Darmstadt

More information

A Correctness Verification Approach of the BPEL Exception Handling CPN Model Based on Temporal Property

A Correctness Verification Approach of the BPEL Exception Handling CPN Model Based on Temporal Property JOURNAL OF NETWORKS, VOL. 9, NO. 10, OCTOBER 2014 2743 A Coectne Veification Appoach of the BPEL Exception Handling CPN Model Baed on Tempoal Popety Guan Hua 1, 2, Ying Shi 2, and Wang Chunwei 3 1. Netwok

More information

Boundary Preserving Joint Estimation of Optical Flow and Disparity in a Sequence of Stereoscopic Images

Boundary Preserving Joint Estimation of Optical Flow and Disparity in a Sequence of Stereoscopic Images Bounday Peeving oint Etimation of Optical Flow and Dipaity in a Sequence of Steeocopic Image H. Weile, A. Mitiche, A. Manoui INRS-EMT Place Bonaventue 8 ue de la Gauchetiee O. Suite 69 Monteal, Quebec

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

Biometric authentication using digital retinal images

Biometric authentication using digital retinal images Biometic authentication uing digital etinal image M. ORTEGA, C.MARIÑO, M.G.PENEDO, M.BLANCO, F.GONZÁLEZ Gupo de Viión Atificial y Reconocimiento de Patone Univeity of A Couña Campu de Elviña /n, A Couña,

More information

Lecture overview. Visualisatie BMT. Visualization pipeline. Data representation. Discrete data. Sampling. Data Datasets Interpolation

Lecture overview. Visualisatie BMT. Visualization pipeline. Data representation. Discrete data. Sampling. Data Datasets Interpolation Viualiatie BMT Lectue oveview Data Dataet Intepolation Data epeentation Ajan Kok a.j.f.kok@tue.nl 2 Viualization pipeline Raw Data Data Enichment/Enhancement Deived Data Data epeentation Viualization data

More information

Loss-Cognizant Pricing in Feed-Forward Networks with Greedy Users

Loss-Cognizant Pricing in Feed-Forward Networks with Greedy Users Loss-Cognizant Picing in Feed-Fowad Netwoks with Geedy Uses Ashaf Al Daoud and Muat Alanyali Depatment of Electical and Compute Engineeing Boston Univesity, Boston, MA 02215 Email: {ashaf, alanyali}@bu.edu

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

where f(x, y): input image, g(x, y): processed image, and T: operator Or: s = T(r), where r: input pixel, and s: output pixel

where f(x, y): input image, g(x, y): processed image, and T: operator Or: s = T(r), where r: input pixel, and s: output pixel 3 Intenit Tanfomation and Spatial Filteing - Intenit tanfomation Change the intenit of each piel in ode to enhance the image: g, T[f, ], whee f, : input image, g, : poceed image, and T: opeato O: T, whee

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

Comparisons of Transient Analytical Methods for Determining Hydraulic Conductivity Using Disc Permeameters

Comparisons of Transient Analytical Methods for Determining Hydraulic Conductivity Using Disc Permeameters Compaisons of Tansient Analytical Methods fo Detemining Hydaulic Conductivity Using Disc Pemeametes 1,,3 Cook, F.J. 1 CSRO Land and Wate, ndoooopilly, Queensland The Univesity of Queensland, St Lucia,

More information

On Error Estimation in Runge-Kutta Methods

On Error Estimation in Runge-Kutta Methods Leonado Jounal of Sciences ISSN 1583-0233 Issue 18, Januay-June 2011 p. 1-10 On Eo Estimation in Runge-Kutta Methods Ochoche ABRAHAM 1,*, Gbolahan BOLARIN 2 1 Depatment of Infomation Technology, 2 Depatment

More information

A Coarse-to-Fine Algorithm for 3D Registration based on Wavelet Decomposition

A Coarse-to-Fine Algorithm for 3D Registration based on Wavelet Decomposition A oae-to-fine Algoithm fo 3D Regitation baed on Wavelet Decompoition. TORRE-FERRERO, S. ROBLA, E.G. SARABIA, J.R. LLATA Electonic Technology, Sytem and Automation Engineeing Depatment Univeity of antabia

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

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

Gravitational Shift for Beginners

Gravitational Shift for Beginners Gavitational Shift fo Beginnes This pape, which I wote in 26, fomulates the equations fo gavitational shifts fom the elativistic famewok of special elativity. Fist I deive the fomulas fo the gavitational

More information

Hierarchically Clustered P2P Streaming System

Hierarchically Clustered P2P Streaming System Hieachically Clusteed P2P Steaming System Chao Liang, Yang Guo, and Yong Liu Polytechnic Univesity Thomson Lab Booklyn, NY 11201 Pinceton, NJ 08540 Abstact Pee-to-pee video steaming has been gaining populaity.

More information

Classification by crossing and polling for integrated processing of maps and surveys. An addendum to GRCP-software version 2

Classification by crossing and polling for integrated processing of maps and surveys. An addendum to GRCP-software version 2 Stichting Ondezoek Weeldvoedelvooziening van de Vije Univeiteit Cente fo Wold Food Studie Claification by coing and polling fo integated poceing of map and uvey An addendum to GRCP-oftwae veion 2 by Michiel

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

Worst-Case Delay Bounds for Uniform Load-Balanced Switch Fabrics

Worst-Case Delay Bounds for Uniform Load-Balanced Switch Fabrics Wost-Case Delay Bounds fo Unifom Load-Balanced Switch Fabics Spyidon Antonakopoulos, Steven Fotune, Rae McLellan, Lisa Zhang Bell Laboatoies, 600 Mountain Ave, Muay Hill, NJ 07974 fistname.lastname@alcatel-lucent.com

More information

Topics. Lecture 37: Global Optimization. Issues. A Simple Example: Copy Propagation X := 3 B > 0 Y := 0 X := 4 Y := Z + W A := 2 * 3X

Topics. Lecture 37: Global Optimization. Issues. A Simple Example: Copy Propagation X := 3 B > 0 Y := 0 X := 4 Y := Z + W A := 2 * 3X Lecture 37: Global Optimization [Adapted from note by R. Bodik and G. Necula] Topic Global optimization refer to program optimization that encompa multiple baic block in a function. (I have ued the term

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

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

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

Second Order Edge-preserving Regularized Functional for Recovering Discontinuous Phase Maps in Fringe Patterns with Carrier

Second Order Edge-preserving Regularized Functional for Recovering Discontinuous Phase Maps in Fringe Patterns with Carrier Second Ode Edge-peeving Regulaized Functional fo Recoveing Dicontinuou Phae Map in Finge Patten with Caie Calo A. Galvan a and Maiano Rivea b a Cento Nacional de Metologia, Apdo. Potal 1-1 Cento, Queetao,

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

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

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

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

Lecture Outline. Global flow analysis. Global Optimization. Global constant propagation. Liveness analysis. Local Optimization. Global Optimization

Lecture Outline. Global flow analysis. Global Optimization. Global constant propagation. Liveness analysis. Local Optimization. Global Optimization Lecture Outline Global flow analyi Global Optimization Global contant propagation Livene analyi Adapted from Lecture by Prof. Alex Aiken and George Necula (UCB) CS781(Praad) L27OP 1 CS781(Praad) L27OP

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

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

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

FACE VECTORS OF FLAG COMPLEXES

FACE VECTORS OF FLAG COMPLEXES FACE VECTORS OF FLAG COMPLEXES ANDY FROHMADER Abstact. A conjectue of Kalai and Eckhoff that the face vecto of an abitay flag complex is also the face vecto of some paticula balanced complex is veified.

More information

The Open Biotechnology Journal, 2015, 9, Experimental and Numerical Analysis of Drip Tape Layout for Irrigation of Sugarcane in Latosol

The Open Biotechnology Journal, 2015, 9, Experimental and Numerical Analysis of Drip Tape Layout for Irrigation of Sugarcane in Latosol Send Ode fo Repint to epint@benthamcience.ae The Open Biotechnology Jounal, 215, 9, 265-271 265 Open Acce Expeimental and Numeical Analyi of Dip Tape Layout fo Iigation of Sugacane in Latool Kai Huang

More information

Type-Safe Multithreading in Cyclone

Type-Safe Multithreading in Cyclone Type-Safe Multitheading in Cyclone Dan Goman Compute Science Depatment Conell Univeity Ithaca, NY 14853 danieljg@c.conell.edu ABSTRACT We extend Cyclone, a type-afe polymophic language at the C level of

More information

CS 450: COMPUTER GRAPHICS QUATERNIONS SPRING 2016 DR. MICHAEL J. REALE

CS 450: COMPUTER GRAPHICS QUATERNIONS SPRING 2016 DR. MICHAEL J. REALE CS 45: COMPUTER GRAPHICS QUATERNIONS SPRING 6 DR. MICHAEL J. REALE http://common.ikimedia.og/iki/fi le%3awilliam_roan_hamilton_pot ait_oal_combined.png INTRODUCTION Quatenion inented b Si William Roan

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

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

Modeling a shared medium access node with QoS distinction

Modeling a shared medium access node with QoS distinction Modeling a shaed medium access node with QoS distinction Matthias Gies, Jonas Geutet Compute Engineeing and Netwoks Laboatoy (TIK) Swiss Fedeal Institute of Technology Züich CH-8092 Züich, Switzeland email:

More information

Mobility Pattern Recognition in Mobile Ad-Hoc Networks

Mobility Pattern Recognition in Mobile Ad-Hoc Networks Mobility Patten Recognition in Mobile Ad-Hoc Netwoks S. M. Mousavi Depatment of Compute Engineeing, Shaif Univesity of Technology sm_mousavi@ce.shaif.edu H. R. Rabiee Depatment of Compute Engineeing, Shaif

More information

Conversion Functions for Symmetric Key Ciphers

Conversion Functions for Symmetric Key Ciphers Jounal of Infomation Assuance and Secuity 2 (2006) 41 50 Convesion Functions fo Symmetic Key Ciphes Deba L. Cook and Angelos D. Keomytis Depatment of Compute Science Columbia Univesity, mail code 0401

More information

On the throughput-cost tradeoff of multi-tiered optical network architectures

On the throughput-cost tradeoff of multi-tiered optical network architectures 1 On the thoughput-cost tadeoff of multi-tieed optical netwok achitectues Guy Weichenbeg, Vincent W. S. Chan, and Muiel Médad Laboatoy fo Infomation and Decision Systems Massachusetts Institute of Technology

More information

Automatically Testing Interacting Software Components

Automatically Testing Interacting Software Components Automatically Testing Inteacting Softwae Components Leonad Gallaghe Infomation Technology Laboatoy National Institute of Standads and Technology Gaithesbug, MD 20899, USA lgallaghe@nist.gov Jeff Offutt

More information

A Shape-preserving Affine Takagi-Sugeno Model Based on a Piecewise Constant Nonuniform Fuzzification Transform

A Shape-preserving Affine Takagi-Sugeno Model Based on a Piecewise Constant Nonuniform Fuzzification Transform A Shape-peseving Affine Takagi-Sugeno Model Based on a Piecewise Constant Nonunifom Fuzzification Tansfom Felipe Fenández, Julio Gutiéez, Juan Calos Cespo and Gacián Tiviño Dep. Tecnología Fotónica, Facultad

More information

Development of MITC isotropic triangular shell finite elements

Development of MITC isotropic triangular shell finite elements Compute and Stuctue 8 (4) 945 96 www.elevie.com/locate/compuc Development of MITC iotopic tiangula hell finite element Phill-Seung Lee, Klau-J ugen Bathe * Maachuett Initute of Technology, Cambidge, MA

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

CS 548: COMPUTER GRAPHICS QUATERNIONS SPRING 2015 DR. MICHAEL J. REALE

CS 548: COMPUTER GRAPHICS QUATERNIONS SPRING 2015 DR. MICHAEL J. REALE CS 548: COMPUTER GRAPHICS QUATERNIONS SPRING 5 DR. MICHAEL J. REALE http://common.ikimedia.og/iki/fi le%3awilliam_roan_hamilton_pot ait_oal_combined.png INTRODUCTION Quatenion inented b Si William Roan

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

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

An Efficient Video Copy Detection Method Combining Vocabulary Tree and Inverted File

An Efficient Video Copy Detection Method Combining Vocabulary Tree and Inverted File An Eicient Video Copy Detection Method Combining Vocabulay Tee and Inveted File Xuan Li 1, Bing Li 2, Weiming Hu 2, Jineng Yang 1, 1 College o aviation automation, Civil Aviation Univeity o China, Tianjin,

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

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

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

User Specified non-bonded potentials in gromacs

User Specified non-bonded potentials in gromacs Use Specified non-bonded potentials in gomacs Apil 8, 2010 1 Intoduction On fist appeaances gomacs, unlike MD codes like LAMMPS o DL POLY, appeas to have vey little flexibility with egads to the fom of

More information

1 The secretary problem

1 The secretary problem Thi i new material: if you ee error, pleae email jtyu at tanford dot edu 1 The ecretary problem We will tart by analyzing the expected runtime of an algorithm, a you will be expected to do on your homework.

More information

Package vitality. R topics documented: May 13, 2018

Package vitality. R topics documented: May 13, 2018 Package vitality May 13, 2018 Title Fitting Routine fo the Vitality Family of Motality Model Veion 1.3 Date 2018-05-13 Autho Gego Paolt, Jame J. Andeon, Ting Li, David H. Salinge, David J. Shaow Maintaine

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

Minimum congestion spanning trees in bipartite and random graphs

Minimum congestion spanning trees in bipartite and random graphs Minimum congetion panning tree in bipartite and random graph M.I. Otrovkii Department of Mathematic and Computer Science St. John Univerity 8000 Utopia Parkway Queen, NY 11439, USA e-mail: otrovm@tjohn.edu

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