On Optimization of Network-coded Scalable Multimedia Service Multicasting
1 like370 views
The document outlines the R2D2 project focused on optimizing data delivery for multimedia broadcasting over 4G/5G networks, particularly through error control and resource optimization strategies. It discusses a multi-rate transmission system that delivers layered services to users based on their channel conditions, utilizing techniques like Random Linear Network Coding (RLNC). Additionally, various resource allocation models and heuristic strategies are proposed to enhance the efficient delivery of service levels while minimizing radio resource usage.
On Optimization of Network-coded Scalable Multimedia Service Multicasting
- 1. Bristol, 5th February 2015 R2D2: Network error control for Rapid and Reliable Data Delivery Project supported by EPSRC under the First Grant scheme (EP/L006251/1) On Optimization of Network-coded Scalable Multimedia Service Multicasting University of Bristol Andrea Tassi and Ioannis Chatzigeorgiou {a.tassi, i.chatzigeorgiou}@lancaster.ac.uk andCommunications School of Computing
- 2. Starting Point and Goals ๏ Delivery of multimedia broadcast/multicast services over 4G/5G networks is a challenging task. This has propelled research into delivery schemes. ๏ Multi-rate Transmission (MrT) strategies have been proposed as a means of delivering layered services to users experiencing different downlink channel conditions. ๏ Layered service consists of a basic layer and multiple enhancement layers. Goals ๏ Error control - Ensure that a predetermined fraction of users achieves a certain service level with at least a given probability ๏ Resource optimisation - Reduce the total amount of radio resources needed to deliver a layered service. 2 andCommunications School of Computing
- 3. Index 1. System Parameters and Performance Analysis 2. Multi-Channel Resource Allocation Models and Heuristic Strategies 3. Analytical Results 4. Concluding Remarks 3 andCommunications School of Computing
- 4. 1. System Parameters and Performance Analysis andCommunications School of Computing
- 5. System Model ๏ One-hop wireless communication system composed of one source node and U users 5 UE 1 UE 3 UE 2 UE 4 UE U Source Node ˆB3 ˆB2 ˆB1 subch. 1 subch. 2 subch. 3 ๏ Each PtM layered service is delivered through C orthogonal broadcast erasure subchannels The$same$MCS Capacity$of$subch.$3$ (no.$of$packets) ๏ Each subchannel delivers streams of (en)coded packets (according to the RLNC principle). andCommunications School of Computing
- 6. 6 ๏ Encoding performed over each service layer independently from the others. ๏ The source node will linearly combine the data packets composing the l-th layer and will generate a stream of coded packets , where k1 k2 k3 x1 x2 xK. . .. . . ๏ is a layered source message of K source packets, classified into L service layers x = {x1, . . . , xK} Non-Overlapping Layered RNC andCommunications School of Computing kl xl = {xi}kl i=1 nl kl y = {yj}nl j=1 yj = klX i=1 gj,i xi Coef:icients$of$the$ linear$combination$ are$selected$over$a$ :inite$:ield$of$size$q
- 7. Non-Overlapping Layered RNC ๏ User u recovers layer l if it will collect k_l linearly independent coded packets. The prob. of this event is 7 kl Pl(nl,u) = nl,u X r=kl ✓ nl,u r ◆ pnl,u r (1 p)r h(r) = nl,u X r=kl ✓ nl,u r ◆ pnl,u r (1 p)r kl 1Y i=0 h 1 1 qr i i | {z } h(r) Prob.$of$receiving$r$out$of$nl,u$coded$symbols Prob.$of$decoding$ layer$l ๏ The probability that user u recover the first l service layers is DNO,l(n1,u, . . . , nL,u) = DNO,l(nu) = lY i=1 Pi(ni,u) PEP andCommunications School of Computing
- 8. 8 ๏ The source node (i) linearly combines data packets belonging to the same window, (ii) repeats this process for all windows, and (iii) broadcasts each stream of coded packets over one or more subchannels Expanding Window Layered RNC ๏ We define the l-th window as the set of source packets belonging to the first l service layers. Namely, where Xl Xl ={xj}Kl j=1 Kl = Pl i=1 ki k1 k2 k3 x1 x2 xK. . .. . . Exp. Win. 3 Exp. Win. 2 Exp. Win. 1 andCommunications School of Computing
- 9. Expanding Window Layered RNC ๏ Sums allow us to consider all the possible combinations of received coded packets 9 ๏ The probability of user u recovering the first l layers (namely, the l-th window) can be written as DEW,l L,u) = =DEW,l(Nu) = N1,u X r1=0 · · · Nl 1,u X rl 1=0 Nl,u X rl=rmin,l ✓ N1,u r1 ◆ · · · ✓ Nl,u rl ◆ p Pl i=1(Ni,u ri) (1 p) Pl i=1 ri gl(r) Prob.$of$receiving $out$ of$ coded$symbols Prob.$of$decoding$ window$l DEW,l(Nu) DEW,l(N1,u, . . . , NL,u) = =DEW,l(Nu) = N1,u X r1=0 · · · Nl 1,u X rl 1=0 Nl,u X rl=rmin,l ✓ N1,u r1 ◆ · · · ✓ Nl,u rl ◆ p Pl i=1(Ni,u ri) ( r = {r1, . . . , rl} andCommunications School of Computing
- 10. 2. Multi-Channel Resource Allocation Models andCommunications School of Computing
- 11. Allocation Patterns 11 ˆB3 ˆB2 ˆB1 subchannel 1 subchannel 2 subchannel 3 coded packets from x1 coded packets from x2 coded packets from x3 ˆB3 ˆB2 ˆB1 subchannel 1 subchannel 2 subchannel 3 Separated$ Allocation$ Pattern andCommunications School of Computing
- 12. Mixed$ Allocation$ Pattern ˆB3 ˆB2 ˆB1 coded packets from x3 or X3 coded packets from x2 or X2 coded packets from x1 or X1 subchannel 1 subchannel 2 subchannel 3 Allocation Patterns 11 ˆB3 ˆB2 ˆB1 subchannel 1 subchannel 2 subchannel 3 andCommunications School of Computing
- 13. NO-MA Model 12 andCommunications School of Computing ๏ Consider the variable . It is 1, if u can recover the first l layers with a probability value , otherwise it is 0. u,l = I ⇣ DNO,l(nu) ˆD ⌘ ˆD (NO-MA) min m1,...,mC n(1,c),...,n(L,c) LX l=1 CX c=1 n(l,c) (1) subject to UX u=1 u,l U ˆtl l = 1, . . . , L (2) mc 1 < mc c = 2, . . . , L (3) 0 LX l=1 n(l,c) ˆBc c = 1, . . . , C (4) No.$of$packets$of$layer$l$ delivered$over$cMinimization$of$ resource$footprint ˆB3 ˆB2 ˆB1 subch. 1subch. 2subch. 3
- 14. NO-MA Model 12 andCommunications School of Computing ๏ Consider the variable . It is 1, if u can recover the first l layers with a probability value , otherwise it is 0. u,l = I ⇣ DNO,l(nu) ˆD ⌘ ˆD (NO-MA) min m1,...,mC n(1,c),...,n(L,c) LX l=1 CX c=1 n(l,c) (1) subject to UX u=1 u,l U ˆtl l = 1, . . . , L (2) mc 1 < mc c = 2, . . . , L (3) 0 LX l=1 n(l,c) ˆBc c = 1, . . . , C (4) Each$service$level$shall$be$ achieved$by$a$predetermined$ fraction$of$users No.$of$users Target$fraction$of$users (NO-MA) min m1,...,mC n(1,c),...,n(L,c) LX l=1 CX c=1 n(l,c) (1) subject to UX u=1 u,l U ˆtl l = 1, . . . , L (2) mc 1 < mc c = 2, . . . , L (3) 0 LX l=1 n(l,c) ˆBc c = 1, . . . , C (4) ˆB3 ˆB2 ˆB1 subch. 1subch. 2subch. 3
- 15. NO-MA Model 12 andCommunications School of Computing ๏ Consider the variable . It is 1, if u can recover the first l layers with a probability value , otherwise it is 0. u,l = I ⇣ DNO,l(nu) ˆD ⌘ ˆD (NO-MA) min m1,...,mC n(1,c),...,n(L,c) LX l=1 CX c=1 n(l,c) (1) subject to UX u=1 u,l U ˆtl l = 1, . . . , L (2) mc 1 < mc c = 2, . . . , L (3) 0 LX l=1 n(l,c) ˆBc c = 1, . . . , C (4) DynamicH$and$ systemHrelated$ constraints (NO-MA) min m1,...,mC n(1,c),...,n(L,c) LX l=1 CX c=1 n(l,c) (1) subject to UX u=1 u,l U ˆtl l = 1, . . . , L (2) mc 1 < mc c = 2, . . . , L (3) 0 LX l=1 n(l,c) ˆBc c = 1, . . . , C (4) (NO-MA) min m1,...,mC n(1,c),...,n(L,c) LX l=1 CX c=1 n(l,c) (1) subject to UX u=1 u,l U ˆtl l = 1, . . . , L (2) mc 1 < mc c = 2, . . . , L (3) 0 LX l=1 n(l,c) ˆBc c = 1, . . . , C (4) ˆB3 ˆB2 ˆB1 subch. 1subch. 2subch. 3
- 16. NO-MA Heuristic ๏ The NO-MA is an hard integer optimisation problem because of the coupling constraints among variables ๏ We propose a two-step heuristic strategy i. MCSs optimisation ( ) ii. No. of coded packet per-subchannel optimization ( ) 13 m1, . . . , mC n(1,c) , . . . , n(L,c) ๏ The first step selects the value of such that packets delivered through subch. c are received (at least with a target prob.) by users. mc |U(mc) | U · ˆtc apping Resource Allocation Strategies system where the source node delivers the by means of the NO RNC principle. From (??), indication variable u,l as follows: u,l = I ⇣ DNO,l(nu) ˆD ⌘ . (13) s, u,l = 1, if u can recover the first l layers ity value that is equal to or greater than a target wise u,l = 0. e allocation model that we propose for the Step 1 Subchannel MCSs optimization. 1: c C 2: v mMAX and 3: while c 1 do 4: repeat 5: mc v 6: v v 1 7: until |U(mc) | U · ˆtc or v < mmin 8: c c 1 9: end while because of the nature of the considered a andCommunications School of Computing
- 17. NO-MA Heuristic ๏ The idea behind the second step can be summarised as follows 14 ˆB3 ˆB2 ˆB1 subchannel 1 subchannel 2 subchannel 3 DNO,1(n(1) ) ˆD DNO,2(n(1) , n(2) ) ˆD DNO,3(n(1) , n(2) , n(3) ) ˆD andCommunications School of Computing
- 18. NO-MA Heuristic ๏ The idea behind the second step can be summarised as follows 14 e node delivers the principle. From (6), follows: ˆD ⌘ . (13) ver the first l layers greater than a target we propose for the A (NO-SA) can be (l,c) (14) = 1, . . . , L (15) = 2, . . . , L (16) = 1, . . . , C (17) or l 6= c (18) nts the overall num- d to deliver all the L 1: c C 2: v mMAX and 3: while c 1 do 4: repeat 5: mc v 6: v v 1 7: until |U(mc) | U · ˆtc or v < mmin 8: c c 1 9: end while Step 2 Coded packet allocation for a the NO-MA case. 1: c 1 2: n(l,c) 1 for any l = 1, . . . , L and c = 1, . . . , C 3: n = {n(l) }L l=1, where n(l) 1 for any l = 1, . . . , L 4: for l 1, . . . , L do 5: while DNO,l(n) < ˆD and c C do 6: n(l,c) n(l,c) + 1 7: n(l) PC t=1 n(l,t) for any l = 1, . . . , L 8: if PL t=1 n(t,c) = ˆBc then 9: c c + 1 10: end if 11: end while 12: if DNO,l(n) < ˆD and c > C then 13: no solution can be found. 14: end if 15: end for because of the nature of the considered allocation pattern. Furthermore, without loss of generality, we assume that coded Requires$a$no.$of$steps PC t=1 ˆBt andCommunications School of Computing
- 19. EW-MA Model ๏ We define the indicator variable User u will recover the first l service layers (at least) with probability if any of the windows l, l+1, …, L are recovered (at least) with probability 15 µu,l = I L_ t=l n DEW,t(Nu) ˆD o ! ˆD ˆD ๏ Consider the EW delivery mode andCommunications School of Computing k1 k2 k3 x1 x2 xK. . .. . . Exp. Win. 3 Exp. Win. 2 Exp. Win. 1 ˆB3 ˆB2 ˆB1 subch. 1subch. 2subch. 3
- 20. EW-MA Model ๏ The RA problem for the EW-MA case is 16 (EW-MA) min m1,...,mC N(1,c),...,N(L,c) LX l=1 CX c=1 N(l,c) (1) subject to UX u=1 µu,l U ˆtl l = 1, . . . , L (2) mc 1 < mc c = 2, . . . , L (3) 0 LX l=1 N(l,c) ˆBc c = 1, . . . , C (4) No.$of$packets$of$ window$l$delivered$ over$c ๏ It is still an hard integer optimisation problem but the previously proposed heuristic strategy can be still applied. andCommunications School of Computing ˆB3 ˆB2 ˆB1 subch. 1subch. 2subch. 3
- 21. “Egalitarian” Model 17 andCommunications School of Computing ๏ Previous strategies ensure minimum SLA and minimize the resource footprint. Point of view of the ISP… ๏ Best practice for burglars - To still object with the maximum value and the minimum weight. The profit-cost ratio is maximized. ๏ We define the model for a SA pattern as: (E-SA) maximize m1,...,mL N(1),...,N(L) UX u=1 LX l=1 Qu,l , LX l=1 N(l) (1) subject to UX u=1 Qu,l U ˆtl l = 1, . . . , L (2) 0 N(l) ˆBl l = 1, . . . , L. (3) Pro(it$H$No.$of$video$layers$ recovered$by$any$of$the$users Cost$H$No.$of$ transmissions$needed SLAHrelated$constraint ๏ We can refer to the previous heuristics. ˆB3 ˆB2 ˆB1 subch. 1subch. 2subch. 3
- 22. 3. Analytical Results andCommunications School of Computing
- 23. Analytical Results (part 1) 19 ๏ LTE-A eMBMS scenarios ๏ We compared the proposed strategies with a classic Multi- rate Transmission strategy ๏ System performance was evaluated in terms of It$is$a$maximization$of$the$ sum$of$the$user$QoS Resource$footprint = 8 >>>>< >>>>: LX l=1 CX c=1 n(l,c) , for NO-RNC LX l=1 CX c=1 N(l,c) , for EW-RNC. max m1,...,mL UX u=1 PSNRu No$error$control$strategies$ are$allowed$(ARQ,$RLNC,$etc.) andCommunications School of Computing
- 24. Analytical Results (part 1) 19 ๏ LTE-A eMBMS scenarios ๏ We compared the proposed strategies with a classic Multi- rate Transmission strategy ๏ System performance was evaluated in terms of It$is$a$maximization$of$the$ sum$of$the$user$QoS PSNR$after$recovery$of$the$basic$ and$the$:irst$l$enhancement$layers ⇢(u) = 8 >< >: max l=1,...,L n PSNRl D (u) NO,l o , for NO-RNC max l=1,...,L n PSNRl D (u) EW,l o , for EW-RNC max m1,...,mL UX u=1 PSNRu No$error$control$strategies$ are$allowed$(ARQ,$RLNC,$etc.) andCommunications School of Computing
- 25. Target cellTarget MG eNB Scenario$with$a$high$heterogeneity.$80$ UEs$equally$spaced$and$$placed$along$the$ radial$line$representing$the$symmetry$ axis$of$one$sector$of$the$target$cell Analytical Results (part 1) 20 We$considered$Stream$A$and$B$ which$have$3$layers,$bitrate$of$ A$is$smaller$than$that$of$B andCommunications School of Computing
- 26. Distance (m) MaximumPSNRρ(dB) 90 110 130 150 170 190 210 230 250 270 290 0 5 15 25 35 45 55 ˆt1ˆt2ˆt3 MrT Heu. NOW−MA Heu. EW−MA ⌧ = 60 ⌧ = 43 Analytical Results (part 1) 21 Stream$A andCommunications School of Computing All$the$proposed$ strategies$meet$ the$coverage$ constraints EWHMA NOHMA MrT PSNR layers$1+2+3 PSNR layers$1+2 PSNR layers$1
- 27. Distance (m) MaximumPSNRρ(dB) 90 110 130 150 170 190 210 230 250 270 290 0 5 15 25 35 45 55 ˆt1ˆt2ˆt3 MrT Heu. NOW−MA Heu. EW−MA ⌧ = 73 ⌧ = 88 Analytical Results (part 1) 22 andCommunications School of Computing All$the$proposed$ strategies$meet$ the$coverage$ constraints MrT EWHMA NOHMA PSNR layers$1+2+3 PSNR layers$1+2 PSNR layers$1 Stream$B
- 28. Analytical Results (part 2) 23 andCommunications School of Computing ๏ LTE-A allows multiple contiguous BS to deliver (in a synchronous fashion) the same services by means of the same signals eNB eNB eNB eNB M1/M2 MCE / MBMS-GW SFN 3 2 1 B UE3 UEMUE2 UE1 UE4 −400 −200 0 200 400 600 −200 −100 0 100 200 300 400 500 600 700 0 5 10 15 Spacial$SINR$distribution Single$Frequency$Network 4HBS$SFN,$1700$users$placed$at$ the$vertices$of$a$regular$square$ grid$placed$on$the$playground.$
- 29. Analytical Results (part 2) 24 andCommunications School of Computing x position (m) yposition(m) −500 −300 −100 100 300 500 700 −200 −100 0 100 200 300 400 500 600 700 x position (m) yposition(m) −500 −300 −100 100 300 500 700 −200 −100 0 100 200 300 400 500 600 700 45.8 45.8 45.8 45.8 45.8 45.845.8 45.8 45.8 45.8 35.9 35.9 35.9 35.9 35.9 35.9 35.9 35.9 35.9 35.9 35.9 27.9 27.9 27.9 27.9 27.9 27.9 27.9 27.9 27.9 27.9 27.9 27.9 E"SAMrT 45.8 45.8 3Hlayer$ stream ๏ Also in this case MrT cannot ensure the desired coverage!
- 30. x position (m) yposition(m) −500 −300 −100 100 300 500 700 −200 −100 0 100 200 300 400 500 600 700 x position (m) yposition(m) −500 −300 −100 100 300 500 700 −200 −100 0 100 200 300 400 500 600 700 46.4 46.4 46.4 46.4 46.4 46.4 46.4 46.4 46.4 46.4 46.4 46.4 39.9 39.9 39.9 39.9 39.9 39.9 39.9 39.9 39.939.9 39.9 33.4 33.4 33.4 33.4 33.4 33.4 33.4 33.4 28.1 28.1 28.1 28.1 28.1 28.1 28.1 28.1 28.1 46.4 46.4 E"SAMrT 4Hlayer$ stream Analytical Results (part 2) 24 andCommunications School of Computing ๏ Also in this case MrT cannot ensure the desired coverage!
- 31. 4. Concluding Remarks andCommunications School of Computing
- 32. Concluding Remarks 26 ๏ Definition of a generic system model that can be easily adapted to practical scenarios and different viewpoints (ISP vs users). ๏ Derivation of the theoretical framework to assess user QoS ๏ Definition of efficient resource allocation frameworks, that can jointly optimise both system parameters and the error control strategy in use ๏ Development of efficient heuristic strategies that can derive good quality solutions in a finite number of steps. andCommunications School of Computing
- 33. Thank you for your attention andCommunications School of Computing For more information http://arxiv.org/abs/1411.5547 or http://goo.gl/Z4Y9YF A. Tassi, I. Chatzigeorgiou, and D. Vukobratović, “Resource Allocation Frameworks for Network-coded Layered Multimedia Multicast Services”, IEEE Journal on Selected Areas in Communications, Special Issue on “Fundamental Approaches to Network Coding in Wireless Communication Systems”, in press.
- 34. Bristol, 5th February 2015 R2D2: Network error control for Rapid and Reliable Data Delivery Project supported by EPSRC under the First Grant scheme (EP/L006251/1) On Optimization of Network-coded Scalable Multimedia Service Multicasting University of Bristol Andrea Tassi and Ioannis Chatzigeorgiou {a.tassi, i.chatzigeorgiou}@lancaster.ac.uk andCommunications School of Computing