Explicit and Implicit Temperature Constraints in Energy Harvesting Communications

Size: px
Start display at page:

Download "Explicit and Implicit Temperature Constraints in Energy Harvesting Communications"

Transcription

1 Explct and Implct Temperature Constrants n Energy Harvestng Communcatons Abdulrahman Baknna, Omur Ozel 2, and Sennur Ulukus Department of Electrcal and Computer Engneerng, Unversty of Maryland, College Park, MD 2 Department of Electrcal and Computer Engneerng, Carnege Mellon Unversty, Pttsburgh, PA Abstract We consder an energy harvestng communcaton system where the temperature dynamcs s governed by the transmsson power polcy. Dfferent from the prevous work, we consder a dscrete tme system where transmsson power s kept constant n each slot. We consder two models that capture dfferent effects of temperature. In the frst model, the temperature s constraned to be below a crtcal temperature at all tme nstants; we con ths model as explct temperature constrant model. We nvestgate throughput optmal power allocaton for multple energy arrvals under general, as well as temperature and energy lmted regmes. In the second model, we consder the effect of the temperature on the channel qualty; we con ths model as mplct temperature constrant model. As the dynamc range of the temperature changes sgnfcantly, the change n the thermal nose becomes non-neglgble, affectng the sgnal to nterference plus nose rato SINR. In effect, transmtted sgnals contrbute as nterference for all subsequent slots. In ths case, we nvestgate throughput optmal power allocaton under general, as well as low and hgh SINR regmes. I. INTRODUCTION We study two dfferent effects of temperature change on the optmal power allocaton, and hence, on the system performance of a sngle-user energy harvestng system. We consder explct and mplct temperature constraned systems. In the explct temperature constraned model, a peak temperature constrant prevents the system from overheatng. In the mplct temperature constraned model, the effect of the temperature on the channel qualty controls the system temperature. The schedulng-theoretc approach for energy harvestng communcatons was studed n varous settngs, see [] [5]. Prevous works consdered sngle-user channel [] [4], broadcast channel [5], multple access channel [6], [7], nterference channel [8], two-hop channel [9], [], two-way channel [], [2], and damond channel [3]. Temperature effects are studed n [4], [5]. In [4], a peak temperature constrant s consdered and the optmal contnuous power allocaton s studed. Although extensve nsghts and propertes of the optmal polcy were derved, only the sngle energy arrval case was fully solved. The fact that the power can take any contnuous functon makes the problem challengng, as the problem then becomes a functonal optmzaton problem and the optmal such functon should be obtaned. In ths paper, we consder the slotted verson of the temperature evoluton model consdered n [4], [5]. Ths work was supported by NSF Grants CNS , CCF 4-22, CCF , and CNS envronment temperature T e data queue E Tx Fg.. System model: the system heats up due to data transmsson. In the frst model we consder here, whch we con as the explct temperature constraned model, we consder an explct peak temperature constrant. Increasng the transmsson power ncreases the throughput and the temperature. Hgher temperature levels mean smaller admssble transmsson power levels for future slots. Ths model s the slotted verson of [4]. We study the optmal power allocaton for the multple energy arrval case, for whch we develop a generalzed waterfllng algorthm. Then, we study suffcent condtons under whch the system operates n the lmtng cases of non-bndng temperature or energy constrants. When the temperature constrant s not bndng, the problem reduces to the sngle-user energy harvestng channel studed n []. When the energy constrant s not bndng, the system effectvely becomes a temperature constraned system wth all the needed energy arrvng before the communcaton begns. In ths case, we show that the optmal powers are bounded and non-ncreasng, and the temperature of the system s non-decreasng. In the second model we consder here, whch we con as the mplct temperature constraned model, the temperature s not explctly constraned, however, the temperature affects the thermal nose whch affects the channel qualty. Ths arses when the dynamc range of the temperature s large, and s smlar to that presented n [6] but n a schedulng-theoretc settng. In ths case, the transmt powers used n prevous channel uses affect the thermal nose and therefore the channel qualty n future channel uses, and hence, the channel becomes a use dependent or acton dependent channel, see [7] [9]. Specfcally, each slot sees the prevous transmssons as nterference through a certan temperature flter. Ths flter arses naturally from dscretzng tme nto tme slots. For the general sgnal to nterference plus nose rato SINR, the problem s non-convex and s a sgnomal problem for whch we acheve a local optmal soluton usng the sngle N + Rx

2 condensaton method proposed n [2]. We then propose a heurstc algorthm whch mproves upon the local optmal soluton and may acheve the global optmal soluton. Then, we study the extreme settngs of low and hgh SINR regmes. We show that for low SINR regme t s optmal to transmt zero powers for all the slots except for the last, n whch all the harvested energy s transmtted. For the hgh SINR regme, the optmal soluton can be found usng geometrc programmng. II. SYSTEM MODEL We consder an energy harvestng communcaton system n whch the transmtter harvests energy Ẽ n the th slot, see Fg.. We consder the same temperature model consdered at [4], [5]. In ths model, the temperature, Tt, s defned by the followng dfferental equaton, dtt = apt btt T e dt where T e s the envronment temperature, Tt s the temperature at tme t, and a,b are non-negatve constants. Wth the ntal temperature T = T e, the soluton of s: t Tt = e bt e bτ apτdτ +T e 2 In what follows we assume that the duraton of each slot s equal to, whch can take any postve value. Let us defne T T as the temperature level by the end of the th slot, P as the power level used n the th slot. Usng 2, T can be expressed as: T = αt + +γ 3 where α = e b, = a b [ α] and γ = T e[ α]. The effect of n 3 appears through the constantsα,,γ. As the slot duraton ncreases, the values of, γ ncrease whle the value of α decreases; as the slot duraton ncreases, the temperature at the end of the slot becomes more dependent on the power transmtted wthn ths slot and less dependent on the ntal temperature at the begnnng of the slot. We now elmnate the prevous temperature readngs n T makng the temperature a functon of the powers only. We can do ths by recursvely substtutng by T n T to have T k = α k +T e 4 Ths formula shows that the temperature at the end of each slot depends on the power transmtted n ths slot and all prevous slots through an exponentally decayng temperature flter. We note that ths s the same formula that was developed n [6] n whch the slot duraton was assumed to be unty. III. EXPLICIT PEAK TEMPERATURE CONSTRAINT We now consder the model n whch we have an energy harvestng transmtter wth a peak temperature constrant. The nose varance s the same throughout the communcaton sesson and s set to σ 2. We consder a slotted system wth a constant power per slot. It follows from [4, equaton 47], that the temperature s monotone wthn the slot duraton. Hence, for the peak temperature constraned case, t suffces to constran the temperature only at the end of each slot; we begn the communcaton wth the system havng temperature T e. In ths case, the problem can be wrtten as, max {} T k T c, 2 log + P σ 2 Ẽ, k 5 where n the objectve functon and the energy constrant s the slot duraton. In what follows, wthout loss of generalty, we drop snce t s just a constant multpled n the objectve functon and we defne E = Ẽ. We rewrte problem 5 makng use of 4 as, max {} 2 log + P σ 2 α k T c T e, E, k 6 In the last slot, ether the temperature or the energy constrant has to be satsfed wth equalty. Otherwse, we can ncrease one of the powers untl one of the constrants s met wth equalty and ths strctly ncreases the objectve functon. Ths problem s a convex problem n the powers, whch can be solved optmally usng the KKTs and the Lagrangan s: L = log + P σ 2 + k α k=λ k T c T e + k E 7 k=µ Dfferentatng wth respect to and equatng to zero we get, = α D k= λ kα k + D k= µ σ 2 8 k In the optmal soluton, f nether constrant was tght n slot < D, then the power n slot + s strctly less than the power n slot. Ths follows from complementary slackness snce f at slot, f both constrants were not tght then we have λ = µ = whch, usng 8, mples that > +. To fnd the optmal soluton, we begn wth an ntal feasble power allocaton. If for ths power allocaton there exst non-negatve Lagrange multplers whch satsfy 8 and the complementary slackness condtons, then ths s the optmal power allocaton. Otherwse, these power allocatons should be modfed by pourng water away from the slots wth negatve Lagrange multplers to the slots wth hgher Lagrange multplers untl non-negatve Lagrange multplers are found and the complementary slackness condtons are satsfed. Snce ths problem s a convex optmzaton problem, any soluton for the KKTs acheve the global maxmum.

3 A. Energy Lmted Case In ths subsecton, we study a suffcent condton under whch the system becomes energy lmted. For all slots j n whch the followng s satsfed j E T c T e the temperature constrant cannot be tght. In partcular, when t s satsfed for k = D, then the temperature constrant can be completely removed from the system. To prove ths, we assume for the sake of contradcton that we have at slot j j E Tc Te whle the temperature constrant s tght, whch mples: T c T e = j α j < j 9 j E whch contradcts the assumpton j E Tc Te. The strct nequalty follows snce α <. The structure of the optmal soluton for ths case s studed n []. B. Temperature Lmted Case In ths part, we frst study a suffcent condton for problem 6 to be temperature lmted. The energy constrant s never tght f the followng condton s satsfed: T c T e k < E, k {,...,D} k For the temperature lmted case, an upper bound on the transmsson powers s equal to Tc Te. Hence, s suffcent to satsfy k < k E. In what follows, we study the structure of the optmal polcy for the temperature lmted case. In the last slot, the temperature constrant s satsfed wth equalty. The optmal powers are monotoncally decreasng n tme. The proof follows by contradcton. Assume for some ndex j that we have Pj < P j+. We now form another polcy, denoted as { }, whch has = P for all slots j,j +, whle we change the powers of slots j,j + to be P j = Pj + δ and P j+ = Pj+ δ for small enough δ. Ths δ always exsts as Pj < Pj+ mples that j k= αj k Pk < Tc Te. Snce the objectve functon s strctly concave, ths new polcy yelds a strctly hgher objectve functon, whch contradcts the optmalty of Pj < P j+. Now t remans to check that wth ths new polcy, the temperature constrant s stll feasble for any slot k j + whch follows from:, j,j+ < =, j,j+ α k P +α k j Pj +α k j Pj+ α k P +αk j P j +αk j P j+ 2 α k P < T c T e 3 Moreover, the optmal temperature levels are non-decreasng n tme. To prove ths, t suffces to show that: k+ α k P α k+ P, k = {,...,D } 4 We rewrte 4 as follows, α α k P Pk+, k = {,...,D } 5 Snce, we know that the last slot has to be satsfed wth equalty then we know D αd P = Tc Te. Hence, for the constrant at k = D we have: D α D P T c T e = α D P 6 whch can be wrtten as follows, D α α D P PD 7 whch proves 5 for k = D. Now assume for the sake of contradcton that 5 s false for k = D 2,.e.: D 2 PD < α Substtutng ths n 7, we get: α D 2 P 8 P D = αp D + αp D 9 D 2 < α α α D 2 P + αpd 2 D = α α D P PD 2 But snce we know that n the optmal polcy the powers are non-decreasng, ths s a contradcton and 5 holds for k = D 2. The same argument follows for any k < D 2. In optmal soluton, f the constrant s satsfed wth equalty for two consecutve slots then the power n the second slot must be equal to α Tc Te. To obtan ths, the two consecutve constrants whch are satsfed wth equalty are solved smultaneously for the power n the second slot. Hence, when the temperature hts the crtcal temperature, the optmal transmsson power n all the subsequent slots becomes constant and equal to α Tc Te. Ths follows snce the temperature s ncreasng, thus whenever the constrant becomes tght, t remans tght for all subsequent slots. We now conclude that the transmsson power s bounded as α T c T e T c T e, = {,...,D} 22 The lower bound follows from the dscusson above whle the upper bound follows from the feasblty of the constrants. We then proceed to fnd the optmal power allocatons. Snce the problem s convex, a necessary and suffcent condton s

4 to fnd a soluton satsfyng the KKTs. The optmal power s gven by settng {µ } = n 8. To ths end we present an approach to obtan the optmal powers. We use lne search to search over the tme slot at whch the temperature constrant becomes tght, whch we denote as. Then slots = { +,...,D} have power allocaton equal to α Tc Te, whle the power allocatons for slots = {,..., } are strctly decreasng and strctly hgher than α Tc Te. Hence, we ntalze = D and search for a soluton for the powers satsfyng the KKTs. If we obtan a soluton then we stop and ths s the optmal soluton. Otherwse, we decrease by one and repeat the search. IV. IMPLICIT TEMPERATURE CONSTRAINT We now consder the case when the dynamc range of the temperature ncreases. In ths case, we need to consder the change n the thermal nose of the system due to temperature changes. The thermal nose s lnearly proportonal to the temperature [2, Chapter ]. The objectve functon s: 2 log + ct +σ 2 23 where c s the proportonalty constant between the thermal nose and the temperature. In ths settng, the nose varance n each slot s determned by the value of the temperature at the begnnng of the slot. The maxmum temperature the system can reach s equal to T max D E + T e. Ths occurs when the transmtter transmts all ts energy arrvals n the last slot. The value of T max s useful n determnng the maxmum possble temperature on the system. As we show n the low SINR case later, the optmal power allocaton results n system temperature equal to T max. Usng 4 n 23, the problem can now be wrtten n terms of only transmsson powers as follows: max {} 2 log + c k= α k P k +T e +σ 2 E, P k, k {,...,D} 24 The problem n ths form hghlghts the effect of prevous transmssons on subsequent slots. The transmsson power at tme appears as an nterferng term at slot ndces greater than wth an exponentally decayng weght due to the flterng n the temperature. Ths problem s non-convex and determnng the global optmal soluton s generally a dffcult task. Next, we adapt the sgnomal programmng based teratve algorthm n [2] for the energy harvestng case. Ths algorthm provably converges to a local optmum pont. The problem n 24 can be wrtten n the followng equvalent sgnomal mnmzaton problem mn {} D c k= α k P k +T e +σ 2 c k= α k P k +T e +σ 2 + E, P k, k {,...,D} 25 The objectve functon n 25 s a sgnomal functon whch s a rato between two posynomals. Note also that the energy harvestng constrants n 25 are posynomals n p. In each teraton we approxmate the objectve by a posynomal. We do ths by approxmatng the posynomal n the denomnator by a monomal. Approprate choce of an approxmaton whch satsfes the condtons n [22] guarantees convergence to a local optmal soluton. Let us denote the posynomal n the th denomnator evaluated usng a power vector P by u P,.e., we have + u P vkp = c α k P k + +ct e +σ 2 26 k= k= where for k = {,..., } we have v k P = cα k P k, v P = and v + P = ct e +σ 2. Usng the arthmetc-geometrc mean nequalty we approxmate each posynomal by a monomal as follows: cα k θ k P θ k cte +σ 2 θ + u P k= θ k θ θ + 27 where + k= θ k = for all = {,...,D}. We now solve the prevous problem teratvely where we ntalze the power allocaton to any feasble power allocaton P. Then, we approxmate the posynomals u P usng the arthmetc-geometrc mean nequalty shown above. In each teraton j where the power allocaton s P j we choose θk as a functon of the posynomals and power allocaton as follows: θ k Pj = v k Pj u P j 28 whch satsfes + k= θ k Pj =. Ths choce of θ k Pj guarantees that the teratons converge to a KKT pont of the orgnal problem [22]. In partcular, for each teraton ths s a geometrc program and as requred by [22], ths can be transformed nto a convex problem; see also [2]. The above teratve approach converges to a local optmal soluton. Achevng the global optmal soluton s of exponental complexty. Alternatvely, to get to the optmal soluton, an approach ntroduced n [23] can be used. Ths approach solves the followng problem teratvely: mn {},t t OP t, t t α, E, P k 29 where OP s the objectve functon of 25 and α s chosen to be a number whch s slghtly more than and t can be ntalzed to be the soluton of problem 25 and then updated as the optmal solutons resultng from 29.

5 A. Low SINR Case The low SINR case occurs when the ncomng energes are very small wth respect to the nose varance. In ths case, an approxmaton to the logarthm functon n the objectve functon s the lnear functon. Hence, the objectve functon can be rewrtten as, whch s smlar to [24, equaton 4]: c 3 k= α k P k +T e +σ 2 In ths case, the optmal allocaton s zero for all tme slots, except n slot D n whch t s equal to P D = D E. Ths follows snce 3 can be upper bounded by D ct e +σ 2 E ct e +σ 2 3 and t s acheved by the specfed allocaton. Hence, a suffcent condton to have a low SINR regme s D E ct e +σ 2. Note also that the non-causal energy arrval knowledge s not necessary here as all energy arrvals are used only at the very last slot. The temperature at the end of the communcaton sesson s equal to T max = D E +T e. B. Hgh SINR Case When the values of c and σ are small, SINR s hgh and we approxmate the objectve functon by gnorng nsde the logarthm. In ths case, the problem n 24 has the Lagrangan: L = 2 log c + k k=µ k= α k P k +T e +σ 2 E 32 Takng the dervatve wth respect to gves, L = + j=+ and then equatng to zero gves: j=+ cα j c + j k= αj k P k +T e +σ 2 cα j c = j k= αj k P k +T e +σ 2 k= µ k 33 µ k 34 k= Although ths problem s non-convex, t s a geometrc program and we show next that any local optmal soluton for ths problem s also a global optmal soluton. To show ths we propose the followng equvalent problem: mn {x } 2 log c k= α k e x k +T e +σ 2 e x e x E, x R k {,...,D} 35 Ths equvalent problem s obtaned by substtutng = e x. Problem n 35 s a convex optmzaton problem snce the objectve s a convex functon n the form of a log-sumexponent and the constrant set s a convex set [25]. Hence, the KKTs are necessary and suffcent for global optmalty. We next show that any soluton for the KKTs of the orgnal non-convex problem also yelds a global optmal soluton. To prove ths, we frst wrte the Lagrangan of problem 35 as: L = 2 log c k= α k e x k +T e +σ 2 e x + ν k e x E 36 k= Takng the dervatve wth respect to x gves, L x = + j=+ cα j e x c j k= αj k e x k +Te +σ 2 D +e x ν k 37 k= whch provdes the followng necessary condton: e x + j=+ cα j c + j k= αj k e x k +Te +σ 2 ν k = k= 38 Usng the transformaton = logx and settng ν = µ, we observe that any soluton of 34 satsfes 38. Also, complementary slackness correspondng to 32 s satsfed f and only f t s satsfed by those for 35. Snce 35 s convex, any soluton satsfyng the KKTs s global optmal and through the transformaton = logx,µ = ν s also global optmal n the orgnal problem. The equvalent problem n 35 can be solved usng any convex optmzaton toolbox. V. NUMERICAL RESULTS We frst consder the explct peak temperature constraned model. As shown n Fg. 2, n general the power allocaton does not possess any monotoncty. The optmal power allocaton s close to the mnmum of the power allocaton of the energy and temperature lmted cases. We study the temperature lmted case n Fg. 3. When temperature s strctly ncreasng power s strctly decreasng. When the temperature reaches the crtcal temperature, the power remans constant. We then study the mplct temperature constraned model. For the general SINR case, we ntalze the sgnomal programmng problem usng a feasble power allocaton of = mn E n all slots. For the case shown n Fg. 4, the objectve functon takes the value.895 at the the global optmal and our experment verfes that the sngle condensaton method also yelds.895. In general, we observe that the sngle condensaton gves solutons close the global optmal. We then present the hgh SINR case n Fg. 5. In all cases we studed,

6 4 a =., b =.3, Tc =38, E = [2, 5, 3, 4, ] 3 a =.2, b =, E = [2,, 3, ] 3 2 P 2 optmal allocaton mnmum of ener. and temp. lmted P T 37.5 T Fg. 2. Smulaton for explct temperature constrant: general case. a =., b =.3, Tc = Fg. 4. Smulaton for mplct temperature constrant: general case. a =2, b =.3 E = [,3, 2,.5] 6.5 P 4 P T 37.5 T Fg. 3. Smulaton for explct temperature constrant: temperature lmted. we observe that the power allocaton s non-decreasng whch ponts to a monotoncty structure n the optmal allocaton. REFERENCES [] J. Yang and S. Ulukus. Optmal packet schedulng n an energy harvestng communcaton system. IEEE Trans. Comm., 6:22 23, January 22. [2] K. Tutuncuoglu and A. Yener. Optmum transmsson polces for battery lmted energy harvestng nodes. IEEE Trans. Wreless Comm., 3:8 89, March 22. [3] O. Ozel, K. Tutuncuoglu, J. Yang, S. Ulukus, and A. Yener. Transmsson wth energy harvestng nodes n fadng wreless channels: Optmal polces. IEEE JSAC, 298: , September 2. [4] C. K. Ho and R. Zhang. Optmal energy allocaton for wreless communcatons wth energy harvestng constrants. IEEE Trans. Sgnal Proc., 69: , September 22. [5] J. Yang, O. Ozel, and S. Ulukus. Broadcastng wth an energy harvestng rechargeable transmtter. IEEE Trans. Wreless Comm., 2:57 583, February 22. [6] J. Yang and S. Ulukus. Optmal packet schedulng n a multple access channel wth energy harvestng transmtters. Journal of Comm. and Networks, 42:4 5, Aprl 22. [7] Z. Wang, V. Aggarwal, and X. Wang. Iteratve dynamc water-fllng for fadng multple-access channels wth energy harvestng. IEEE JSAC, 333: , March 25. [8] K. Tutuncuoglu and A. Yener. Sum-rate optmal power polces for energy harvestng transmtters n an nterference channel. Journal of Comm. Networks, 42:5 6, Aprl 22. [9] D. Gunduz and B. Devllers. Two-hop communcaton wth energy harvestng. In IEEE CAMSAP, December 2. [] C. Huang, R. Zhang, and S. Cu. Throughput maxmzaton for the gaussan relay channel wth energy harvestng constrants. IEEE JSAC, 38: , August 23. [] B. Varan and A. Yener. Delay constraned energy harvestng networks wth lmted energy and data storage. IEEE JSAC, 345:55 564, May Fg. 5. Smulaton for mplct temperature constrant: hgh SINR case. [2] A. Arafa, A. Baknna, and S. Ulukus. Energy harvestng two-way channels wth decodng and processng costs. IEEE Trans. on Green Comm. and Networkng, :3 6, March 27. [3] B. Gurakan and S. Ulukus. Cooperatve damond channel wth energy harvestng nodes. IEEE JSAC, 345:64 67, May 26. [4] O. Ozel, S. Ulukus, and P. Grover. Energy harvestng transmtters that heat up: Throughput maxmzaton under temperature constrants. IEEE Trans. Wreless Comm., 58: , August 26. [5] A. Baknna, O. Ozel, and S. Ulukus. Energy harvestng communcatons under temperature constrants. In UCSD ITA, February 26. [6] T. Koch, A. Lapdoth, and P. P. Sotrads. Channels that heat up. IEEE Trans. Info. Theory, 558: , July 29. [7] D. Ward, N. Martns, and B. M. Sadler. Optmal remote estmaton over acton dependent swtchng channels: Managng workload and bas of a human operator. In ACC, July 25. [8] T. Wessman. Capacty of channels wth acton-dependent states. IEEE Trans. on Info. Theory, 56: , October 2. [9] B. Ahmad and O. Smeone. On channels wth acton-dependent states. In IEEE ISIT, July 22. [2] M. Chang, C. W. Tan, D. P. Palomar, D. Onell, and D. Julan. Power control by geometrc programmng. IEEE Trans. Wreless Comm., 67, July 27. [2] P. R. Gray and R. G. Meyer. Analyss and desgn of analog ntegrated crcuts. John Wley & Sons Inc., 99. [22] B. R. Marks and G. P. Wrght. A general nner approxmaton algorthm for nonconvex mathematcal programs. Operatons Research, 264:68 683, July/August 978. [23] M. Chang. Geometrc programmng for communcaton systems. Foundatons and Trends n Communcatons and Informaton Theory, 2 2: 54, 25. [24] S. Ulukus and L. J. Greensten. Throughput maxmzaton n CDMA uplnks usng adaptve spreadng and power control. In IEEE Sxth ISSTA, September 2. [25] S. P. Boyd and L. Vandenberghe. Convex Optmzaton. Cambrdge Unversty Press, 24.

Two-Way and Multiple-Access Energy Harvesting Systems with Energy Cooperation

Two-Way and Multiple-Access Energy Harvesting Systems with Energy Cooperation Two-Way and Multple-Access Energy Harvestng Systems wth Energy Cooperaton Berk Gurakan, Omur Ozel, Jng Yang 2, and Sennur Ulukus Department of Electrcal and Computer Engneerng, Unversty of Maryland, College

More information

ECE559VV Project Report

ECE559VV Project Report ECE559VV Project Report (Supplementary Notes Loc Xuan Bu I. MAX SUM-RATE SCHEDULING: THE UPLINK CASE We have seen (n the presentaton that, for downlnk (broadcast channels, the strategy maxmzng the sum-rate

More information

Energy Harvesting Multiple Access Channel with Peak Temperature Constraints

Energy Harvesting Multiple Access Channel with Peak Temperature Constraints Energy Harvesting Multiple Access Channel with Peak Temperature Constraints Abdulrahman Baknina, Omur Ozel 2, and Sennur Ulukus Department of Electrical and Computer Engineering, University of Maryland,

More information

Energy Harvesting Communications under Explicit and Implicit Temperature Constraints

Energy Harvesting Communications under Explicit and Implicit Temperature Constraints Energy Harvesting Communications under Explicit and Implicit Temperature Constraints Abdulrahman Baknina, Omur Ozel 2, and Sennur Ulukus Department of Electrical and Computer Engineering, University of

More information

MMA and GCMMA two methods for nonlinear optimization

MMA and GCMMA two methods for nonlinear optimization MMA and GCMMA two methods for nonlnear optmzaton Krster Svanberg Optmzaton and Systems Theory, KTH, Stockholm, Sweden. krlle@math.kth.se Ths note descrbes the algorthms used n the author s 2007 mplementatons

More information

princeton univ. F 17 cos 521: Advanced Algorithm Design Lecture 7: LP Duality Lecturer: Matt Weinberg

princeton univ. F 17 cos 521: Advanced Algorithm Design Lecture 7: LP Duality Lecturer: Matt Weinberg prnceton unv. F 17 cos 521: Advanced Algorthm Desgn Lecture 7: LP Dualty Lecturer: Matt Wenberg Scrbe: LP Dualty s an extremely useful tool for analyzng structural propertes of lnear programs. Whle there

More information

Assortment Optimization under MNL

Assortment Optimization under MNL Assortment Optmzaton under MNL Haotan Song Aprl 30, 2017 1 Introducton The assortment optmzaton problem ams to fnd the revenue-maxmzng assortment of products to offer when the prces of products are fxed.

More information

x = x 1 + :::+ x K and the nput covarance matrces are of the form ± = E[x x y ]. 3.2 Dualty Next, we ntroduce the concept of dualty wth the followng t

x = x 1 + :::+ x K and the nput covarance matrces are of the form ± = E[x x y ]. 3.2 Dualty Next, we ntroduce the concept of dualty wth the followng t Sum Power Iteratve Water-fllng for Mult-Antenna Gaussan Broadcast Channels N. Jndal, S. Jafar, S. Vshwanath and A. Goldsmth Dept. of Electrcal Engg. Stanford Unversty, CA, 94305 emal: njndal,syed,srram,andrea@wsl.stanford.edu

More information

4884 IEEE TRANSACTIONS ON COMMUNICATIONS, VOL. 61, NO. 12, DECEMBER Energy Cooperation in Energy Harvesting Communications

4884 IEEE TRANSACTIONS ON COMMUNICATIONS, VOL. 61, NO. 12, DECEMBER Energy Cooperation in Energy Harvesting Communications 4884 IEEE TRANSACTIONS ON COMMUNICATIONS, VOL. 6, NO., DECEMBER 3 Energy Cooperaton n Energy Harvestng Communcatons Ber Guraan, Student Member, IEEE, Omur Ozel, Student Member, IEEE, Jng Yang, Member,

More information

Problem Set 9 Solutions

Problem Set 9 Solutions Desgn and Analyss of Algorthms May 4, 2015 Massachusetts Insttute of Technology 6.046J/18.410J Profs. Erk Demane, Srn Devadas, and Nancy Lynch Problem Set 9 Solutons Problem Set 9 Solutons Ths problem

More information

Welfare Properties of General Equilibrium. What can be said about optimality properties of resource allocation implied by general equilibrium?

Welfare Properties of General Equilibrium. What can be said about optimality properties of resource allocation implied by general equilibrium? APPLIED WELFARE ECONOMICS AND POLICY ANALYSIS Welfare Propertes of General Equlbrum What can be sad about optmalty propertes of resource allocaton mpled by general equlbrum? Any crteron used to compare

More information

College of Computer & Information Science Fall 2009 Northeastern University 20 October 2009

College of Computer & Information Science Fall 2009 Northeastern University 20 October 2009 College of Computer & Informaton Scence Fall 2009 Northeastern Unversty 20 October 2009 CS7880: Algorthmc Power Tools Scrbe: Jan Wen and Laura Poplawsk Lecture Outlne: Prmal-dual schema Network Desgn:

More information

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur Module 3 LOSSY IMAGE COMPRESSION SYSTEMS Verson ECE IIT, Kharagpur Lesson 6 Theory of Quantzaton Verson ECE IIT, Kharagpur Instructonal Objectves At the end of ths lesson, the students should be able to:

More information

Solutions HW #2. minimize. Ax = b. Give the dual problem, and make the implicit equality constraints explicit. Solution.

Solutions HW #2. minimize. Ax = b. Give the dual problem, and make the implicit equality constraints explicit. Solution. Solutons HW #2 Dual of general LP. Fnd the dual functon of the LP mnmze subject to c T x Gx h Ax = b. Gve the dual problem, and make the mplct equalty constrants explct. Soluton. 1. The Lagrangan s L(x,

More information

The Minimum Universal Cost Flow in an Infeasible Flow Network

The Minimum Universal Cost Flow in an Infeasible Flow Network Journal of Scences, Islamc Republc of Iran 17(2): 175-180 (2006) Unversty of Tehran, ISSN 1016-1104 http://jscencesutacr The Mnmum Unversal Cost Flow n an Infeasble Flow Network H Saleh Fathabad * M Bagheran

More information

IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, VOL. 33, NO. 3, MARCH

IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, VOL. 33, NO. 3, MARCH IEEE JOURAL O SELECTED AREAS I COMMUICATIOS, VOL. 33, O. 3, MARCH 205 467 Optmum Polces for an Energy Harvestng Transmtter Under Energy Storage Losses Kaya Tutuncuoglu, Student Member, IEEE, Ayln Yener,

More information

Which Separator? Spring 1

Which Separator? Spring 1 Whch Separator? 6.034 - Sprng 1 Whch Separator? Mamze the margn to closest ponts 6.034 - Sprng Whch Separator? Mamze the margn to closest ponts 6.034 - Sprng 3 Margn of a pont " # y (w $ + b) proportonal

More information

Lagrange Multipliers Kernel Trick

Lagrange Multipliers Kernel Trick Lagrange Multplers Kernel Trck Ncholas Ruozz Unversty of Texas at Dallas Based roughly on the sldes of Davd Sontag General Optmzaton A mathematcal detour, we ll come back to SVMs soon! subject to: f x

More information

Resource Allocation with a Budget Constraint for Computing Independent Tasks in the Cloud

Resource Allocation with a Budget Constraint for Computing Independent Tasks in the Cloud Resource Allocaton wth a Budget Constrant for Computng Independent Tasks n the Cloud Wemng Sh and Bo Hong School of Electrcal and Computer Engneerng Georga Insttute of Technology, USA 2nd IEEE Internatonal

More information

Maximizing the number of nonnegative subsets

Maximizing the number of nonnegative subsets Maxmzng the number of nonnegatve subsets Noga Alon Hao Huang December 1, 213 Abstract Gven a set of n real numbers, f the sum of elements of every subset of sze larger than k s negatve, what s the maxmum

More information

Lecture Notes on Linear Regression

Lecture Notes on Linear Regression Lecture Notes on Lnear Regresson Feng L fl@sdueducn Shandong Unversty, Chna Lnear Regresson Problem In regresson problem, we am at predct a contnuous target value gven an nput feature vector We assume

More information

Economics 101. Lecture 4 - Equilibrium and Efficiency

Economics 101. Lecture 4 - Equilibrium and Efficiency Economcs 0 Lecture 4 - Equlbrum and Effcency Intro As dscussed n the prevous lecture, we wll now move from an envronment where we looed at consumers mang decsons n solaton to analyzng economes full of

More information

Some modelling aspects for the Matlab implementation of MMA

Some modelling aspects for the Matlab implementation of MMA Some modellng aspects for the Matlab mplementaton of MMA Krster Svanberg krlle@math.kth.se Optmzaton and Systems Theory Department of Mathematcs KTH, SE 10044 Stockholm September 2004 1. Consdered optmzaton

More information

Module 9. Lecture 6. Duality in Assignment Problems

Module 9. Lecture 6. Duality in Assignment Problems Module 9 1 Lecture 6 Dualty n Assgnment Problems In ths lecture we attempt to answer few other mportant questons posed n earler lecture for (AP) and see how some of them can be explaned through the concept

More information

Closed form solutions for water-filling problems in optimization and game frameworks

Closed form solutions for water-filling problems in optimization and game frameworks Closed form solutons for water-fllng problems n optmzaton and game frameworks Etan Altman INRIA BP93 24 route des Lucoles 692 Sopha Antpols FRANCE altman@sopha.nra.fr Konstantn Avrachenkov INRIA BP93 24

More information

Joint Scheduling and Resource Allocation in CDMA Systems

Joint Scheduling and Resource Allocation in CDMA Systems TO APPEAR IEEE TRANSACTIONS ON INFORMATION THEORY Jont Schedulng and Resource Allocaton n CDMA Systems Vjay G. Subramanan, Randall A. Berry, and Rajeev Agrawal Abstract In ths paper, the schedulng and

More information

An Upper Bound on SINR Threshold for Call Admission Control in Multiple-Class CDMA Systems with Imperfect Power-Control

An Upper Bound on SINR Threshold for Call Admission Control in Multiple-Class CDMA Systems with Imperfect Power-Control An Upper Bound on SINR Threshold for Call Admsson Control n Multple-Class CDMA Systems wth Imperfect ower-control Mahmoud El-Sayes MacDonald, Dettwler and Assocates td. (MDA) Toronto, Canada melsayes@hotmal.com

More information

Dynamic Programming. Preview. Dynamic Programming. Dynamic Programming. Dynamic Programming (Example: Fibonacci Sequence)

Dynamic Programming. Preview. Dynamic Programming. Dynamic Programming. Dynamic Programming (Example: Fibonacci Sequence) /24/27 Prevew Fbonacc Sequence Longest Common Subsequence Dynamc programmng s a method for solvng complex problems by breakng them down nto smpler sub-problems. It s applcable to problems exhbtng the propertes

More information

Energy-Efficient Power Allocation for M2M Communications with Energy Harvesting Transmitter

Energy-Efficient Power Allocation for M2M Communications with Energy Harvesting Transmitter nergy-ffcent ower Allocaton for M2M Communcatons wth nergy Harvestng Transmtter Derrck Wng Kwan Ng and Robert Schober Char of Dgtal Communcatons, The Unverstät rlangen-nürnberg Abstract In ths paper, power

More information

Joint Scheduling and Resource Allocation in CDMA Systems

Joint Scheduling and Resource Allocation in CDMA Systems TECHNICAL REPORT - JUNE 2009 1 Jont Schedulng and Resource Allocaton n CDMA Systems Vjay G. Subramanan, Randall A. Berry, and Rajeev Agrawal Expanded Techncal Report: A shorter verson of ths paper wll

More information

2E Pattern Recognition Solutions to Introduction to Pattern Recognition, Chapter 2: Bayesian pattern classification

2E Pattern Recognition Solutions to Introduction to Pattern Recognition, Chapter 2: Bayesian pattern classification E395 - Pattern Recognton Solutons to Introducton to Pattern Recognton, Chapter : Bayesan pattern classfcaton Preface Ths document s a soluton manual for selected exercses from Introducton to Pattern Recognton

More information

Appendix for Causal Interaction in Factorial Experiments: Application to Conjoint Analysis

Appendix for Causal Interaction in Factorial Experiments: Application to Conjoint Analysis A Appendx for Causal Interacton n Factoral Experments: Applcaton to Conjont Analyss Mathematcal Appendx: Proofs of Theorems A. Lemmas Below, we descrbe all the lemmas, whch are used to prove the man theorems

More information

Power Allocation/Beamforming for DF MIMO Two-Way Relaying: Relay and Network Optimization

Power Allocation/Beamforming for DF MIMO Two-Way Relaying: Relay and Network Optimization Power Allocaton/Beamformng for DF MIMO Two-Way Relayng: Relay and Network Optmzaton Je Gao, Janshu Zhang, Sergy A. Vorobyov, Ha Jang, and Martn Haardt Dept. of Electrcal & Computer Engneerng, Unversty

More information

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems Numercal Analyss by Dr. Anta Pal Assstant Professor Department of Mathematcs Natonal Insttute of Technology Durgapur Durgapur-713209 emal: anta.bue@gmal.com 1 . Chapter 5 Soluton of System of Lnear Equatons

More information

Lectures - Week 4 Matrix norms, Conditioning, Vector Spaces, Linear Independence, Spanning sets and Basis, Null space and Range of a Matrix

Lectures - Week 4 Matrix norms, Conditioning, Vector Spaces, Linear Independence, Spanning sets and Basis, Null space and Range of a Matrix Lectures - Week 4 Matrx norms, Condtonng, Vector Spaces, Lnear Independence, Spannng sets and Bass, Null space and Range of a Matrx Matrx Norms Now we turn to assocatng a number to each matrx. We could

More information

Additional Codes using Finite Difference Method. 1 HJB Equation for Consumption-Saving Problem Without Uncertainty

Additional Codes using Finite Difference Method. 1 HJB Equation for Consumption-Saving Problem Without Uncertainty Addtonal Codes usng Fnte Dfference Method Benamn Moll 1 HJB Equaton for Consumpton-Savng Problem Wthout Uncertanty Before consderng the case wth stochastc ncome n http://www.prnceton.edu/~moll/ HACTproect/HACT_Numercal_Appendx.pdf,

More information

PHYS 705: Classical Mechanics. Calculus of Variations II

PHYS 705: Classical Mechanics. Calculus of Variations II 1 PHYS 705: Classcal Mechancs Calculus of Varatons II 2 Calculus of Varatons: Generalzaton (no constrant yet) Suppose now that F depends on several dependent varables : We need to fnd such that has a statonary

More information

Simultaneous Optimization of Berth Allocation, Quay Crane Assignment and Quay Crane Scheduling Problems in Container Terminals

Simultaneous Optimization of Berth Allocation, Quay Crane Assignment and Quay Crane Scheduling Problems in Container Terminals Smultaneous Optmzaton of Berth Allocaton, Quay Crane Assgnment and Quay Crane Schedulng Problems n Contaner Termnals Necat Aras, Yavuz Türkoğulları, Z. Caner Taşkın, Kuban Altınel Abstract In ths work,

More information

n α j x j = 0 j=1 has a nontrivial solution. Here A is the n k matrix whose jth column is the vector for all t j=0

n α j x j = 0 j=1 has a nontrivial solution. Here A is the n k matrix whose jth column is the vector for all t j=0 MODULE 2 Topcs: Lnear ndependence, bass and dmenson We have seen that f n a set of vectors one vector s a lnear combnaton of the remanng vectors n the set then the span of the set s unchanged f that vector

More information

Lecture 10 Support Vector Machines II

Lecture 10 Support Vector Machines II Lecture 10 Support Vector Machnes II 22 February 2016 Taylor B. Arnold Yale Statstcs STAT 365/665 1/28 Notes: Problem 3 s posted and due ths upcomng Frday There was an early bug n the fake-test data; fxed

More information

Joint Scheduling and Resource Allocation in CDMA Systems

Joint Scheduling and Resource Allocation in CDMA Systems ACCEPTED TO IEEE TRANSACTIONS ON INFORMATION THEORY 1 Jont Schedulng and Resource Allocaton n CDMA Systems Vjay G. Subramanan, Randall A. Berry, and Rajeev Agrawal Abstract We consder schedulng and resource

More information

State Amplification and State Masking for the Binary Energy Harvesting Channel

State Amplification and State Masking for the Binary Energy Harvesting Channel State Amplfcaton and State Maskng for the Bnary Energy Harvestng Channel Kaya Tutuncuoglu, Omur Ozel 2, Ayln Yener, and Sennur Ulukus 2 Department of Electrcal Engneerng, The Pennsylvana State Unversty,

More information

Lecture 21: Numerical methods for pricing American type derivatives

Lecture 21: Numerical methods for pricing American type derivatives Lecture 21: Numercal methods for prcng Amercan type dervatves Xaoguang Wang STAT 598W Aprl 10th, 2014 (STAT 598W) Lecture 21 1 / 26 Outlne 1 Fnte Dfference Method Explct Method Penalty Method (STAT 598W)

More information

More metrics on cartesian products

More metrics on cartesian products More metrcs on cartesan products If (X, d ) are metrc spaces for 1 n, then n Secton II4 of the lecture notes we defned three metrcs on X whose underlyng topologes are the product topology The purpose of

More information

Chapter Newton s Method

Chapter Newton s Method Chapter 9. Newton s Method After readng ths chapter, you should be able to:. Understand how Newton s method s dfferent from the Golden Secton Search method. Understand how Newton s method works 3. Solve

More information

Minimisation of the Average Response Time in a Cluster of Servers

Minimisation of the Average Response Time in a Cluster of Servers Mnmsaton of the Average Response Tme n a Cluster of Servers Valery Naumov Abstract: In ths paper, we consder task assgnment problem n a cluster of servers. We show that optmal statc task assgnment s tantamount

More information

Solutions to exam in SF1811 Optimization, Jan 14, 2015

Solutions to exam in SF1811 Optimization, Jan 14, 2015 Solutons to exam n SF8 Optmzaton, Jan 4, 25 3 3 O------O -4 \ / \ / The network: \/ where all lnks go from left to rght. /\ / \ / \ 6 O------O -5 2 4.(a) Let x = ( x 3, x 4, x 23, x 24 ) T, where the varable

More information

Joint Scheduling and Power-Allocation for Interference Management in Wireless Networks

Joint Scheduling and Power-Allocation for Interference Management in Wireless Networks Jont Schedulng and Power-Allocaton for Interference Management n Wreless Networks Xn Lu *, Edwn K. P. Chong, and Ness B. Shroff * * School of Electrcal and Computer Engneerng Purdue Unversty West Lafayette,

More information

Managing Capacity Through Reward Programs. on-line companion page. Byung-Do Kim Seoul National University College of Business Administration

Managing Capacity Through Reward Programs. on-line companion page. Byung-Do Kim Seoul National University College of Business Administration Managng Caacty Through eward Programs on-lne comanon age Byung-Do Km Seoul Natonal Unversty College of Busness Admnstraton Mengze Sh Unversty of Toronto otman School of Management Toronto ON M5S E6 Canada

More information

Joint Scheduling and Resource Allocation in CDMA Systems

Joint Scheduling and Resource Allocation in CDMA Systems SUBMITTED TO IEEE TRANSACTIONS ON INFORMATION THEORY 1 Jont Schedulng and Resource Allocaton n CDMA Systems Vjay Subramanan, Randall A. Berry, and Rajeev Agrawal Abstract We consder schedulng and resource

More information

Perfect Competition and the Nash Bargaining Solution

Perfect Competition and the Nash Bargaining Solution Perfect Competton and the Nash Barganng Soluton Renhard John Department of Economcs Unversty of Bonn Adenauerallee 24-42 53113 Bonn, Germany emal: rohn@un-bonn.de May 2005 Abstract For a lnear exchange

More information

Single-Facility Scheduling over Long Time Horizons by Logic-based Benders Decomposition

Single-Facility Scheduling over Long Time Horizons by Logic-based Benders Decomposition Sngle-Faclty Schedulng over Long Tme Horzons by Logc-based Benders Decomposton Elvn Coban and J. N. Hooker Tepper School of Busness, Carnege Mellon Unversty ecoban@andrew.cmu.edu, john@hooker.tepper.cmu.edu

More information

On the Multicriteria Integer Network Flow Problem

On the Multicriteria Integer Network Flow Problem BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 5, No 2 Sofa 2005 On the Multcrtera Integer Network Flow Problem Vassl Vasslev, Marana Nkolova, Maryana Vassleva Insttute of

More information

Chapter - 2. Distribution System Power Flow Analysis

Chapter - 2. Distribution System Power Flow Analysis Chapter - 2 Dstrbuton System Power Flow Analyss CHAPTER - 2 Radal Dstrbuton System Load Flow 2.1 Introducton Load flow s an mportant tool [66] for analyzng electrcal power system network performance. Load

More information

EEL 6266 Power System Operation and Control. Chapter 3 Economic Dispatch Using Dynamic Programming

EEL 6266 Power System Operation and Control. Chapter 3 Economic Dispatch Using Dynamic Programming EEL 6266 Power System Operaton and Control Chapter 3 Economc Dspatch Usng Dynamc Programmng Pecewse Lnear Cost Functons Common practce many utltes prefer to represent ther generator cost functons as sngle-

More information

Distributed Non-Autonomous Power Control through Distributed Convex Optimization

Distributed Non-Autonomous Power Control through Distributed Convex Optimization Dstrbuted Non-Autonomous Power Control through Dstrbuted Convex Optmzaton S. Sundhar Ram and V. V. Veeravall ECE Department and Coordnated Scence Lab Unversty of Illnos at Urbana-Champagn Emal: {ssrnv5,vvv}@llnos.edu

More information

6.854J / J Advanced Algorithms Fall 2008

6.854J / J Advanced Algorithms Fall 2008 MIT OpenCourseWare http://ocw.mt.edu 6.854J / 18.415J Advanced Algorthms Fall 2008 For nformaton about ctng these materals or our Terms of Use, vst: http://ocw.mt.edu/terms. 18.415/6.854 Advanced Algorthms

More information

ON A DETERMINATION OF THE INITIAL FUNCTIONS FROM THE OBSERVED VALUES OF THE BOUNDARY FUNCTIONS FOR THE SECOND-ORDER HYPERBOLIC EQUATION

ON A DETERMINATION OF THE INITIAL FUNCTIONS FROM THE OBSERVED VALUES OF THE BOUNDARY FUNCTIONS FOR THE SECOND-ORDER HYPERBOLIC EQUATION Advanced Mathematcal Models & Applcatons Vol.3, No.3, 2018, pp.215-222 ON A DETERMINATION OF THE INITIAL FUNCTIONS FROM THE OBSERVED VALUES OF THE BOUNDARY FUNCTIONS FOR THE SECOND-ORDER HYPERBOLIC EUATION

More information

PROBLEM SET 7 GENERAL EQUILIBRIUM

PROBLEM SET 7 GENERAL EQUILIBRIUM PROBLEM SET 7 GENERAL EQUILIBRIUM Queston a Defnton: An Arrow-Debreu Compettve Equlbrum s a vector of prces {p t } and allocatons {c t, c 2 t } whch satsfes ( Gven {p t }, c t maxmzes βt ln c t subject

More information

U.C. Berkeley CS294: Beyond Worst-Case Analysis Luca Trevisan September 5, 2017

U.C. Berkeley CS294: Beyond Worst-Case Analysis Luca Trevisan September 5, 2017 U.C. Berkeley CS94: Beyond Worst-Case Analyss Handout 4s Luca Trevsan September 5, 07 Summary of Lecture 4 In whch we ntroduce semdefnte programmng and apply t to Max Cut. Semdefnte Programmng Recall that

More information

VQ widely used in coding speech, image, and video

VQ widely used in coding speech, image, and video at Scalar quantzers are specal cases of vector quantzers (VQ): they are constraned to look at one sample at a tme (memoryless) VQ does not have such constrant better RD perfomance expected Source codng

More information

CSci 6974 and ECSE 6966 Math. Tech. for Vision, Graphics and Robotics Lecture 21, April 17, 2006 Estimating A Plane Homography

CSci 6974 and ECSE 6966 Math. Tech. for Vision, Graphics and Robotics Lecture 21, April 17, 2006 Estimating A Plane Homography CSc 6974 and ECSE 6966 Math. Tech. for Vson, Graphcs and Robotcs Lecture 21, Aprl 17, 2006 Estmatng A Plane Homography Overvew We contnue wth a dscusson of the major ssues, usng estmaton of plane projectve

More information

Yong Joon Ryang. 1. Introduction Consider the multicommodity transportation problem with convex quadratic cost function. 1 2 (x x0 ) T Q(x x 0 )

Yong Joon Ryang. 1. Introduction Consider the multicommodity transportation problem with convex quadratic cost function. 1 2 (x x0 ) T Q(x x 0 ) Kangweon-Kyungk Math. Jour. 4 1996), No. 1, pp. 7 16 AN ITERATIVE ROW-ACTION METHOD FOR MULTICOMMODITY TRANSPORTATION PROBLEMS Yong Joon Ryang Abstract. The optmzaton problems wth quadratc constrants often

More information

University of Alberta. Library Release Form. Title of Thesis: Joint Bandwidth and Power Allocation in Wireless Communication Networks

University of Alberta. Library Release Form. Title of Thesis: Joint Bandwidth and Power Allocation in Wireless Communication Networks Unversty of Alberta Lbrary Release Form Name of Author: Xaowen Gong Ttle of Thess: Jont Bandwdth and Power Allocaton n Wreless Communcaton Networks Degree: Master of Scence Year ths Degree Granted: 2010

More information

Hila Etzion. Min-Seok Pang

Hila Etzion. Min-Seok Pang RESERCH RTICLE COPLEENTRY ONLINE SERVICES IN COPETITIVE RKETS: INTINING PROFITILITY IN THE PRESENCE OF NETWORK EFFECTS Hla Etzon Department of Technology and Operatons, Stephen. Ross School of usness,

More information

The Second Anti-Mathima on Game Theory

The Second Anti-Mathima on Game Theory The Second Ant-Mathma on Game Theory Ath. Kehagas December 1 2006 1 Introducton In ths note we wll examne the noton of game equlbrum for three types of games 1. 2-player 2-acton zero-sum games 2. 2-player

More information

Numerical Heat and Mass Transfer

Numerical Heat and Mass Transfer Master degree n Mechancal Engneerng Numercal Heat and Mass Transfer 06-Fnte-Dfference Method (One-dmensonal, steady state heat conducton) Fausto Arpno f.arpno@uncas.t Introducton Why we use models and

More information

Maximal Margin Classifier

Maximal Margin Classifier CS81B/Stat41B: Advanced Topcs n Learnng & Decson Makng Mamal Margn Classfer Lecturer: Mchael Jordan Scrbes: Jana van Greunen Corrected verson - /1/004 1 References/Recommended Readng 1.1 Webstes www.kernel-machnes.org

More information

Lecture 10 Support Vector Machines. Oct

Lecture 10 Support Vector Machines. Oct Lecture 10 Support Vector Machnes Oct - 20-2008 Lnear Separators Whch of the lnear separators s optmal? Concept of Margn Recall that n Perceptron, we learned that the convergence rate of the Perceptron

More information

Downlink Throughput Maximization: TDMA versus CDMA

Downlink Throughput Maximization: TDMA versus CDMA 00 Conference on Informaton Scences and Systems, The Johns Hopkns Unversty, March, 00 Downlnk Throughput Maxmzaton: TDMA versus CDMA Changyoon Oh and Ayln Yener Department of Electrcal Engneerng The Pennsylvana

More information

Energy Efficient Resource Allocation in Machine-to-Machine Communications with Multiple Access and Energy Harvesting for IoT

Energy Efficient Resource Allocation in Machine-to-Machine Communications with Multiple Access and Energy Harvesting for IoT 1 Energy Effcent Resource Allocaton n Machne-to-Machne Communcatons wth Multple Access and Energy Harvestng for IoT Zhaohu Yang, We Xu, Senor Member, IEEE, Yjn Pan, Cunhua Pan, and Mng Chen arxv:1711.10776v1

More information

Kernel Methods and SVMs Extension

Kernel Methods and SVMs Extension Kernel Methods and SVMs Extenson The purpose of ths document s to revew materal covered n Machne Learnng 1 Supervsed Learnng regardng support vector machnes (SVMs). Ths document also provdes a general

More information

Suggested solutions for the exam in SF2863 Systems Engineering. June 12,

Suggested solutions for the exam in SF2863 Systems Engineering. June 12, Suggested solutons for the exam n SF2863 Systems Engneerng. June 12, 2012 14.00 19.00 Examner: Per Enqvst, phone: 790 62 98 1. We can thnk of the farm as a Jackson network. The strawberry feld s modelled

More information

CHAPTER 5 NUMERICAL EVALUATION OF DYNAMIC RESPONSE

CHAPTER 5 NUMERICAL EVALUATION OF DYNAMIC RESPONSE CHAPTER 5 NUMERICAL EVALUATION OF DYNAMIC RESPONSE Analytcal soluton s usually not possble when exctaton vares arbtrarly wth tme or f the system s nonlnear. Such problems can be solved by numercal tmesteppng

More information

Feature Selection: Part 1

Feature Selection: Part 1 CSE 546: Machne Learnng Lecture 5 Feature Selecton: Part 1 Instructor: Sham Kakade 1 Regresson n the hgh dmensonal settng How do we learn when the number of features d s greater than the sample sze n?

More information

Outline. Communication. Bellman Ford Algorithm. Bellman Ford Example. Bellman Ford Shortest Path [1]

Outline. Communication. Bellman Ford Algorithm. Bellman Ford Example. Bellman Ford Shortest Path [1] DYNAMIC SHORTEST PATH SEARCH AND SYNCHRONIZED TASK SWITCHING Jay Wagenpfel, Adran Trachte 2 Outlne Shortest Communcaton Path Searchng Bellmann Ford algorthm Algorthm for dynamc case Modfcatons to our algorthm

More information

An Interactive Optimisation Tool for Allocation Problems

An Interactive Optimisation Tool for Allocation Problems An Interactve Optmsaton ool for Allocaton Problems Fredr Bonäs, Joam Westerlund and apo Westerlund Process Desgn Laboratory, Faculty of echnology, Åbo Aadem Unversty, uru 20500, Fnland hs paper presents

More information

Hidden Markov Models

Hidden Markov Models Hdden Markov Models Namrata Vaswan, Iowa State Unversty Aprl 24, 204 Hdden Markov Model Defntons and Examples Defntons:. A hdden Markov model (HMM) refers to a set of hdden states X 0, X,..., X t,...,

More information

Errors for Linear Systems

Errors for Linear Systems Errors for Lnear Systems When we solve a lnear system Ax b we often do not know A and b exactly, but have only approxmatons  and ˆb avalable. Then the best thng we can do s to solve ˆx ˆb exactly whch

More information

The Expectation-Maximization Algorithm

The Expectation-Maximization Algorithm The Expectaton-Maxmaton Algorthm Charles Elan elan@cs.ucsd.edu November 16, 2007 Ths chapter explans the EM algorthm at multple levels of generalty. Secton 1 gves the standard hgh-level verson of the algorthm.

More information

Temperature. Chapter Heat Engine

Temperature. Chapter Heat Engine Chapter 3 Temperature In prevous chapters of these notes we ntroduced the Prncple of Maxmum ntropy as a technque for estmatng probablty dstrbutons consstent wth constrants. In Chapter 9 we dscussed the

More information

Pricing and Resource Allocation Game Theoretic Models

Pricing and Resource Allocation Game Theoretic Models Prcng and Resource Allocaton Game Theoretc Models Zhy Huang Changbn Lu Q Zhang Computer and Informaton Scence December 8, 2009 Z. Huang, C. Lu, and Q. Zhang (CIS) Game Theoretc Models December 8, 2009

More information

Portfolios with Trading Constraints and Payout Restrictions

Portfolios with Trading Constraints and Payout Restrictions Portfolos wth Tradng Constrants and Payout Restrctons John R. Brge Northwestern Unversty (ont wor wth Chrs Donohue Xaodong Xu and Gongyun Zhao) 1 General Problem (Very) long-term nvestor (eample: unversty

More information

Generalized Linear Methods

Generalized Linear Methods Generalzed Lnear Methods 1 Introducton In the Ensemble Methods the general dea s that usng a combnaton of several weak learner one could make a better learner. More formally, assume that we have a set

More information

Outline and Reading. Dynamic Programming. Dynamic Programming revealed. Computing Fibonacci. The General Dynamic Programming Technique

Outline and Reading. Dynamic Programming. Dynamic Programming revealed. Computing Fibonacci. The General Dynamic Programming Technique Outlne and Readng Dynamc Programmng The General Technque ( 5.3.2) -1 Knapsac Problem ( 5.3.3) Matrx Chan-Product ( 5.3.1) Dynamc Programmng verson 1.4 1 Dynamc Programmng verson 1.4 2 Dynamc Programmng

More information

14 Lagrange Multipliers

14 Lagrange Multipliers Lagrange Multplers 14 Lagrange Multplers The Method of Lagrange Multplers s a powerful technque for constraned optmzaton. Whle t has applcatons far beyond machne learnng t was orgnally developed to solve

More information

Sum Throughput Maximization for Heterogeneous Multicell Networks with RF-Powered Relays

Sum Throughput Maximization for Heterogeneous Multicell Networks with RF-Powered Relays Sum Throughput Maxmzaton for Heterogeneous Multcell etworks wth RF-Powered Relays Al A. asr, Duy T. go, Xangyun Zhou, Rodney A. Kennedy, and Salman Durran Research School of Engneerng, Australan atonal

More information

Foundations of Arithmetic

Foundations of Arithmetic Foundatons of Arthmetc Notaton We shall denote the sum and product of numbers n the usual notaton as a 2 + a 2 + a 3 + + a = a, a 1 a 2 a 3 a = a The notaton a b means a dvdes b,.e. ac = b where c s an

More information

First day August 1, Problems and Solutions

First day August 1, Problems and Solutions FOURTH INTERNATIONAL COMPETITION FOR UNIVERSITY STUDENTS IN MATHEMATICS July 30 August 4, 997, Plovdv, BULGARIA Frst day August, 997 Problems and Solutons Problem. Let {ε n } n= be a sequence of postve

More information

How Bad Is Suboptimal Rate Allocation?

How Bad Is Suboptimal Rate Allocation? How Bad Is Suboptmal Rate Allocaton? Tan Lan, Xaojun Ln 2, Mung Chang, Ruby Lee Department of Electrcal Engneerng, Prnceton Unversty, NJ 08544, USA 2 School of Electrcal and Computer Engneerng, Purdue

More information

Optimal Bandwidth and Power Allocation for

Optimal Bandwidth and Power Allocation for Optmal Bandwdth and Power Allocaton for 1 Sum Ergodc Capacty under Fadng Channels n Cogntve Rado Networks arxv:1006.5061v1 [cs.it] 25 Jun 2010 Xaowen Gong, Student Member, IEEE, Sergy A. Vorobyov, Senor

More information

Real-Time Systems. Multiprocessor scheduling. Multiprocessor scheduling. Multiprocessor scheduling

Real-Time Systems. Multiprocessor scheduling. Multiprocessor scheduling. Multiprocessor scheduling Real-Tme Systems Multprocessor schedulng Specfcaton Implementaton Verfcaton Multprocessor schedulng -- -- Global schedulng How are tasks assgned to processors? Statc assgnment The processor(s) used for

More information

Winter 2008 CS567 Stochastic Linear/Integer Programming Guest Lecturer: Xu, Huan

Winter 2008 CS567 Stochastic Linear/Integer Programming Guest Lecturer: Xu, Huan Wnter 2008 CS567 Stochastc Lnear/Integer Programmng Guest Lecturer: Xu, Huan Class 2: More Modelng Examples 1 Capacty Expanson Capacty expanson models optmal choces of the tmng and levels of nvestments

More information

COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS

COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS Avalable onlne at http://sck.org J. Math. Comput. Sc. 3 (3), No., 6-3 ISSN: 97-537 COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS

More information

Cognitive Access Algorithms For Multiple Access Channels

Cognitive Access Algorithms For Multiple Access Channels 203 IEEE 4th Workshop on Sgnal Processng Advances n Wreless Communcatons SPAWC) Cogntve Access Algorthms For Multple Access Channels Ychuan Hu and Alejandro Rbero, Department of Electrcal and Systems Engneerng,

More information

A SEPARABLE APPROXIMATION DYNAMIC PROGRAMMING ALGORITHM FOR ECONOMIC DISPATCH WITH TRANSMISSION LOSSES. Pierre HANSEN, Nenad MLADENOVI]

A SEPARABLE APPROXIMATION DYNAMIC PROGRAMMING ALGORITHM FOR ECONOMIC DISPATCH WITH TRANSMISSION LOSSES. Pierre HANSEN, Nenad MLADENOVI] Yugoslav Journal of Operatons Research (00) umber 57-66 A SEPARABLE APPROXIMATIO DYAMIC PROGRAMMIG ALGORITHM FOR ECOOMIC DISPATCH WITH TRASMISSIO LOSSES Perre HASE enad MLADEOVI] GERAD and Ecole des Hautes

More information

The Study of Teaching-learning-based Optimization Algorithm

The Study of Teaching-learning-based Optimization Algorithm Advanced Scence and Technology Letters Vol. (AST 06), pp.05- http://dx.do.org/0.57/astl.06. The Study of Teachng-learnng-based Optmzaton Algorthm u Sun, Yan fu, Lele Kong, Haolang Q,, Helongang Insttute

More information

Lecture 14: Bandits with Budget Constraints

Lecture 14: Bandits with Budget Constraints IEOR 8100-001: Learnng and Optmzaton for Sequental Decson Makng 03/07/16 Lecture 14: andts wth udget Constrants Instructor: Shpra Agrawal Scrbed by: Zhpeng Lu 1 Problem defnton In the regular Mult-armed

More information

Technical Note: Capacity Constraints Across Nests in Assortment Optimization Under the Nested Logit Model

Technical Note: Capacity Constraints Across Nests in Assortment Optimization Under the Nested Logit Model Techncal Note: Capacty Constrants Across Nests n Assortment Optmzaton Under the Nested Logt Model Jacob B. Feldman, Huseyn Topaloglu School of Operatons Research and Informaton Engneerng, Cornell Unversty,

More information

Capacity Constraints Across Nests in Assortment Optimization Under the Nested Logit Model

Capacity Constraints Across Nests in Assortment Optimization Under the Nested Logit Model Capacty Constrants Across Nests n Assortment Optmzaton Under the Nested Logt Model Jacob B. Feldman School of Operatons Research and Informaton Engneerng, Cornell Unversty, Ithaca, New York 14853, USA

More information