On the Capacity of Symmetric Gaussian Interference Channels with Feedback

Size: px
Start display at page:

Download "On the Capacity of Symmetric Gaussian Interference Channels with Feedback"

Transcription

1 O the Capacity of Symmetric Gaussia Iterferece Chaels with Feedback La V Truog Iformatio Techology Specializatio Departmet ITS FPT Uiversity, Haoi, Vietam latv@fpteduv Hirosuke Yamamoto Dept of Complexity Sciece ad Egieerig The Uiversity of Tokyo, Japa hirosuke@ieeeorg arxiv: v5 [csit] 21 Apr 2015 Abstract I this paper, we propose a ew codig scheme for symmetric Gaussia iterferece chaels with feedback based o the ideas of time-varyig codig schemes The proposed scheme improves the Suh-Tse ad Kramer ier bouds of the chael capacity for the cases of weak ad ot very strog iterferece This improvemet is more sigificat whe the sigal-to-oise ratio SNR is ot very high It is show theoretically ad umerically that our codig scheme ca outperform the Kramer code I additio, the geeralized degrees-of-freedom of our proposed codig scheme is equal to the Suh-Tse scheme i the strog iterferece case The umerical results show that our codig scheme ca attai better performace tha the Suh-Tse codig scheme for all chael parameters Furthermore, the simplicity of the ecodig/decodig algorithms is aother strog poit of our proposed codig scheme compared with the Suh-Tse codig scheme More importatly, our results show that a optimal codig scheme for the symmetric Gaussia iterferece chaels with feedback ca be achieved by usig oly margial posterior distributios uder a better cooperatio strategy betwee trasmitters Idex Terms Gaussia Iterferece Chael, Feedback, Posterior Matchig, Iterated Fuctio Systems I INTRODUCTION The capacity of the iterferece chael with feedback has bee still ukow for may decades although there were some progresses toward solvig this problem Kramer developed a feedback strategy ad derived a outer boud of the Gaussia chael [6], [9] However, the gap betwee the outer ad the ier bouds becomes large uboudedly as sigal to oise ratio SNR ad iterferece to oise ratio INR icrease Furthermore, Suh ad Tse [1], [2] characterized the capacity regio withi 2 bits/s/hz ad the symmetric capacity withi 1 bit/s/hz for the two-user Gaussia iterferece chael with feedback They also idicated that feedback provides multiplicative gai at high SNR However, their codig scheme does ot work well whe the SNR is close to the INR Its symmetric codig rate becomes less tha the Kramer code whe this coditio happes I additio, it also has lower performace tha the Kramer code whe α = log INR/ log SNR is ot very large ad the SNR is low cf Figs 1-2 of this paper or Fig 14 i [2] Later, the Suh-Tse codig scheme was exteded to M-user Gaussia iterferece chaels with feedback for M 3 [7] I this paper, we propose a codig scheme which achieves better performace tha the Suh-Tse code whe α = log INR/ log SNR is ot very large See Figs1-2 of this paper I additio, our code ca attai better symmetric rate tha the Kramer code for all chael Figure 1 Figure 2 Symmetric Rate Comparisio at High SNR Symmetric Rate Comparisio at Low SNR parameters, ad therefore it overcomes all the weak-poits of the Suh-Tse codig scheme ad improves the Suh- Tse ad Kramer ier bouds For the strog iterferece case, our code ca achieve the same geeralized degreesof-freedom as the Suh-Tse codig scheme Furthermore, umerical results show that our codig scheme ideed has better/equal performace tha/to the Suh-Tse codig scheme for all chael parameters II CHANNEL MODEL We cosider the Gaussia iterferece chael show i Fig 3, which has two seders ad two receivers Seder 1 seds a source of iformatio message poits Θ 1, which

2 Figure 3 Gaussia Iterferece Chael with Feedback is uiformly distributed i 0,1, to receiver 1 Seder 2 seds a source of iformatio poits Θ 2, which is also uiformly distributed i 0,1, to receiver 2 Assumig that Θ 1 is idepedet of Θ 2 ad each chael iterferes with the other Specially, we assume that Y 1 = X 1 + ax 2 + Z 1, Y 2 = X 2 + ax 1 + Z 2, where Z 1 N 0, σ1 2 ad Z 2 N 0, σ2 2 are Gaussia oise radom variables, ad the iput power costraits are P 1 ad P 2, respectively I this paper, we cosider the symmetric iterferece chael such that P 1 = P 2 = P ad σ1 2 = σ2 2 = 1 Hece, the sigal-to-oise ratio ad iterferece-to-oise ratio ca capture chael gais SNR = P ad INR = a 2 P We also assume that output symbols are casually feedbacked to the correspodig seder ad the trasmitted symbol X m at time ca deped o both the message Θ m ad the previous chael output sequece Y 1,m := Y m 1, Y m 2,, Y m 1 for m {1, 2} A trasmissio scheme for the two-user Gaussia iterferece chael with feedback is sequeces of measurable fuctios {g m : 0, 1 R 1 R} =1, m {1, 2} so that the iput to the chael geerated by the trasmitter is give by X m = g m Θ m, Y 1,m A decodig rule for the two-user Gaussia iterferece chael with feedback are sequeces of measurable mappigs { m : R E} =1, m {1, 2} where E is the set of all ope itervals i 0, 1 ad m y,m refers to the decoded iterval at receiver m The error probabilities at time associated with a trasmissio scheme ad a decodig rule, is defied as p m e := PΘ m / m Y,m, m {1, 2}, ad the correspodig rate pair R 1, R 2 at time is defied by R m := 1 log m Y,m We say that a trasmissio scheme together with a decodig rule achieves a rate pair R 1, R 2 over a Gaussia iterferece chael if for m {1, 2} we have lim P R m < R m = 0, lim pm e = 0 1 The rate pair is achieved withi iput power costraits P 1, P 2 if the followig is satisfied: 1 lim sup E[X m k ] 2 P m, m {1, 2} 2 k=1 The symmetric capacity is defied by C sym := sup{r : R, R is achievable} A optimal fixed rate decodig rule for the two-user Gaussia iterferece chael with feedback for rate pair R 1, R 2 is the oe that decodes a pair of fixed legth itervals {J 1, J 2 : J m = 2 Rm for m {1, 2}}, which maximizes posteriori probabilities, ie, m y,m = argmax J m E: J m =2 Rm P Θm Y J m y,m It is easy to see that the optimal fixed rate decodig rule for the Gaussia iterferece chael with feedback is the traditioal MAP, MMSE decodig rule A optimal variable rate decodig rule with target error probabilities p m e = δ m is the oe that decodes a pair of miimal-legth itervals J 1, J 2 such that accumulated margial posteriori probabilities exceeds correspodig targets, ie, m y,m = mi J m E:P Θm Y J m y,m 1 δ m J m Both decodig rules use the margial posterior distributio of the message poit P Θm Y which ca be calculated olie at the trasmitters ad the receivers Refer [6] for more details Lemma 1: The achievability i the defiitio 1 ad 2 implies the achievability i the stadard framework Proof: See the detailed proof i [3], [4], ad [5] Notatios: The cumulative distributio fuctio cdf of a radom variable X is give by F X x = P X, x], ad their iverse cdf is defied as F 1 X t := if{x : F Xx > t} The compositio fuctio is defied by f gx = fgx, ad the sig fuctio is defied as sgx := 1 if x 0 ad sgx := 1 if x < 0 We also use x + := maxx, 0 ad log + x := maxlog x, 0 III A CODING SCHEME FOR GAUSSIAN INTERFERENCE CHANNELS WITH FEEDBACK I this sectio, we propose a time-varyig codig scheme for the symmetric Gaussia iterferece chael with feedback as followig: A Ecodig Step 1: Trasmitter m seds X m 1 = F 1 X Θ m where m {1, 2} ad X N 0, P 1 for some P 1 > 0 We also set ρ 1 := E[F 1 X Step + 1 for 1: Both trasmitters estimate Θ 1F 1 X Θ 2] = 0 P 1 ρ +1 = 1 β 2 {ρ 2b sgρ ρ + a +b 2 sgρ [ ρ 1 + a a ] }

3 Trasmitter 1 seds X 1 +1 sgρ +1 where X 1 +1 := 1 β X 1 b sgρ Y 1 Trasmitter 2 seds X 2 +1sga where X 2 +1 := 1 β X 2 b sgay 2 Receiver 1 receives Y 1 +1 = X1 +1 sgρ +1 + a X Z1 +1 Receiver 2 receives Y 2 +1 = sgax a sgρ +1X Z2 +1 Both receivers feedback their received sigals to the correspodig trasmitters Here, β > 0, b should be chose to satisfy the followig costraits: lim sup N B Decodig 1 N N =1 E[X m ] 2 P, m {1, 2} 3 At each time slot, receiver m {1, 2} selects a fixed iterval J m 1 = s m, t m R as the decoded iterval with respect to X m +1 The, set the decoded iterval J m = s m, T m t m, as the decoded iterval T m with respect to X m 1, where T m s := w m 1 w m 2 w m s, s R ad w 1 := β x + b sgρ Y 1, w 2 := β x + b sgay 2 Receiver m sets the decoded iterval for the message Θ m as follows: m Y,m = F Xm J m We call this codig strategy the Gaussia iterferece time-varyig feedback codig strategy, which is a optimal variable rate decodig rule with doubly expoetial decay of targeted error probabilities see the proof of the Lemma 2 i this paper IV A NEW ACHIEVABLE RATE REGION Lemma 2: Uder the coditio that 0 < lim sup β < 1, the time-varyig codig scheme for the symmetric Gaussia iterferece chael with feedback achieves the followig symmetric rate: R sym = lim sup log β bits/chael use Proof: We provide a sketch of the proof The detailed oe ca be foud i papers [3], [4], ad [5] Defie β := lim sup β It is easy to see that R sym = log β 1 For ay R < R sym, we ca fid a ɛ > 0 such that R < logβ + ɛ 1 Choose a N ɛ N such that sup Nɛ β < β + ɛ Usig the Law of Iterated Expectatios Fubii s Theorem, we ca show that PR m We ca also show that < R A ɛ 2 R β + ɛ Nɛ J m 1 p m e = O exp m J 1 2 8P By choosig J m 1 = o 2 logβ+ɛ 1 R o 2 Rsym R, ad t m = s m = J m 1 /2, the aforemetioed coditio 1 is satisfied Here, symbols Ox ad ox are Ladau symbols Besides, the choice of sequeces b, β followig the rule 3 leads to the fact that the iput power costraits are also satisfied Theorem 1: The o-degraded symmetric Gaussia iterferece chael a 0 ca achieve the followig symmetric rate: [ log where ad Deote R sym bits/chael use = 1 2 max ρ [0,ρ 0], b {b 1,b 2 } P P + b 2 [1 + P + a 2 P + 2 a P ρ] 2bP [1 + a ρ] 0 < ρ 0 := a 2 P 2 + P P [2a 2 P 2 + P ] a 2 P 2 < 1, b 2P ρ + a P + a P ρ 2 1,2 = 2 a P + 2P ρ + 2 a 2 P ρ + ρ + 2 a P ρ 2 P 2 a ± 2 ρ 4 2ρ 2 a 2 P 2 + P + a 2 P 2 2 a P + 2P ρ + 2 a 2 P ρ + ρ + 2 a P ρ 2 Proof: From our trasmissio strategy, we have X 1 +1 = 1 β X 1 b sgρ Y 1, 4 X 2 +1 = 1 β X 2 b sgay 2 5 P = E[X 1 ] 2 = E[X 2 ] 2, ρ = E[X1 X 2 ] P We ca show by iductio that E[X 1 ] 2 = E[X 2 ] 2 for all Therefore, it is easy to see that E[X 1 Y 2 E[X 2 Y 1 ] = P ρ + a, ] = P sgasgρ ρ + a, E[Y 1 Y 2 ] = P sga [ ρ 1 + a a ] Note that E[X 1 +1 X2 +1 ] = 1 { β 2 E[X 1 X 2 ] b sgρ E[X 2 Y 1 ] b sgae[x 1 +b 2 sgρ sgae[y 1 Y 2 ] } Y 2 ] ],

4 Fially, we obtai P +1 ρ +1 = P sgρ 1 β 2 { ρ 2b ρ + a Similarly, observe that +b 2 [ ρ 1 + a a ] } 6 E[X 1 Y 1 ] = P sgρ [1 + a ρ ], E[X 2 Y 2 ] = P sga[1 + a ρ ], E[Y 1 ] 2 = 1 + P + a 2 P + 2 a ρ P, E[Y 2 ] 2 = 1 + P + a 2 P + 2 a ρ P From the relatios 4 ad 5 we have P +1 = 1 +b 2 β 2 {P 2P b [1 + a ρ ] [ 1 + P + a 2 P + 2 a ρ P ]} 7 From 6 ad 7, if we ca force P P, b b, β β ad ρ ρ [0, 1], we obtai the followig equatios with three ukows b, ρ, β: P = 1 β 2 [ P 2bP 1 + a ρ + b P + a 2 P + 2 a ρp ], ρ = 1 β 2 [ ρ 2bρ + a + b 2 ρ1 + a a ] 8a 8b By elimiatig β ad cosiderig ρ as a ruig variable, we have the followig quadratic equatio i b for each fixed choice of ρ: b 2 [2 a P + 2P ρ + 2 a 2 P ρ + ρ + 2 a P ρ 2 ] 2b[2P ρ + a P + P a ρ 2 ] + 2P ρ = 0 9 After some simple derivatios, the discrimiat of this quadratic equatio is give by = P 2 a 2 ρ 4 2ρ 2 a 2 P 2 +P + a 2 P 2 := fρ 10 Sice f0 = a 2 P 2 > 0 ad f1 = 2P < 0, there exists the miimum value ρ 0 0, 1 such that fρ 0 = 0 Specifically, the value of ρ 0 is give by a 0 < ρ 0 = 2 P 2 + P P [2a 2 P 2 + P ] a 2 P 2 < 1 O the other had, sice the derivative of fρ satisfies f ρ = 4P 2 a 2 ρ 3 ρ 4P ρ 0, for all ρ [0, 1, we have = fρ 0 for all ρ [0, ρ 0 ] For all these values of ρ, we ca easily show that 9 ca have two positive solutios b 1, b 2 as the theorem statemet As a result, 8a ad 8b have at least oe solutio b, β for each fixed ρ [0, ρ 0 ] Therefore, if we force all the sequeces b, β, P, ρ to coverges to b, β, P, ρ, the symmetric Gaussia iterferece chael with feedback ca achieve the followig rate by the Lemma 2: + R sym = log mi β = 1 max ρ [0,ρ 0] 2 ρ [0,ρ 0],b {b 1,b 2 } log + P P + b 2 [1 + P + a 2 P + 2 a P ρ] 2bP [1 + a ρ] Note that sice Lemma 2 holds oly for 0 < β < 1, the superscript + is ecessary to deal with geeral cases To complete the proof, we show a procedure to force ρ = ρ ρ = 1 ρ for ay ρ [0, ρ 0 ], b = b, P = P, β = β for ay 2 Ideed, from 6 ad 7, we firstly force P 2 = P, ρ 2 = ρ, ad β 2 = β by settig P ρ = P 1 β1 2 {2 a b 2 1 b 1 }, 11 P = 1 β1 2 {P 1 2b 1 P 1 + b P 1 + a 2 P 1 } 12 This procedure is feasible because 11 ad 12 have at least oe solutio which is a triplet b 1, P 1 > 0, β 1 > 0 for each ρ [0, ρ 0 ] Ideed, For ρ = 0, we ca choose b 1 = 0, P 1 = P, β 1 = 1 For ρ 0, from 11 ad 12 we have the followig quadratic equatio i b 1 b 2 1[1 + P 1 + a 2 P 1 ρ 2 a P 1 ] 2ρ a P 1 b 1 + P 1 ρ = 0 13 The discrimiat of this quadratic equatio ca easily be show to be equal to Eρ = a 2 1 ρ 2 P 2 1 P 1 ρ 2 For the case ρ a, we ca choose P 1 such that this discrimiat is equal to zero by settig P 1 = ρ 2 /a 2 1 ρ 2 By this choice of P 1, we obtai b 1 = ρ a P P 1 + a 2 P 1 ρ 2 a P 1 I order for 11 ad 12 to have solutio β 1, we eed to show that the above choices of P 1, b 1 satisfy b 2 1 b 1 > 0 Clearly for ρ < a, this requiremet is satisfied by otig that b 1 < 0 sice ρ a < 0 ad ρ > 2 a ρ2 a 2 + ρ 2 = 2 a P P 1 + a 2 P 1 For ρ > a a < 1, of course, observe that P 1 = ρ 2 a 2 1 ρ 2 > ρ a a 2 ρ It follows that ρ a P 1 > 1+P 1 +a 2 P 1 ρ 2 a P 1 > 0 or b 1 > 1 This also meas that b 2 1 b 1 > 0 For the case ρ = a ay choice of P 1 > 1/1 a 2 works sice we have E a > 0 Besides, the sum of two solutios of the quadratic equatio 13 i b 1 is equal to zero, ad their product is ot equal to zero by usig Vieta s formula Hece, we must fid at least oe b 1 < 0 or b 2 1 b 1 > 0 Fially, we oly eed to set β = β, P = P, b = b which is a solutio of 8a ad 8b for each choice of ρ [0, ρ 0 ] ad for all 3 Here b should be chose to miimize β for each choice of ρ [0, ρ 0 ] i order to maximize the achievable symmetric rate of our codig scheme Last but ot least, we ca show that our code has better performace tha the Kramer code [6], ad therefore the superscript + ca be got rid of from the achievable symmetric rate formula Remark 1: For the degraded Gaussia iterferece chael with feedback a = 0, 8a ad 8b become ρ = 1 β 2 ρb 12, P = 1 β 2 [P b 12 + b 2 ] 14

5 From 14, we must have ρ = 0 ad the achievable symmetric rate becomes R sym = 1 [ ] 2 max log P b R P b b 2 = 1 log1 + P 2 This result coicides with the well-kow capacity of this chael with o iterferece Corollary 1: The proposed time-varyig code outperforms the Kramer code for all chael parameters Proof: A variat of Kramer code is costructed by choosig triplet b, β, ρ, which is a solutio of 8a ad 8b, as follows: b = P 1 + a ρ P 1 + a a ρ + 1, β = a 2 P 1 ρ P 1 + a a ρ + 1, ad ρ is the uique solutio i 0, 1 of the ext equatio: 2 a 3 P 2 ρ 4 + a 2 P ρ 3 4 a P a 2 P + 1ρ 2 2a 2 P + P + 2ρ + 2 a P a 2 P + 1 = 0 15 See also i [2], [6], [9] Therefore, it is iferior to the proposed code i this paper Remark 2: The choice of b ad β for the Kramer code is to maximize the achievable rate or miimize β by usig oly 8a for each fixed value of ρ I order to satisfy 8b, the choice of b, β i the above proof is applicable oly to the fixed value of ρ which is the uique solutio i 0, 1 of 14 For other values of ρ, from 8a ad 8b, we see that b ad β are give by two differet fuctios of ρ Elargig the set of possible choices of ρ the b, β icreases the achievable symmetric rate i this paper Corrolary 2: For α = log INR/ log SNR > 1, the geeralized degrees of freedom of the proposed codig scheme is give by dα := R sym SNR, INR lim = α SNR,INR log SNR 2 Here, the uit of R sym is bits/s/hz As a cosequece, the proposed codig scheme has the same geeralized degrees of freedom as the Suh-Tse codig scheme [1], [2] ad also provides the multiplicative gai at high SNR Note that the Kramer code achieves oly dα = 1 + α/4 for α 1 cf 48 [2] Proof: Sice α = log INR/ log SNR > 1, we have a 2 P = P α, or a = P α 1/2 For P sufficietly large ad ρ 1, observe that b 1,2 P α+1/2 1 + ρ 2 2P α ρ ± P α+1/2 1 ρ 2 2P α ρ By choosig b = b 1 P 1 α/2 ρ, we obtai R sym ρ log + P P + b 2 P α 2bP 1 + P α 1/2 ρ = log 1 ρ 2 2ρP 1 α/2 bits/s/hz 16 From Theorem 1, we also have ρ 2 0 = 1 + P α 2P α + P 2α > 1 3P α/2, ad the ρ, that maximizes the achievable rate, must be i the iterval [0, ρ 0 ] This coditio is satisfied by settig ρ = 1 P γ + P α 1 P α 1/2 for a arbitrary positive umber γ < α/2 although this ρ may ot be optimal Ideed, for P sufficietly large, we have ρ 1 ad ρ < 1 P γ + P α 1 P α 1/2 = 1 P γ Hece ρ 2 < 1 P γ < 1 3P α/2 < ρ 2 0 O the other had, sice 1 ρ 2 2ρ P 1 α/2 = P γ, we obtai dα R sym ρ lim SNR,INR log SNR = γ Sice γ ca take ay arbitrary value less tha α/2, we have dα α/2 From the result of [2], it is kow that dα α/2 Hece, we must have dα = α/2 V NUMERICAL EVALUATION I order to evaluate R sym i Theorem 1 umerically, we determied the optimal ρ by icreasig it from zero by icremetal step 10 5 The umerical results are show i Figs 1-2 i Sectio I We ote that our proposed code ca achieve better/equal symmetric rate tha/to the Suh- Tse ad Kramer codes for all chael parameter a, P This improvemet is more sigificat whe SNR is ot too high VI CONCLUSION The ier boud of the capacity is improved compared with the Suh-Tse ad Kramer ier bouds by aalyzig the performace of our proposed code as a optimized form of the Kramer code Our result also shows that a optimal codig scheme for the Gaussia iterferece chael with feedback ca be achieved by usig margial posterior distributios ACKNOWLEDGMENT This work was supported i part by JSPS KAKENHI Grat Number REFERENCES [1] Chagho Suh ad David Tse, Symmetric Feedback Capacity of the Gaussia Iterferece Chael to Withi Oe Bit, i Proc It Symp Iformatio Theory, Ju 2009 [2] Chagho Suh ad David Tse, Feedback Capacity of the Gaussia Iterferece Chael to Withi 2 Bits, IEEE Tras If Theory, vol 57, No 5, pp , May 2011 [3] La V Truog, Posterior Matchig Scheme for Gaussia Multiple Access Chael with Feedback, [Olie] Available: [4] La V Truog, Posterior Matchig Scheme for Gaussia Multiple Access Chael with Feedback, i Proc IEEE Iformatio Theory Workshop, Nov 2014 [5] La V Truog, A Novel Time-Varyig Codig Scheme for the Gaussia Broadcast Chael with Feedback, [Olie] Available: [6] Gerhard Kramer, Feedback Strategies for White Gaussia Iterferece Networks, IEEE Tras If Theory, vol 48, pp , Ja 2002 [7] Ravi Tado, Soheil Mohajer, ad H Vicet Poor, O the Symmetric Feedback Capacity of the K-User Cyclic Z- Iterferece Chael, IEEE Tras If Theory, vol 59, o5, pp , May 2013 [8] Ravi Tado, Soheil Mohajer, ad H Vicet Poor, O the Feedback Capacity of the Fully Coected K-User Iterferece Chael, IEEE Tras If Theory, vol 59, o5, pp , May 2013 [9] G Kramer, Correctio to Feedback Strategies for White Gaussia Iterferece Networks, ad a Capacity Theorem for Gaussia Iterferece Chaels with Feedback, IEEE Tras If Theory, vol 50, o 6, pp , Ju 2004 [10] O Shayevitz ad M Feder, Optimal Feedback Commuicatio via Posterior Matchig, IEEE Tras If Theory, vol 57, o3, pp , Mar 2011

Posterior Matching Scheme for Gaussian Multiple Access Channel with Feedback

Posterior Matching Scheme for Gaussian Multiple Access Channel with Feedback 1 Posterior Matchig Scheme for Gaussia Multiple Access Chael with Feedback La V. Truog, Member, IEEE, Iformatio Techology Specializatio Departmet ITS, FPT Uiversity, Haoi, Vietam E-mail: latv@fpt.edu.v

More information

Symmetric Two-User Gaussian Interference Channel with Common Messages

Symmetric Two-User Gaussian Interference Channel with Common Messages Symmetric Two-User Gaussia Iterferece Chael with Commo Messages Qua Geg CSL ad Dept. of ECE UIUC, IL 680 Email: geg5@illiois.edu Tie Liu Dept. of Electrical ad Computer Egieerig Texas A&M Uiversity, TX

More information

The Maximum-Likelihood Decoding Performance of Error-Correcting Codes

The Maximum-Likelihood Decoding Performance of Error-Correcting Codes The Maximum-Lielihood Decodig Performace of Error-Correctig Codes Hery D. Pfister ECE Departmet Texas A&M Uiversity August 27th, 2007 (rev. 0) November 2st, 203 (rev. ) Performace of Codes. Notatio X,

More information

Randomized Algorithms I, Spring 2018, Department of Computer Science, University of Helsinki Homework 1: Solutions (Discussed January 25, 2018)

Randomized Algorithms I, Spring 2018, Department of Computer Science, University of Helsinki Homework 1: Solutions (Discussed January 25, 2018) Radomized Algorithms I, Sprig 08, Departmet of Computer Sciece, Uiversity of Helsiki Homework : Solutios Discussed Jauary 5, 08). Exercise.: Cosider the followig balls-ad-bi game. We start with oe black

More information

Asymptotic Coupling and Its Applications in Information Theory

Asymptotic Coupling and Its Applications in Information Theory Asymptotic Couplig ad Its Applicatios i Iformatio Theory Vicet Y. F. Ta Joit Work with Lei Yu Departmet of Electrical ad Computer Egieerig, Departmet of Mathematics, Natioal Uiversity of Sigapore IMS-APRM

More information

Information Theory Tutorial Communication over Channels with memory. Chi Zhang Department of Electrical Engineering University of Notre Dame

Information Theory Tutorial Communication over Channels with memory. Chi Zhang Department of Electrical Engineering University of Notre Dame Iformatio Theory Tutorial Commuicatio over Chaels with memory Chi Zhag Departmet of Electrical Egieerig Uiversity of Notre Dame Abstract A geeral capacity formula C = sup I(; Y ), which is correct for

More information

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.436J/15.085J Fall 2008 Lecture 19 11/17/2008 LAWS OF LARGE NUMBERS II THE STRONG LAW OF LARGE NUMBERS

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.436J/15.085J Fall 2008 Lecture 19 11/17/2008 LAWS OF LARGE NUMBERS II THE STRONG LAW OF LARGE NUMBERS MASSACHUSTTS INSTITUT OF TCHNOLOGY 6.436J/5.085J Fall 2008 Lecture 9 /7/2008 LAWS OF LARG NUMBRS II Cotets. The strog law of large umbers 2. The Cheroff boud TH STRONG LAW OF LARG NUMBRS While the weak

More information

Lecture 7: Channel coding theorem for discrete-time continuous memoryless channel

Lecture 7: Channel coding theorem for discrete-time continuous memoryless channel Lecture 7: Chael codig theorem for discrete-time cotiuous memoryless chael Lectured by Dr. Saif K. Mohammed Scribed by Mirsad Čirkić Iformatio Theory for Wireless Commuicatio ITWC Sprig 202 Let us first

More information

Lecture 27. Capacity of additive Gaussian noise channel and the sphere packing bound

Lecture 27. Capacity of additive Gaussian noise channel and the sphere packing bound Lecture 7 Ageda for the lecture Gaussia chael with average power costraits Capacity of additive Gaussia oise chael ad the sphere packig boud 7. Additive Gaussia oise chael Up to this poit, we have bee

More information

On Random Line Segments in the Unit Square

On Random Line Segments in the Unit Square O Radom Lie Segmets i the Uit Square Thomas A. Courtade Departmet of Electrical Egieerig Uiversity of Califoria Los Ageles, Califoria 90095 Email: tacourta@ee.ucla.edu I. INTRODUCTION Let Q = [0, 1] [0,

More information

Let us give one more example of MLE. Example 3. The uniform distribution U[0, θ] on the interval [0, θ] has p.d.f.

Let us give one more example of MLE. Example 3. The uniform distribution U[0, θ] on the interval [0, θ] has p.d.f. Lecture 5 Let us give oe more example of MLE. Example 3. The uiform distributio U[0, ] o the iterval [0, ] has p.d.f. { 1 f(x =, 0 x, 0, otherwise The likelihood fuctio ϕ( = f(x i = 1 I(X 1,..., X [0,

More information

Lecture 11: Channel Coding Theorem: Converse Part

Lecture 11: Channel Coding Theorem: Converse Part EE376A/STATS376A Iformatio Theory Lecture - 02/3/208 Lecture : Chael Codig Theorem: Coverse Part Lecturer: Tsachy Weissma Scribe: Erdem Bıyık I this lecture, we will cotiue our discussio o chael codig

More information

Lecture 15: Strong, Conditional, & Joint Typicality

Lecture 15: Strong, Conditional, & Joint Typicality EE376A/STATS376A Iformatio Theory Lecture 15-02/27/2018 Lecture 15: Strog, Coditioal, & Joit Typicality Lecturer: Tsachy Weissma Scribe: Nimit Sohoi, William McCloskey, Halwest Mohammad I this lecture,

More information

This exam contains 19 pages (including this cover page) and 10 questions. A Formulae sheet is provided with the exam.

This exam contains 19 pages (including this cover page) and 10 questions. A Formulae sheet is provided with the exam. Probability ad Statistics FS 07 Secod Sessio Exam 09.0.08 Time Limit: 80 Miutes Name: Studet ID: This exam cotais 9 pages (icludig this cover page) ad 0 questios. A Formulae sheet is provided with the

More information

62. Power series Definition 16. (Power series) Given a sequence {c n }, the series. c n x n = c 0 + c 1 x + c 2 x 2 + c 3 x 3 +

62. Power series Definition 16. (Power series) Given a sequence {c n }, the series. c n x n = c 0 + c 1 x + c 2 x 2 + c 3 x 3 + 62. Power series Defiitio 16. (Power series) Give a sequece {c }, the series c x = c 0 + c 1 x + c 2 x 2 + c 3 x 3 + is called a power series i the variable x. The umbers c are called the coefficiets of

More information

7.1 Convergence of sequences of random variables

7.1 Convergence of sequences of random variables Chapter 7 Limit Theorems Throughout this sectio we will assume a probability space (, F, P), i which is defied a ifiite sequece of radom variables (X ) ad a radom variable X. The fact that for every ifiite

More information

1 Introduction to reducing variance in Monte Carlo simulations

1 Introduction to reducing variance in Monte Carlo simulations Copyright c 010 by Karl Sigma 1 Itroductio to reducig variace i Mote Carlo simulatios 11 Review of cofidece itervals for estimatig a mea I statistics, we estimate a ukow mea µ = E(X) of a distributio by

More information

SUCCESSIVE INTERFERENCE CANCELLATION DECODING FOR THE K -USER CYCLIC INTERFERENCE CHANNEL

SUCCESSIVE INTERFERENCE CANCELLATION DECODING FOR THE K -USER CYCLIC INTERFERENCE CHANNEL Joural of Theoretical ad Applied Iformatio Techology 31 st December 212 Vol 46 No2 25-212 JATIT & LLS All rights reserved ISSN: 1992-8645 wwwatitorg E-ISSN: 1817-3195 SCCESSIVE INTERFERENCE CANCELLATION

More information

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2018 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

Sequences and Series of Functions

Sequences and Series of Functions Chapter 6 Sequeces ad Series of Fuctios 6.1. Covergece of a Sequece of Fuctios Poitwise Covergece. Defiitio 6.1. Let, for each N, fuctio f : A R be defied. If, for each x A, the sequece (f (x)) coverges

More information

Problem Set 4 Due Oct, 12

Problem Set 4 Due Oct, 12 EE226: Radom Processes i Systems Lecturer: Jea C. Walrad Problem Set 4 Due Oct, 12 Fall 06 GSI: Assae Gueye This problem set essetially reviews detectio theory ad hypothesis testig ad some basic otios

More information

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2018 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

Increasing timing capacity using packet coloring

Increasing timing capacity using packet coloring 003 Coferece o Iformatio Scieces ad Systems, The Johs Hopkis Uiversity, March 4, 003 Icreasig timig capacity usig packet colorig Xi Liu ad R Srikat[] Coordiated Sciece Laboratory Uiversity of Illiois e-mail:

More information

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2016 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

EECS564 Estimation, Filtering, and Detection Hwk 2 Solns. Winter p θ (z) = (2θz + 1 θ), 0 z 1

EECS564 Estimation, Filtering, and Detection Hwk 2 Solns. Winter p θ (z) = (2θz + 1 θ), 0 z 1 EECS564 Estimatio, Filterig, ad Detectio Hwk 2 Sols. Witer 25 4. Let Z be a sigle observatio havig desity fuctio where. p (z) = (2z + ), z (a) Assumig that is a oradom parameter, fid ad plot the maximum

More information

Chapter 6 Infinite Series

Chapter 6 Infinite Series Chapter 6 Ifiite Series I the previous chapter we cosidered itegrals which were improper i the sese that the iterval of itegratio was ubouded. I this chapter we are goig to discuss a topic which is somewhat

More information

Machine Learning Brett Bernstein

Machine Learning Brett Bernstein Machie Learig Brett Berstei Week Lecture: Cocept Check Exercises Starred problems are optioal. Statistical Learig Theory. Suppose A = Y = R ad X is some other set. Furthermore, assume P X Y is a discrete

More information

Generalized Semi- Markov Processes (GSMP)

Generalized Semi- Markov Processes (GSMP) Geeralized Semi- Markov Processes (GSMP) Summary Some Defiitios Markov ad Semi-Markov Processes The Poisso Process Properties of the Poisso Process Iterarrival times Memoryless property ad the residual

More information

ECE 901 Lecture 12: Complexity Regularization and the Squared Loss

ECE 901 Lecture 12: Complexity Regularization and the Squared Loss ECE 90 Lecture : Complexity Regularizatio ad the Squared Loss R. Nowak 5/7/009 I the previous lectures we made use of the Cheroff/Hoeffdig bouds for our aalysis of classifier errors. Hoeffdig s iequality

More information

STAT Homework 1 - Solutions

STAT Homework 1 - Solutions STAT-36700 Homework 1 - Solutios Fall 018 September 11, 018 This cotais solutios for Homework 1. Please ote that we have icluded several additioal commets ad approaches to the problems to give you better

More information

An Introduction to Randomized Algorithms

An Introduction to Randomized Algorithms A Itroductio to Radomized Algorithms The focus of this lecture is to study a radomized algorithm for quick sort, aalyze it usig probabilistic recurrece relatios, ad also provide more geeral tools for aalysis

More information

1 of 7 7/16/2009 6:06 AM Virtual Laboratories > 6. Radom Samples > 1 2 3 4 5 6 7 6. Order Statistics Defiitios Suppose agai that we have a basic radom experimet, ad that X is a real-valued radom variable

More information

x = Pr ( X (n) βx ) =

x = Pr ( X (n) βx ) = Exercise 93 / page 45 The desity of a variable X i i 1 is fx α α a For α kow let say equal to α α > fx α α x α Pr X i x < x < Usig a Pivotal Quatity: x α 1 < x < α > x α 1 ad We solve i a similar way as

More information

Cooperative Communication Fundamentals & Coding Techniques

Cooperative Communication Fundamentals & Coding Techniques 3 th ICACT Tutorial Cooperative commuicatio fudametals & codig techiques Cooperative Commuicatio Fudametals & Codig Techiques 0..4 Electroics ad Telecommuicatio Research Istitute Kiug Jug 3 th ICACT Tutorial

More information

Lecture 7: MIMO Architectures Theoretical Foundations of Wireless Communications 1. Overview. CommTh/EES/KTH

Lecture 7: MIMO Architectures Theoretical Foundations of Wireless Communications 1. Overview. CommTh/EES/KTH : Theoretical Foudatios of Wireless Commuicatios 1 Thursday, May 19, 2016 12:30-15:30, Coferece Room SIP 1 Textbook: D. Tse ad P. Viswaath, Fudametals of Wireless Commuicatio 1 / 1 Overview Lecture 6:

More information

Optimization Methods MIT 2.098/6.255/ Final exam

Optimization Methods MIT 2.098/6.255/ Final exam Optimizatio Methods MIT 2.098/6.255/15.093 Fial exam Date Give: December 19th, 2006 P1. [30 pts] Classify the followig statemets as true or false. All aswers must be well-justified, either through a short

More information

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.265/15.070J Fall 2013 Lecture 2 9/9/2013. Large Deviations for i.i.d. Random Variables

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.265/15.070J Fall 2013 Lecture 2 9/9/2013. Large Deviations for i.i.d. Random Variables MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.265/15.070J Fall 2013 Lecture 2 9/9/2013 Large Deviatios for i.i.d. Radom Variables Cotet. Cheroff boud usig expoetial momet geeratig fuctios. Properties of a momet

More information

ECE 564/645 - Digital Communication Systems (Spring 2014) Final Exam Friday, May 2nd, 8:00-10:00am, Marston 220

ECE 564/645 - Digital Communication Systems (Spring 2014) Final Exam Friday, May 2nd, 8:00-10:00am, Marston 220 ECE 564/645 - Digital Commuicatio Systems (Sprig 014) Fial Exam Friday, May d, 8:00-10:00am, Marsto 0 Overview The exam cosists of four (or five) problems for 100 (or 10) poits. The poits for each part

More information

Optimally Sparse SVMs

Optimally Sparse SVMs A. Proof of Lemma 3. We here prove a lower boud o the umber of support vectors to achieve geeralizatio bouds of the form which we cosider. Importatly, this result holds ot oly for liear classifiers, but

More information

Lecture 11 and 12: Basic estimation theory

Lecture 11 and 12: Basic estimation theory Lecture ad 2: Basic estimatio theory Sprig 202 - EE 94 Networked estimatio ad cotrol Prof. Kha March 2 202 I. MAXIMUM-LIKELIHOOD ESTIMATORS The maximum likelihood priciple is deceptively simple. Louis

More information

Math 140A Elementary Analysis Homework Questions 3-1

Math 140A Elementary Analysis Homework Questions 3-1 Math 0A Elemetary Aalysis Homework Questios -.9 Limits Theorems for Sequeces Suppose that lim x =, lim y = 7 ad that all y are o-zero. Detarime the followig limits: (a) lim(x + y ) (b) lim y x y Let s

More information

A Non-Asymptotic Achievable Rate for the AWGN Energy-Harvesting Channel using Save-and-Transmit

A Non-Asymptotic Achievable Rate for the AWGN Energy-Harvesting Channel using Save-and-Transmit 016 IEEE Iteratioal Symposium o Iformatio Theory A No-Asymptotic Achievable Rate for the AWGN Eergy-Harvestig Chael usig Save-ad-Trasmit Silas L. Fog, Vicet Y. F. Ta, ad Jig Yag Departmet of Electrical

More information

Confidence interval for the two-parameter exponentiated Gumbel distribution based on record values

Confidence interval for the two-parameter exponentiated Gumbel distribution based on record values Iteratioal Joural of Applied Operatioal Research Vol. 4 No. 1 pp. 61-68 Witer 2014 Joural homepage: www.ijorlu.ir Cofidece iterval for the two-parameter expoetiated Gumbel distributio based o record values

More information

5.1. The Rayleigh s quotient. Definition 49. Let A = A be a self-adjoint matrix. quotient is the function. R(x) = x,ax, for x = 0.

5.1. The Rayleigh s quotient. Definition 49. Let A = A be a self-adjoint matrix. quotient is the function. R(x) = x,ax, for x = 0. 40 RODICA D. COSTIN 5. The Rayleigh s priciple ad the i priciple for the eigevalues of a self-adjoit matrix Eigevalues of self-adjoit matrices are easy to calculate. This sectio shows how this is doe usig

More information

ACO Comprehensive Exam 9 October 2007 Student code A. 1. Graph Theory

ACO Comprehensive Exam 9 October 2007 Student code A. 1. Graph Theory 1. Graph Theory Prove that there exist o simple plaar triagulatio T ad two distict adjacet vertices x, y V (T ) such that x ad y are the oly vertices of T of odd degree. Do ot use the Four-Color Theorem.

More information

Introduction to Optimization Techniques. How to Solve Equations

Introduction to Optimization Techniques. How to Solve Equations Itroductio to Optimizatio Techiques How to Solve Equatios Iterative Methods of Optimizatio Iterative methods of optimizatio Solutio of the oliear equatios resultig form a optimizatio problem is usually

More information

THE interference channel problem describes a setup where multiple pairs of transmitters and receivers share a communication

THE interference channel problem describes a setup where multiple pairs of transmitters and receivers share a communication Trade-off betwee Commuicatio ad Cooperatio i the Iterferece Chael Farhad Shirai EECS Departmet Uiversity of Michiga A Arbor,USA Email: fshirai@umich.edu S. Sadeep Pradha EECS Departmet Uiversity of Michiga

More information

MAS111 Convergence and Continuity

MAS111 Convergence and Continuity MAS Covergece ad Cotiuity Key Objectives At the ed of the course, studets should kow the followig topics ad be able to apply the basic priciples ad theorems therei to solvig various problems cocerig covergece

More information

Channel coding, linear block codes, Hamming and cyclic codes Lecture - 8

Channel coding, linear block codes, Hamming and cyclic codes Lecture - 8 Digital Commuicatio Chael codig, liear block codes, Hammig ad cyclic codes Lecture - 8 Ir. Muhamad Asial, MSc., PhD Ceter for Iformatio ad Commuicatio Egieerig Research (CICER) Electrical Egieerig Departmet

More information

Random Variables, Sampling and Estimation

Random Variables, Sampling and Estimation Chapter 1 Radom Variables, Samplig ad Estimatio 1.1 Itroductio This chapter will cover the most importat basic statistical theory you eed i order to uderstad the ecoometric material that will be comig

More information

Infinite Sequences and Series

Infinite Sequences and Series Chapter 6 Ifiite Sequeces ad Series 6.1 Ifiite Sequeces 6.1.1 Elemetary Cocepts Simply speakig, a sequece is a ordered list of umbers writte: {a 1, a 2, a 3,...a, a +1,...} where the elemets a i represet

More information

Non-Asymptotic Achievable Rates for Gaussian Energy-Harvesting Channels: Best-Effort and Save-and-Transmit

Non-Asymptotic Achievable Rates for Gaussian Energy-Harvesting Channels: Best-Effort and Save-and-Transmit No-Asymptotic Achievable Rates for Gaussia Eergy-Harvestig Chaels: Best-Effort ad Save-ad-Trasmit Silas L. Fog, Jig Yag, ad Ayli Yeer arxiv:805.089v [cs.it] 30 May 08 Abstract A additive white Gaussia

More information

6.3 Testing Series With Positive Terms

6.3 Testing Series With Positive Terms 6.3. TESTING SERIES WITH POSITIVE TERMS 307 6.3 Testig Series With Positive Terms 6.3. Review of what is kow up to ow I theory, testig a series a i for covergece amouts to fidig the i= sequece of partial

More information

Since X n /n P p, we know that X n (n. Xn (n X n ) Using the asymptotic result above to obtain an approximation for fixed n, we obtain

Since X n /n P p, we know that X n (n. Xn (n X n ) Using the asymptotic result above to obtain an approximation for fixed n, we obtain Assigmet 9 Exercise 5.5 Let X biomial, p, where p 0, 1 is ukow. Obtai cofidece itervals for p i two differet ways: a Sice X / p d N0, p1 p], the variace of the limitig distributio depeds oly o p. Use the

More information

7.1 Convergence of sequences of random variables

7.1 Convergence of sequences of random variables Chapter 7 Limit theorems Throughout this sectio we will assume a probability space (Ω, F, P), i which is defied a ifiite sequece of radom variables (X ) ad a radom variable X. The fact that for every ifiite

More information

ECE 8527: Introduction to Machine Learning and Pattern Recognition Midterm # 1. Vaishali Amin Fall, 2015

ECE 8527: Introduction to Machine Learning and Pattern Recognition Midterm # 1. Vaishali Amin Fall, 2015 ECE 8527: Itroductio to Machie Learig ad Patter Recogitio Midterm # 1 Vaishali Ami Fall, 2015 tue39624@temple.edu Problem No. 1: Cosider a two-class discrete distributio problem: ω 1 :{[0,0], [2,0], [2,2],

More information

Fall 2013 MTH431/531 Real analysis Section Notes

Fall 2013 MTH431/531 Real analysis Section Notes Fall 013 MTH431/531 Real aalysis Sectio 8.1-8. Notes Yi Su 013.11.1 1. Defiitio of uiform covergece. We look at a sequece of fuctios f (x) ad study the coverget property. Notice we have two parameters

More information

Basics of Probability Theory (for Theory of Computation courses)

Basics of Probability Theory (for Theory of Computation courses) Basics of Probability Theory (for Theory of Computatio courses) Oded Goldreich Departmet of Computer Sciece Weizma Istitute of Sciece Rehovot, Israel. oded.goldreich@weizma.ac.il November 24, 2008 Preface.

More information

Application to Random Graphs

Application to Random Graphs A Applicatio to Radom Graphs Brachig processes have a umber of iterestig ad importat applicatios. We shall cosider oe of the most famous of them, the Erdős-Réyi radom graph theory. 1 Defiitio A.1. Let

More information

Statistical Inference (Chapter 10) Statistical inference = learn about a population based on the information provided by a sample.

Statistical Inference (Chapter 10) Statistical inference = learn about a population based on the information provided by a sample. Statistical Iferece (Chapter 10) Statistical iferece = lear about a populatio based o the iformatio provided by a sample. Populatio: The set of all values of a radom variable X of iterest. Characterized

More information

Singular Continuous Measures by Michael Pejic 5/14/10

Singular Continuous Measures by Michael Pejic 5/14/10 Sigular Cotiuous Measures by Michael Peic 5/4/0 Prelimiaries Give a set X, a σ-algebra o X is a collectio of subsets of X that cotais X ad ad is closed uder complemetatio ad coutable uios hece, coutable

More information

Chapter 6 Principles of Data Reduction

Chapter 6 Principles of Data Reduction Chapter 6 for BST 695: Special Topics i Statistical Theory. Kui Zhag, 0 Chapter 6 Priciples of Data Reductio Sectio 6. Itroductio Goal: To summarize or reduce the data X, X,, X to get iformatio about a

More information

IIT JAM Mathematical Statistics (MS) 2006 SECTION A

IIT JAM Mathematical Statistics (MS) 2006 SECTION A IIT JAM Mathematical Statistics (MS) 6 SECTION A. If a > for ad lim a / L >, the which of the followig series is ot coverget? (a) (b) (c) (d) (d) = = a = a = a a + / a lim a a / + = lim a / a / + = lim

More information

Convergence of random variables. (telegram style notes) P.J.C. Spreij

Convergence of random variables. (telegram style notes) P.J.C. Spreij Covergece of radom variables (telegram style otes).j.c. Spreij this versio: September 6, 2005 Itroductio As we kow, radom variables are by defiitio measurable fuctios o some uderlyig measurable space

More information

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.265/15.070J Fall 2013 Lecture 3 9/11/2013. Large deviations Theory. Cramér s Theorem

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.265/15.070J Fall 2013 Lecture 3 9/11/2013. Large deviations Theory. Cramér s Theorem MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.265/5.070J Fall 203 Lecture 3 9//203 Large deviatios Theory. Cramér s Theorem Cotet.. Cramér s Theorem. 2. Rate fuctio ad properties. 3. Chage of measure techique.

More information

1 Convergence in Probability and the Weak Law of Large Numbers

1 Convergence in Probability and the Weak Law of Large Numbers 36-752 Advaced Probability Overview Sprig 2018 8. Covergece Cocepts: i Probability, i L p ad Almost Surely Istructor: Alessadro Rialdo Associated readig: Sec 2.4, 2.5, ad 4.11 of Ash ad Doléas-Dade; Sec

More information

Lecture 19: Convergence

Lecture 19: Convergence Lecture 19: Covergece Asymptotic approach I statistical aalysis or iferece, a key to the success of fidig a good procedure is beig able to fid some momets ad/or distributios of various statistics. I may

More information

(A sequence also can be thought of as the list of function values attained for a function f :ℵ X, where f (n) = x n for n 1.) x 1 x N +k x N +4 x 3

(A sequence also can be thought of as the list of function values attained for a function f :ℵ X, where f (n) = x n for n 1.) x 1 x N +k x N +4 x 3 MATH 337 Sequeces Dr. Neal, WKU Let X be a metric space with distace fuctio d. We shall defie the geeral cocept of sequece ad limit i a metric space, the apply the results i particular to some special

More information

Topic 5 [434 marks] (i) Find the range of values of n for which. (ii) Write down the value of x dx in terms of n, when it does exist.

Topic 5 [434 marks] (i) Find the range of values of n for which. (ii) Write down the value of x dx in terms of n, when it does exist. Topic 5 [44 marks] 1a (i) Fid the rage of values of for which eists 1 Write dow the value of i terms of 1, whe it does eist Fid the solutio to the differetial equatio 1b give that y = 1 whe = π (cos si

More information

Econ 325 Notes on Point Estimator and Confidence Interval 1 By Hiro Kasahara

Econ 325 Notes on Point Estimator and Confidence Interval 1 By Hiro Kasahara Poit Estimator Eco 325 Notes o Poit Estimator ad Cofidece Iterval 1 By Hiro Kasahara Parameter, Estimator, ad Estimate The ormal probability desity fuctio is fully characterized by two costats: populatio

More information

Linear regression. Daniel Hsu (COMS 4771) (y i x T i β)2 2πσ. 2 2σ 2. 1 n. (x T i β y i ) 2. 1 ˆβ arg min. β R n d

Linear regression. Daniel Hsu (COMS 4771) (y i x T i β)2 2πσ. 2 2σ 2. 1 n. (x T i β y i ) 2. 1 ˆβ arg min. β R n d Liear regressio Daiel Hsu (COMS 477) Maximum likelihood estimatio Oe of the simplest liear regressio models is the followig: (X, Y ),..., (X, Y ), (X, Y ) are iid radom pairs takig values i R d R, ad Y

More information

Lecture 14: Graph Entropy

Lecture 14: Graph Entropy 15-859: Iformatio Theory ad Applicatios i TCS Sprig 2013 Lecture 14: Graph Etropy March 19, 2013 Lecturer: Mahdi Cheraghchi Scribe: Euiwoog Lee 1 Recap Bergma s boud o the permaet Shearer s Lemma Number

More information

Non-Asymptotic Achievable Rates for Gaussian Energy-Harvesting Channels: Best-Effort and Save-and-Transmit

Non-Asymptotic Achievable Rates for Gaussian Energy-Harvesting Channels: Best-Effort and Save-and-Transmit 08 IEEE Iteratioal Symposium o Iformatio Theory ISIT No-Asymptotic Achievable Rates for Gaussia Eergy-Harvestig Chaels: Best-Effort ad Save-ad-Trasmit Silas L Fog Departmet of Electrical ad Computer Egieerig

More information

ECONOMETRIC THEORY. MODULE XIII Lecture - 34 Asymptotic Theory and Stochastic Regressors

ECONOMETRIC THEORY. MODULE XIII Lecture - 34 Asymptotic Theory and Stochastic Regressors ECONOMETRIC THEORY MODULE XIII Lecture - 34 Asymptotic Theory ad Stochastic Regressors Dr. Shalabh Departmet of Mathematics ad Statistics Idia Istitute of Techology Kapur Asymptotic theory The asymptotic

More information

Introduction to Machine Learning DIS10

Introduction to Machine Learning DIS10 CS 189 Fall 017 Itroductio to Machie Learig DIS10 1 Fu with Lagrage Multipliers (a) Miimize the fuctio such that f (x,y) = x + y x + y = 3. Solutio: The Lagragia is: L(x,y,λ) = x + y + λ(x + y 3) Takig

More information

arxiv: v1 [cs.it] 13 Jul 2012

arxiv: v1 [cs.it] 13 Jul 2012 O the Sum Capacity of the Discrete Memoryless Iterferece Chael with Oe-Sided Weak Iterferece ad Mixed Iterferece Fagfag Zhu ad Biao Che Syracuse Uiversity Departmet of EECS Syracuse, NY 3244 Email: fazhu{biche}@syr.edu

More information

The variance of a sum of independent variables is the sum of their variances, since covariances are zero. Therefore. V (xi )= n n 2 σ2 = σ2.

The variance of a sum of independent variables is the sum of their variances, since covariances are zero. Therefore. V (xi )= n n 2 σ2 = σ2. SAMPLE STATISTICS A radom sample x 1,x,,x from a distributio f(x) is a set of idepedetly ad idetically variables with x i f(x) for all i Their joit pdf is f(x 1,x,,x )=f(x 1 )f(x ) f(x )= f(x i ) The sample

More information

It is always the case that unions, intersections, complements, and set differences are preserved by the inverse image of a function.

It is always the case that unions, intersections, complements, and set differences are preserved by the inverse image of a function. MATH 532 Measurable Fuctios Dr. Neal, WKU Throughout, let ( X, F, µ) be a measure space ad let (!, F, P ) deote the special case of a probability space. We shall ow begi to study real-valued fuctios defied

More information

Stochastic Simulation

Stochastic Simulation Stochastic Simulatio 1 Itroductio Readig Assigmet: Read Chapter 1 of text. We shall itroduce may of the key issues to be discussed i this course via a couple of model problems. Model Problem 1 (Jackso

More information

32 estimating the cumulative distribution function

32 estimating the cumulative distribution function 32 estimatig the cumulative distributio fuctio 4.6 types of cofidece itervals/bads Let F be a class of distributio fuctios F ad let θ be some quatity of iterest, such as the mea of F or the whole fuctio

More information

1.010 Uncertainty in Engineering Fall 2008

1.010 Uncertainty in Engineering Fall 2008 MIT OpeCourseWare http://ocw.mit.edu.00 Ucertaity i Egieerig Fall 2008 For iformatio about citig these materials or our Terms of Use, visit: http://ocw.mit.edu.terms. .00 - Brief Notes # 9 Poit ad Iterval

More information

Chapter 3. Strong convergence. 3.1 Definition of almost sure convergence

Chapter 3. Strong convergence. 3.1 Definition of almost sure convergence Chapter 3 Strog covergece As poited out i the Chapter 2, there are multiple ways to defie the otio of covergece of a sequece of radom variables. That chapter defied covergece i probability, covergece i

More information

4. Partial Sums and the Central Limit Theorem

4. Partial Sums and the Central Limit Theorem 1 of 10 7/16/2009 6:05 AM Virtual Laboratories > 6. Radom Samples > 1 2 3 4 5 6 7 4. Partial Sums ad the Cetral Limit Theorem The cetral limit theorem ad the law of large umbers are the two fudametal theorems

More information

Definition 4.2. (a) A sequence {x n } in a Banach space X is a basis for X if. unique scalars a n (x) such that x = n. a n (x) x n. (4.

Definition 4.2. (a) A sequence {x n } in a Banach space X is a basis for X if. unique scalars a n (x) such that x = n. a n (x) x n. (4. 4. BASES I BAACH SPACES 39 4. BASES I BAACH SPACES Sice a Baach space X is a vector space, it must possess a Hamel, or vector space, basis, i.e., a subset {x γ } γ Γ whose fiite liear spa is all of X ad

More information

Lecture Chapter 6: Convergence of Random Sequences

Lecture Chapter 6: Convergence of Random Sequences ECE5: Aalysis of Radom Sigals Fall 6 Lecture Chapter 6: Covergece of Radom Sequeces Dr Salim El Rouayheb Scribe: Abhay Ashutosh Doel, Qibo Zhag, Peiwe Tia, Pegzhe Wag, Lu Liu Radom sequece Defiitio A ifiite

More information

Limit distributions for products of sums

Limit distributions for products of sums Statistics & Probability Letters 62 (23) 93 Limit distributios for products of sums Yogcheg Qi Departmet of Mathematics ad Statistics, Uiversity of Miesota-Duluth, Campus Ceter 4, 7 Uiversity Drive, Duluth,

More information

Math 220B Final Exam Solutions March 18, 2002

Math 220B Final Exam Solutions March 18, 2002 Math 0B Fial Exam Solutios March 18, 00 1. (1 poits) (a) (6 poits) Fid the Gree s fuctio for the tilted half-plae {(x 1, x ) R : x 1 + x > 0}. For x (x 1, x ), y (y 1, y ), express your Gree s fuctio G(x,

More information

Econ 325/327 Notes on Sample Mean, Sample Proportion, Central Limit Theorem, Chi-square Distribution, Student s t distribution 1.

Econ 325/327 Notes on Sample Mean, Sample Proportion, Central Limit Theorem, Chi-square Distribution, Student s t distribution 1. Eco 325/327 Notes o Sample Mea, Sample Proportio, Cetral Limit Theorem, Chi-square Distributio, Studet s t distributio 1 Sample Mea By Hiro Kasahara We cosider a radom sample from a populatio. Defiitio

More information

INFINITE SEQUENCES AND SERIES

INFINITE SEQUENCES AND SERIES INFINITE SEQUENCES AND SERIES INFINITE SEQUENCES AND SERIES I geeral, it is difficult to fid the exact sum of a series. We were able to accomplish this for geometric series ad the series /[(+)]. This is

More information

Sequences and Limits

Sequences and Limits Chapter Sequeces ad Limits Let { a } be a sequece of real or complex umbers A ecessary ad sufficiet coditio for the sequece to coverge is that for ay ɛ > 0 there exists a iteger N > 0 such that a p a q

More information

Lecture 3 : Random variables and their distributions

Lecture 3 : Random variables and their distributions Lecture 3 : Radom variables ad their distributios 3.1 Radom variables Let (Ω, F) ad (S, S) be two measurable spaces. A map X : Ω S is measurable or a radom variable (deoted r.v.) if X 1 (A) {ω : X(ω) A}

More information

Multiterminal source coding with complementary delivery

Multiterminal source coding with complementary delivery Iteratioal Symposium o Iformatio Theory ad its Applicatios, ISITA2006 Seoul, Korea, October 29 November 1, 2006 Multitermial source codig with complemetary delivery Akisato Kimura ad Tomohiko Uyematsu

More information

1 Duality revisited. AM 221: Advanced Optimization Spring 2016

1 Duality revisited. AM 221: Advanced Optimization Spring 2016 AM 22: Advaced Optimizatio Sprig 206 Prof. Yaro Siger Sectio 7 Wedesday, Mar. 9th Duality revisited I this sectio, we will give a slightly differet perspective o duality. optimizatio program: f(x) x R

More information

Sequences A sequence of numbers is a function whose domain is the positive integers. We can see that the sequence

Sequences A sequence of numbers is a function whose domain is the positive integers. We can see that the sequence Sequeces A sequece of umbers is a fuctio whose domai is the positive itegers. We ca see that the sequece 1, 1, 2, 2, 3, 3,... is a fuctio from the positive itegers whe we write the first sequece elemet

More information

Monte Carlo Integration

Monte Carlo Integration Mote Carlo Itegratio I these otes we first review basic umerical itegratio methods (usig Riema approximatio ad the trapezoidal rule) ad their limitatios for evaluatig multidimesioal itegrals. Next we itroduce

More information

1 Approximating Integrals using Taylor Polynomials

1 Approximating Integrals using Taylor Polynomials Seughee Ye Ma 8: Week 7 Nov Week 7 Summary This week, we will lear how we ca approximate itegrals usig Taylor series ad umerical methods. Topics Page Approximatig Itegrals usig Taylor Polyomials. Defiitios................................................

More information

A statistical method to determine sample size to estimate characteristic value of soil parameters

A statistical method to determine sample size to estimate characteristic value of soil parameters A statistical method to determie sample size to estimate characteristic value of soil parameters Y. Hojo, B. Setiawa 2 ad M. Suzuki 3 Abstract Sample size is a importat factor to be cosidered i determiig

More information

Lesson 10: Limits and Continuity

Lesson 10: Limits and Continuity www.scimsacademy.com Lesso 10: Limits ad Cotiuity SCIMS Academy 1 Limit of a fuctio The cocept of limit of a fuctio is cetral to all other cocepts i calculus (like cotiuity, derivative, defiite itegrals

More information

Real Variables II Homework Set #5

Real Variables II Homework Set #5 Real Variables II Homework Set #5 Name: Due Friday /0 by 4pm (at GOS-4) Istructios: () Attach this page to the frot of your homework assigmet you tur i (or write each problem before your solutio). () Please

More information

Comparison Study of Series Approximation. and Convergence between Chebyshev. and Legendre Series

Comparison Study of Series Approximation. and Convergence between Chebyshev. and Legendre Series Applied Mathematical Scieces, Vol. 7, 03, o. 6, 3-337 HIKARI Ltd, www.m-hikari.com http://d.doi.org/0.988/ams.03.3430 Compariso Study of Series Approimatio ad Covergece betwee Chebyshev ad Legedre Series

More information