On the multivariate conditional probability density of a vector perturbed by Gaussian noise
|
|
- Damian Brown
- 6 years ago
- Views:
Transcription
1 IEEE TRANS. ON INFORMATION THEORY VOL. X NO. XX XXXX On the multivariate conditional probability density of a vector perturbed by Gaussian noise Yuriy S. Shmaliy Senior Member IEEE Abstract This paper examines the joint conditional probability density function (pdf) of the main variables (envelope phase and their n- order time derivatives) of a time-varying random signal in the presence of additive Gaussian noise. The main variables are conditioned with respect to the given variables which are the amplitude phase and their derivatives of the signal alone. We prove a theorem stating that some of the conditional pdfs of the main variables do not depend on some of the given variables. This theorem together with Bayes s theorem can substantially simplify the derivations of conditional pdfs and give alternative forms of them. Both theorems can also help in finding reasonable approximations as we demonstrate for the phase and first time derivative of the envelope. Index Terms Joint probability density function Phase Random signal Time derivative of the envelope Time-varying vector. I. INTRODUCTION Statistical properties of an information bearing narrowband signal perturbed by noise have been studied for decades beginning with the early works of Rice. In spite of this as will become clear in the sequel one of the important problems still remains unsolved. Most generally a signal is represented in the form of s(t) S(t) cosω t + ϑ(t)] U c(t) cos ω t U s(t) sin ω t () where U c U cos ϑ U s U sin ϑ S(t) is an instantaneous power and ω is an angular carrier frequency. Here U(t) S(t) and ϑ(t) are the signal amplitude and phase respectively. A common case is that at a receiver () is perturbed by narrowband Gaussian noise whose model is ξ(t) A(t) cosω t + φ(t)] A c(t) cos ω t A s(t) sin ω t () where A c A cos φ and A s A sin φ are orthogonal low-pass stationary and zero-mean Gaussian processes with equal variances σ σ c σ s. Also it is supposed that ξ(t) is continuous and multiply differentiable. Both s(t) and ξ(t) are mixed at the receiver additively so that the signal becomes noisy v(t) V (t) cosω t + ϕ(t)] (3a) V c(t) cos ω t V s(t) sin ω t (3b) (U c + A c) cos ω t (U s + A s) sin ω t (3c) where V c V cos ϕ V s V sin ϕ V (t) is the envelope and ϕ(t) is the phase. Equating the amplitudes of the harmonic functions in (3b) and (3c) produces the Gaussian variables A c(t) V (t) cos ϕ(t) U(t) cos ϑ(t) (4) A s(t) V (t) sin ϕ(t) U(t) sin ϑ(t) (5) allowing the investigation of statistical properties of V ϕ and their n-order time derivatives. The procedure begins with the joint probability density function (pdf) of A c A s and their n-order time Manuscript received...; revised... Yu. S. Shmaliy is with the Guanajuato University FIMEE Salamanca Gto. Mexico ( shmaliy@salamanca.ugto.mx). Also with Kharkiv National University of Radio Electronics Ukraine. Fig.. Time behaviors of the cosine components U c and V c of s(t) and v(t) respectively: (I) unmodulated signal and (II) modulated signal. derivatives thereafter transferring to that for V ϕ and their n-order time derivatives. When the unnecessary variables are integrated out the pdf becomes marginal representing the required variable. By Shannon s philosophy ] the marginal pdf is commonly used to estimate differential entropy ] 3] at a fixed time instant of a given variable with noisy signaling fading and fluctuations in communication channels radars wireless systems etc. II. MULTIVARIATE CONDITIONAL PDF AND PROBLEM FORMULATION With zero signal the joint pdf representes Gaussian noise that was considered by many authors 4] 9]. If a signal exists U c and U s can be supposed to be either time-invariant (unmodulated) or timevarying (modulated). In view of that two different behaviors of a signal can be recognized as shown in Fig. for cosine components. In the most general case represented by the region (II) U c and U s are modulated and their n-order time derivatives at some point t commonly exist. If these amplitudes are random at t then U ϑ and their n-order time derivatives appear as given variables in the joint pdf 8] 3] 4] 6] so that the latter can be written as p(v V V... ϕ ϕ ϕ... U U Ü... ϑ ϑ ϑ...) (6) In the region (I) both U c and U s are constant although randomly valued. Therefore all of the time derivatives are removed in (6) from the given variables. This case was analyzed in 8] ] ]. If the marginal pdfs of each of the given variables p(u) p( U)... p(ϑ) p( ϑ)... are supposed to be known the joint unconditional pdf is obtained via (6) by... p(v V... ϕ ϕ...) p(v V... ϕ ϕ... U U... ϑ ϑ...) p(u)p( U)... p(ϑ)p( ϑ)... dud U... dϑd ϑ... (7) If s(t) is deterministic then each of the given variables U U... ϑ ϑ... has a delta-shaped pdf; that is p(u U... ϑ ϑ...) δ(u U )δ( U U ) δ(ϑ ϑ )δ( ϑ ϑ ). By the sifting property of the delta function δ(x) all of the deterministic quantities are commonly accounted for as coefficients in the joint pdf. Now assume that we have a complete general probabilistic picture of v(t) sketched by (6) and would like to obtain for example the marginal pdf of V. If we integrate out in (6) all other variables we formally get a marginal pdf that is conditional
2 IEEE TRANS. ON INFORMATION THEORY VOL. X NO. XX XXXX... p(v U U... ϑ ϑ...) p(v V... ϕ ϕ... U U... ϑ ϑ...) d V... dϕd ϕ... On the other hand we know that the envelope V has the Rice pdf p(v U) if the only given variable is U. Note that Rice derived his pdf for a time-invariant vector. To the best of our knowledge the pdf of V was never derived from (6). The question then arises whether the conditional pdf of V with respect to all the given variables has a form that is more general than the Rice p(v U)? Certainly a similar question can be asked for other variables. The problem now formulates as follows. Given the general joint conditional pdf (6) we would like to know whether the marginal pdf of each of the main variables conditioned on all the given variables depends only on a subset of the given variables. The rest of the paper is organized as follows. In Section III we derive a six-variable (V V V ϕ ϕ ϕ ) joint conditional pdf of a modulated random signal in the presence of Gaussian noise; this pdf does not seem to have been published before. By integrating out step-by-step the high-order time derivatives of main variables we arrive at an important rule that is generalized in the theorem of Section IV. As examples of applications in Section V we use the theorem to derive alternative and approximate pdf forms for ϕ and V. Finally concluding remarks are drawn in Section VI. III. SIX VARIABLE JOINT CONDITIONAL PDF Information bearing signals are typically characterized by the first three terms in the Taylor series expansion of the envelope and phase. In phase channels for instance of prime importance are statistical properties of the phase ϕ (or time delay) linear phase drift ϕ (Doppler frequency) and linear frequency drift ϕ (Doppler rate) 5] 6]. If a signal is randomly valued (by signalling fading and/or fluctuations) and perturbed by Gaussian noise at the receiver it is described by (6) having six main variables (V V V ϕ ϕ ϕ) and six given variables (U U Ü ϑ ϑ ϑ). Let us find this pdf. It is known 7] that at a fixed time instant the Gaussian variables A c da c/dt and A s da s/dt are not correlated with each other and also are not correlated with A c A s Ä c d A c/dt and Äs d A s/dt. However A c Ä c and A s Ä s are mutually correlated. Therefore for a symmetric one-sided power spectral density ( (PSD) S ξ (f) of the Gaussian noise the covariance matrix of A c A ) c Äc As A s Äs can be written as R() σ σ ρ σ σ ρ σ ρ σ ρ σ ρ σ ρ (4) σ ρ σ ρ (4) (8) where the time derivatives of the normalized envelope ρ(τ) R(τ)/σ of the correlation function of ξ(t) ρ d ρ(τ) dτ τ and ρ (4) d 4 ρ(τ)/dτ 4 τ are defined by 7] respectively ρ σ σ b b < ρ (4) σ σ b4 b >. Here σ and σ are variances of the first and second time derivatives of ξ(t) and the mth spectral moment of ξ(t) is calculated by 5] b m (π) m (f f ) m S ξ (f)df where m is an integer. By (8) the joint pdf of the Gaussian variables is written as p(a c A c Äc As A s Äs) p( A c)p( A s)p(a c Äc)p(As Äs) e σ g h(a s +A c )+q(ȧ s +Ȧ c ) ρ (Ä s +Ä c )+ ρ (AsÄs+AcÄc)] (πσ ) 3 ( g) (9) where g ρ(ρ (4) ρ ) σ (σ σ σ) 4 < σ 6 q ρ (4) ρ σ σ σ 4 > h ρ ρ (4) σ 4 σ σ <. σ 4 Using the well-known trick of transferring to the polar coordinates V ϕ and their time derivatives 4] we define the determinant of the Jacobian of the transformation (Ac A s A c A det ] s Äc Äs) (V ϕ V ϕ V V 3 ϕ) and derive the required pdf p(z ż z ϕ ϕ ϕ γ γ γ ϑ ϑ ϑ) π 3 ( ρ)(ρ (4) ρ ) ρ ρ (4) γ ρ(γ ϑ + γ + γ ϑ 4 +4 exp ρ(ρ (4) ρ ) ϑ ϑ γ γ + 4 ϑ γ ϑ γ γ) +(ρ (4) ρ )(γ ϑ + γ) + ρ ( γ γ γ ϑ ) ρ ρ (4) z + (ρ (4) ρ )ż + (ρ (4) 3 ρ )z ϕ ρ ( z + 4ż ϕ z z ϕ + 4zż ϕ ϕ + z ϕ + z ϕ 4 ) + ρ z z + ρ ( ρ γ ϑ ρ γ +ρ (4) γ)z + ρ( ρ γ γ + ϑ γ)(z ϕ z) + ρ ( γ ϑ + γ ϑ)(ż ϕ + z ϕ) (ρ (4) ρ ) exp (4) ρ(ρ (4) ρ ) γ ϑz ϕ (ρ ρ ) γż] cos(ϕ ϑ) + ρ ( γ ϑ + γ ϑ)z + ρ ( ρ γ + γ ϑ γ) (ż ϕ + z ϕ) + ρ ( γ ϑ + γ ϑ) ( z z ϕ ) (ρ (4) ρ ) γ ϑż +(ρ (4) ρ ) γz ϕ] sin(ϕ ϑ) () in which z V/ σ ż V / σ z V / σ γ(t) U (t) σ z 3 S(t) σ () is the instantaneous signal-to-noise ratio (SNR) γ U /σ and γ Ü /σ. Several particular cases of () have earlier been observed in the literature. The case of γ γ γ ϑ ϑ ϑ corresponding to zero signal returns us to the works of Rice 4] ] and some more recent papers 6] 8] 9] in which the unconditional joint pdf p(z ż z ϕ ϕ ϕ) of a narrowband Gaussian noise was of concern. The case of γ and ϑ ω was studied in 8] in the presence
3 IEEE TRANS. ON INFORMATION THEORY VOL. X NO. XX XXXX 3 of Gaussian noise with an arbitrary PSD shape. In 8] we also find the pdf for a symmetric one-sided noise PSD that is identical to the relevant degenerate version of (). For our purposes we now integrate out first z and ϕ and then ż and ϕ. Lemma : Given the six variable joint conditional pdf (). Integrating out ϕ and z yields the four variable joint conditional pdf p(z ż ϕ ϕ γ γ ϑ ϑ) z π ( ρ exp ρ ) ( ρ γ γ ϑ ] γ) exp ( ρ ) z ϕ ρ z + ż ( γ ϑz ϕ + γż ρ γz) cos(ϕ ϑ) +( γz ϕ γ ϑż) sin(ϕ ϑ). () Proof: First integrate () over ϕ from to. Thereafter by an identity ( ) e px ±qx π q dx p exp p > (3) 4p integrate over z from to and arrive at () that was earlier derived in 4]. Lemma : Given the four variable conditional pdf (). Integrating out ϕ and ż leads to the bivariate joint conditional pdf p(z ϕ γ ϑ) z γ+ γz cos(ϕ ϑ) π e z. (4) Proof: Integrate () by (3) over ϕ from to and thereafter over ż from to. Finally arrive at (4). Theorem : Given the joint conditional pdf (7) of the envelope z phase ϕ and their n-order time derivatives z (n) (t) dz (n) (t)/dt (n) and ϕ (n) (t) dϕ (n) (t)/dt (n) n N] respectively of a timevarying signal perturbed by Gaussian noise with randomly valued instantaneous SNR γ and phase ϑ and their n-order time derivatives γ (n) (t) dγ (n) (t)/dt (n) and ϑ (n) (t) dϑ (n) (t)/dt (n) n N] respectively. Then the conditional pdf... dz (N M+)... dz (N)... dϕ (N K+)... dϕ (N) } {{ } } {{ } M K p(z z ()... z (N) ϕ ϕ ()... ϕ (N) γ γ ()... γ (N) ϑ ϑ ()... ϑ (N) ) (8) does not depend on γ (N L+)... γ N and ϑ (N L+)... ϑ N where L min (M K). Proof: Let L M K. Substituting in (7) N by N L yields p(z z ()... z (N L) ϕ ϕ ()... ϕ (N L) γ γ ()... γ (N L) ϑ ϑ ()... ϑ (N L) ). (9) On the other hand by L M K (8) formally produces p(z z ()... z (N L) ϕ ϕ ()... ϕ (N L) γ γ ()... γ (N) ϑ ϑ ()... ϑ (N) ). () Now using the integral representation of the modified Bessel function of the first kind and zeroth order I (x) π π π ex cos(ϕ ϑ)dϕ and integrating (4) over ϕ from π to π lead to the conditional Rice pdf of the envelope ] p(z γ) ze z γ I ( γz) (5) that turns out to be independent on ϑ. On the other hand an identity ( xe px +qx dx + q π p p eq /4p Φ q p )] where Φ(x) π x e t / dt is the probability integral transforms (4) to the conditional Bennett s pdf of the random phase mod π ] p(ϕ γ ϑ) e γ γ π + sin (ϕ ϑ) ] e γ Φ γ cos(ϕ ϑ) cos(ϕ ϑ) (6) π that is conditional on both γ and ϑ. An important inference follows instantly. Integrating p(z ż z ϕ ϕ ϕ γ γ γ ϑ ϑ ϑ) over z and ϕ yields p(z ż ϕ ϕ γ γ ϑ ϑ) without γ and ϑ. Further integrating p(z ż ϕ ϕ γ γ ϑ ϑ ) over ż and ϕ produces p(z ϕ γ ϑ) without γ and ϑ. Note that the terms with γ ϑ γ and ϑ are compensated while integrating (). Below we extend this observation to the multivariate joint conditional pdf (6) proving an important theorem. IV. THEOREM Consider the multivariate joint conditional pdf () for the n-order time derivatives n N] of all of the main and given variables; that is p(z z ()... z (N) ϕ ϕ ()... ϕ (N) γ γ ()... γ (N) ϑ ϑ ()... ϑ (N) ). (7) An identity of (9) and () exists if and only if the variables γ (N L+)... γ N ϑ (N L+)... ϑ N vanish in () and the proof for M K is complete. The proof is supported by Lemma and Lemma. If K M then integrating out one of the excess variables z (n) or ϕ (n) does not necessarily remove the corresponding given variable and we still rely on the minimum of M and K. In other words the theorem states the following: In the joint pdf the maximum order of the time derivatives of given variables cannot exceed the maximum order of the time derivatives of main variables; that is for example p(ż ϕ γ γ ϑ ϑ) but not p(ż ϕ γ γ γ ϑ ϑ ϑ) p(z ϕ γ ϑ) but not p(z ϕ γ γ ϑ ϑ). (a) (b) In the marginal pdf the maximum order of the time derivatives of given variables cannot exceed the order of the time derivative of the main variable. For example p( ϕ γ γ ϑ ϑ) but not p( ϕ γ γ γ ϑ ϑ ϑ) (c) p(z γ) but not p(z γ γ ϑ ϑ) p(ϕ γ ϑ) but not p(ϕ γ γ ϑ ϑ). (d) (e) This means by extension that the Rice pdf p(z γ) and Bennett pdf p(ϕ γ ϑ) are both unique for any signal () and their marginal pdfs are explicitly specified by respectively p(ϕ) p(z) dγ π π dγ p(z γ)p(γ) dϑ p(ϕ γ ϑ)p(ϑ)p(γ) (a) (b) where p(γ) represents random variations in the SNR and p(ϑ) in the signal phase.
4 IEEE TRANS. ON INFORMATION THEORY VOL. X NO. XX XXXX 4 V. APPLICATIONS Below we apply Theorem to derive several alternative and approximate pdf forms for ϕ and ż. A. An Alternative Form of Bennett s pdf By Theorem and (d) an exhaustive form of the conditional Rice s pdf is p(z γ). By virtue of that the Bayes s theorem offers via (4) an alternative form of the conditional Bennett s pdf (6); that is p(ϕ z γ ϑ) p(z ϕ γ ϑ) p(z γ ϑ) p(z ϕ γ ϑ) p(z γ) πi γ cos(ϕ ϑ) (z γ) ez (3) in which z is Ricean (5). Equation (3) is the von Mises/Tikhonov conditional circular normal pdf 7] with a random parameter. Substituting z with its mean value 8] 8] π z(γ) ( ) F ; γ (4) where F (a b; x) is a degenerate hypergeometric function leads to an approximation p a(ϕ γ ϑ) πi e z γ cos(ϕ ϑ) (5) ( z γ) exhibiting a maximum fractional error p a(ϕ) p(ϕ)]/p(ϑ) of about 4.7 % at γ. The error decreases when γ. B. Alternative pdf forms for ż Brown showed in 9] and Isley proved in ] that the pdf of ż of an unmodulated signal with Gaussian noise is normal. However for the modulated and random signal the relevant pdf cannot be found in closed form. Instead Theorem offers several useful solutions. Let us first use (3) and integrate (4) over ϕ from to p(z ż ϕ γ γ ϑ ϑ) exp ( ρ ) z π π ρ exp ( ρ ϑ )γ γ ρ ż ρ z ( γż ρ γz) cos(ϕ ϑ) γ ϑż sin(ϕ ϑ) + ϑ γ γ sin (ϕ ϑ) + ( γ γ ϑ ) cos (ϕ ϑ) Then integrating (6) over ϕ from π to π by a series e ±a cos ϕ I (a) + (±) n I n(a) cos nϕ n ]. (6) yields p(z ż γ γ ϑ ϑ) +γ) ze (z exp ż + γ + γ ϑ ] π( ρ) ( ρ ) ε ni n(r )I n(r ) cos n(ψ ψ ) (7) n { if n where ε n if n > R r ( ρ ) R r ( ρ ) r ( γż ρ γz) + γ ϑ ż r 4 ϑ γ γ + ( γ γ ϑ ) ψ arcsin γ ϑż r and ψ arcsin ϑ γ γ r. Now by the Bayes s theorem we have p(z ż γ γ ϑ ϑ) p(z γ γ ϑ ϑ)p(ż z γ γ ϑ ϑ) and by Theorem and (d) we must replace p(z γ γ ϑ ϑ) by p(z γ). The conditional pdf of ż can then easily be found to be p(ż z γ γ ϑ ϑ) p(z ż γ γ ϑ ϑ) p(z γ) π( ρ)i( γz) exp ż + γ + γ ϑ ] ( ρ ) ε ni n(r )I n(r ) cos n(ψ ψ ) (8) n in which z is Ricean ((5). Like the case of (5) the mean value (4) offers for (8) a nice approximation p a(ż γ γ ϑ ϑ) π( ρ)i( γ z) exp ż + γ + γ ϑ ] ( ρ ) ε ni n(r )I n(r ) cos n(ψ ψ ) (9) n which maximum fractional error p a(ż) p(ż)]/p( ż) is several percent; that is about.9% at γ γ 3 ϑ.7 and ϑ. One can also exploit the fact that by Theorem and (b) the bivariate conditional pdf of z and ϕ is exhaustively represented by p(z ϕ γ ϑ). The Bayes s theorem then easily produces by (3) and (4) p(ż ϕ z ϕ γ γ ϑ ϑ) p(z ż ϕ ϕ γ γ ϑ ϑ) p(z ϕ γ γ ϑ ϑ) exp ( ρ ) p(z ż ϕ ϕ γ γ ϑ ϑ) p(z ϕ γ ϑ) z π( ρ) e γ ϑ + γ ρ ż + z ϕ ( γ ϑz ϕ + γż) cos(ϕ ϑ) +( γz ϕ γ ϑż) sin(ϕ ϑ). (3) Now integrating (3) over ϕ from to by () leads to the conditional normal pdf p(ż ϕ γ γ ϑ ϑ) ] ż γ + γ ϑ cos(ϕ ϑ ς) exp π( ρ) ( ρ ) (3) where ς arccos γ/( γ + γ ϑ ) with the variance ( ρ )/ and a random mean value. Surprisingly (3) does not depend on z. With γ ϑ the pdf (3) acquires a zero mean and readily degenerates to the normal pdf found by Brown 9] and Isley ]. The phase ϕ in (3) has the Bennett distribution (6) or approximately the von Mises/Tikhonov distribution (5). In the latter case the following alternative approximation can be useful π π p b (ż γ γ ϑ ϑ) πi ( z γ) π( ρ ) e + ż γ+γ cos(ϕ ϑ ς)] ϑ ρ z γ cos(ϕ ϑ) ( ρ ) dϕ (3) allowing like the case of (9) a maximum fractional error of several percent.
5 IEEE TRANS. ON INFORMATION THEORY VOL. X NO. XX XXXX 5 VI. CONCLUDING REMARKS An important merit of the above-proved Theorem resides in the fact that it is fundamental for modulated and random narrowband signals perturbed by additive Gaussian noise with an arbitrary PSD shape. The theorem ascertains which of the given variables the joint and marginal pdfs of the main variables (envelope phase and their n-order time derivatives) can depend on. This theorem together with Bayes s theorem can substantially simplify the derivations of conditional pdfs and give alternative forms of them. A nice illustration is the normal conditional pdf with a random mean value derived for ż (3). Both theorems can also help in finding reasonable approximations as we demonstrated by deriving (9) and (3) for ż. ] C. T. Isley Probability distribution for the derivative of the envelope of signal and Gaussian noise IRE Trans. Inform. Theory vol. IT-6 pp Mar. 96. ACKNOWLEDGMENT The author would like to thank Dr. Charles Greenhall of the Jet Propulsion Laboratory (JPL) California Institute of Technology for assistance in reading and discussions of the results and two anonymous reviewers for valuable comments and remarks. REFERENCES ] C.E. Shannon A mathematical theory of communication Bell Syst. Tech. J. vol. 7 pp ] S.W. Golomb The information generating function of a probability distribution IEEE Trans. Inform. Theory vol. IT- pp Jen ] A.C.G. Verdugo Lazo and P.N. Rathie On the entropy of continuous probability distributions IEEE Trans. Inform. Theory vol. IT-4 no. pp. - Jen ] S.O. Rice Mathematical analysis of random noise in Selected Papers on Noise and Stochastic Processes. N. Wax Ed. New York: Dover 954 pp ] D. Middleton Spurious signals caused by noise in triggered circuits J. Applied Physics vol. 9 no. 9 pp Sep ] B.N. Zvyaghintsev Some statistical properties of the second derivative of the envelope of a normal random process Radio Eng. Elektron. Phys. vol. no. 4 pp Apr ] V.I. Tikhonov Statistical Radio Engineering Moscow: Sovetskoe Radio ] V.I. Tikhonov Nonlinear transformations of random processes Moscow: Radio i sviaz ] N.M. Blackman The distribution of local extrema of Gaussian noise and of its envelope IEEE Trans. Inform. Theory vol. 45. no. 6 pp. 5- Sep ] S.O. Rice Statistical properties of a sine wave plus random noise Bell Syst. Tech. J. vol. 7 no. pp Jan ] W.R. Bennett Methods of solving noise problems Proc. IRE vol. 44 pp May 956. ] V.V. Tsvetnov Statistical properties of signals and noises in twochannel phase systems Radiotekhn. vol. no. 5 pp. -9 May ] J. Salz and S. Stein Distribution of instantaneous frequency for signal plus noise IEEE Trans. Inform. Theory vol. IT- pp Oct ] N.G. Gatkin V.A. Garanin M.I. Karnovskiy L.G. Krasniy and N.I. Cherney Probability density of phase derivative of the sum of a modulated signal and Gaussian noise Radio Eng. Electron. Phys. vol. no. 8 pp. 3-9 Aug ] P. Bello Joint estimation of delay Doppler and Doppler rate IRE Trans. Inform. Theory vol. IT-6 no. 3 pp June 96. 6] Yu.S. Shmaliy Phase Errors of Passive Wireless SAW Sensors with Differential Measurement IEEE Sensors J. vol. 4 no. 6 pp Dec. 4. 7] Yu.S. Shmaliy Von Mises/Tikhonov-based distributions for systems with differential phase measurement Signal Processing vol. 85 no. 4 pp Apr. 5. 8] T. H. Park Moments of the generalized Rayleigh distribution Quart. Appl. Math. vol. 9 no. pp Jan ] W. M. Brown Some results on noise through circuits IRE Trans. Inform. Theory vol. IT-5 pp May 959.
3F1 Random Processes Examples Paper (for all 6 lectures)
3F Random Processes Examples Paper (for all 6 lectures). Three factories make the same electrical component. Factory A supplies half of the total number of components to the central depot, while factories
More informationOn the Average Crossing Rates in Selection Diversity
PREPARED FOR IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS (ST REVISION) On the Average Crossing Rates in Selection Diversity Hong Zhang, Student Member, IEEE, and Ali Abdi, Member, IEEE Abstract This letter
More informationStochastic Processes. M. Sami Fadali Professor of Electrical Engineering University of Nevada, Reno
Stochastic Processes M. Sami Fadali Professor of Electrical Engineering University of Nevada, Reno 1 Outline Stochastic (random) processes. Autocorrelation. Crosscorrelation. Spectral density function.
More informationTHE problem of phase noise and its influence on oscillators
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II: EXPRESS BRIEFS, VOL. 54, NO. 5, MAY 2007 435 Phase Diffusion Coefficient for Oscillators Perturbed by Colored Noise Fergal O Doherty and James P. Gleeson Abstract
More informationAn Analysis of Errors in RFID SAW-Tag Systems with Pulse Position Coding
An Analysis of Errors in RFID SAW-Tag Systems with Pulse Position Coding YURIY S. SHMALIY, GUSTAVO CERDA-VILLAFAÑA, OSCAR IBARRA-MANZANO Guanajuato University, Department of Electronics Salamanca, 36855
More informationECE6604 PERSONAL & MOBILE COMMUNICATIONS. Week 3. Flat Fading Channels Envelope Distribution Autocorrelation of a Random Process
1 ECE6604 PERSONAL & MOBILE COMMUNICATIONS Week 3 Flat Fading Channels Envelope Distribution Autocorrelation of a Random Process 2 Multipath-Fading Mechanism local scatterers mobile subscriber base station
More informationDetecting Parametric Signals in Noise Having Exactly Known Pdf/Pmf
Detecting Parametric Signals in Noise Having Exactly Known Pdf/Pmf Reading: Ch. 5 in Kay-II. (Part of) Ch. III.B in Poor. EE 527, Detection and Estimation Theory, # 5c Detecting Parametric Signals in Noise
More information2.7 The Gaussian Probability Density Function Forms of the Gaussian pdf for Real Variates
.7 The Gaussian Probability Density Function Samples taken from a Gaussian process have a jointly Gaussian pdf (the definition of Gaussian process). Correlator outputs are Gaussian random variables if
More informationDUE to its practical importance in communications, the
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II: EXPRESS BRIEFS, VOL. 52, NO. 3, MARCH 2005 149 An Analytical Formulation of Phase Noise of Signals With Gaussian-Distributed Jitter Reza Navid, Student Member,
More informationEEL 5544 Noise in Linear Systems Lecture 30. X (s) = E [ e sx] f X (x)e sx dx. Moments can be found from the Laplace transform as
L30-1 EEL 5544 Noise in Linear Systems Lecture 30 OTHER TRANSFORMS For a continuous, nonnegative RV X, the Laplace transform of X is X (s) = E [ e sx] = 0 f X (x)e sx dx. For a nonnegative RV, the Laplace
More informationIntroduction to Probability and Stocastic Processes - Part I
Introduction to Probability and Stocastic Processes - Part I Lecture 2 Henrik Vie Christensen vie@control.auc.dk Department of Control Engineering Institute of Electronic Systems Aalborg University Denmark
More informationFormulas for probability theory and linear models SF2941
Formulas for probability theory and linear models SF2941 These pages + Appendix 2 of Gut) are permitted as assistance at the exam. 11 maj 2008 Selected formulae of probability Bivariate probability Transforms
More informationECE6604 PERSONAL & MOBILE COMMUNICATIONS. Week 4. Envelope Correlation Space-time Correlation
ECE6604 PERSONAL & MOBILE COMMUNICATIONS Week 4 Envelope Correlation Space-time Correlation 1 Autocorrelation of a Bandpass Random Process Consider again the received band-pass random process r(t) = g
More informationOPTIMAL SENSOR-TARGET GEOMETRIES FOR DOPPLER-SHIFT TARGET LOCALIZATION. Ngoc Hung Nguyen and Kutluyıl Doğançay
OPTIMAL SENSOR-TARGET GEOMETRIES FOR DOPPLER-SHIFT TARGET LOCALIZATION Ngoc Hung Nguyen and Kutluyıl Doğançay Institute for Telecommunications Research, School of Engineering University of South Australia,
More informationProblems on Discrete & Continuous R.Vs
013 SUBJECT NAME SUBJECT CODE MATERIAL NAME MATERIAL CODE : Probability & Random Process : MA 61 : University Questions : SKMA1004 Name of the Student: Branch: Unit I (Random Variables) Problems on Discrete
More informationPerformance Analysis of Spread Spectrum CDMA systems
1 Performance Analysis of Spread Spectrum CDMA systems 16:33:546 Wireless Communication Technologies Spring 5 Instructor: Dr. Narayan Mandayam Summary by Liang Xiao lxiao@winlab.rutgers.edu WINLAB, Department
More informationEAS 305 Random Processes Viewgraph 1 of 10. Random Processes
EAS 305 Random Processes Viewgraph 1 of 10 Definitions: Random Processes A random process is a family of random variables indexed by a parameter t T, where T is called the index set λ i Experiment outcome
More informationTIME-VARIANT RELIABILITY ANALYSIS FOR SERIES SYSTEMS WITH LOG-NORMAL VECTOR RESPONSE
TIME-VARIANT RELIABILITY ANALYSIS FOR SERIES SYSTEMS WITH LOG-NORMAL VECTOR RESONSE Sayan Gupta, ieter van Gelder, Mahesh andey 2 Department of Civil Engineering, Technical University of Delft, The Netherlands
More informationENGR352 Problem Set 02
engr352/engr352p02 September 13, 2018) ENGR352 Problem Set 02 Transfer function of an estimator 1. Using Eq. (1.1.4-27) from the text, find the correct value of r ss (the result given in the text is incorrect).
More informationBivariate distributions
Bivariate distributions 3 th October 017 lecture based on Hogg Tanis Zimmerman: Probability and Statistical Inference (9th ed.) Bivariate Distributions of the Discrete Type The Correlation Coefficient
More informationMultiple Random Variables
Multiple Random Variables Joint Probability Density Let X and Y be two random variables. Their joint distribution function is F ( XY x, y) P X x Y y. F XY ( ) 1, < x
More informationELEMENTS OF PROBABILITY THEORY
ELEMENTS OF PROBABILITY THEORY Elements of Probability Theory A collection of subsets of a set Ω is called a σ algebra if it contains Ω and is closed under the operations of taking complements and countable
More informationMAS223 Statistical Inference and Modelling Exercises
MAS223 Statistical Inference and Modelling Exercises The exercises are grouped into sections, corresponding to chapters of the lecture notes Within each section exercises are divided into warm-up questions,
More informationconditional cdf, conditional pdf, total probability theorem?
6 Multiple Random Variables 6.0 INTRODUCTION scalar vs. random variable cdf, pdf transformation of a random variable conditional cdf, conditional pdf, total probability theorem expectation of a random
More informationEnvelope PDF in Multipath Fading Channels with Random Number of Paths and Nonuniform Phase Distributions
Envelope PDF in Multipath Fading Channels with andom umber of Paths and onuniform Phase Distributions ALI ABDI AD MOSTAFA KAVEH DEPT. OF ELEC. AD COMP. EG., UIVESITY OF MIESOTA 4-74 EE/CSCI BLDG., UIO
More informationChapter 4: Continuous channel and its capacity
meghdadi@ensil.unilim.fr Reference : Elements of Information Theory by Cover and Thomas Continuous random variable Gaussian multivariate random variable AWGN Band limited channel Parallel channels Flat
More informationGAUSSIAN PROCESS REGRESSION
GAUSSIAN PROCESS REGRESSION CSE 515T Spring 2015 1. BACKGROUND The kernel trick again... The Kernel Trick Consider again the linear regression model: y(x) = φ(x) w + ε, with prior p(w) = N (w; 0, Σ). The
More informationIntelligent Embedded Systems Uncertainty, Information and Learning Mechanisms (Part 1)
Advanced Research Intelligent Embedded Systems Uncertainty, Information and Learning Mechanisms (Part 1) Intelligence for Embedded Systems Ph. D. and Master Course Manuel Roveri Politecnico di Milano,
More informationEEM 409. Random Signals. Problem Set-2: (Power Spectral Density, LTI Systems with Random Inputs) Problem 1: Problem 2:
EEM 409 Random Signals Problem Set-2: (Power Spectral Density, LTI Systems with Random Inputs) Problem 1: Consider a random process of the form = + Problem 2: X(t) = b cos(2π t + ), where b is a constant,
More informationOptimum Transmission Scheme for a MISO Wireless System with Partial Channel Knowledge and Infinite K factor
Optimum Transmission Scheme for a MISO Wireless System with Partial Channel Knowledge and Infinite K factor Mai Vu, Arogyaswami Paulraj Information Systems Laboratory, Department of Electrical Engineering
More informationReliability Theory of Dynamic Loaded Structures (cont.) Calculation of Out-Crossing Frequencies Approximations to the Failure Probability.
Outline of Reliability Theory of Dynamic Loaded Structures (cont.) Calculation of Out-Crossing Frequencies Approximations to the Failure Probability. Poisson Approximation. Upper Bound Solution. Approximation
More informationCommunication Systems Lecture 21, 22. Dong In Kim School of Information & Comm. Eng. Sungkyunkwan University
Communication Systems Lecture 1, Dong In Kim School of Information & Comm. Eng. Sungkyunkwan University 1 Outline Linear Systems with WSS Inputs Noise White noise, Gaussian noise, White Gaussian noise
More informationGaussian processes. Chuong B. Do (updated by Honglak Lee) November 22, 2008
Gaussian processes Chuong B Do (updated by Honglak Lee) November 22, 2008 Many of the classical machine learning algorithms that we talked about during the first half of this course fit the following pattern:
More information144 Chapter 3. Second Order Linear Equations
144 Chapter 3. Second Order Linear Equations PROBLEMS In each of Problems 1 through 8 find the general solution of the given differential equation. 1. y + 2y 3y = 0 2. y + 3y + 2y = 0 3. 6y y y = 0 4.
More informationOn the estimation of the K parameter for the Rice fading distribution
On the estimation of the K parameter for the Rice fading distribution Ali Abdi, Student Member, IEEE, Cihan Tepedelenlioglu, Student Member, IEEE, Mostafa Kaveh, Fellow, IEEE, and Georgios Giannakis, Fellow,
More informationParameter Estimation
1 / 44 Parameter Estimation Saravanan Vijayakumaran sarva@ee.iitb.ac.in Department of Electrical Engineering Indian Institute of Technology Bombay October 25, 2012 Motivation System Model used to Derive
More informationPhysics 403. Segev BenZvi. Propagation of Uncertainties. Department of Physics and Astronomy University of Rochester
Physics 403 Propagation of Uncertainties Segev BenZvi Department of Physics and Astronomy University of Rochester Table of Contents 1 Maximum Likelihood and Minimum Least Squares Uncertainty Intervals
More informationCity, University of London Institutional Repository
City Research Online City, University of London Institutional Repository Citation: Zhao, S., Shmaliy, Y. S., Khan, S. & Liu, F. (2015. Improving state estimates over finite data using optimal FIR filtering
More informationOn the Ratio of Rice Random Variables
JIRSS 9 Vol. 8, Nos. 1-, pp 61-71 On the Ratio of Rice Random Variables N. B. Khoolenjani 1, K. Khorshidian 1, 1 Departments of Statistics, Shiraz University, Shiraz, Iran. n.b.khoolenjani@gmail.com Departments
More information04. Random Variables: Concepts
University of Rhode Island DigitalCommons@URI Nonequilibrium Statistical Physics Physics Course Materials 215 4. Random Variables: Concepts Gerhard Müller University of Rhode Island, gmuller@uri.edu Creative
More informationThe statistics of ocean-acoustic ambient noise
The statistics of ocean-acoustic ambient noise Nicholas C. Makris Naval Research Laboratory, Washington, D.C. 0375, USA Abstract With the assumption that the ocean-acoustic ambient noise field is a random
More informationRevision of Lecture 4
Revision of Lecture 4 We have completed studying digital sources from information theory viewpoint We have learnt all fundamental principles for source coding, provided by information theory Practical
More informationfor valid PSD. PART B (Answer all five units, 5 X 10 = 50 Marks) UNIT I
Code: 15A04304 R15 B.Tech II Year I Semester (R15) Regular Examinations November/December 016 PROBABILITY THEY & STOCHASTIC PROCESSES (Electronics and Communication Engineering) Time: 3 hours Max. Marks:
More informationEE4601 Communication Systems
EE4601 Communication Systems Week 2 Review of Probability, Important Distributions 0 c 2011, Georgia Institute of Technology (lect2 1) Conditional Probability Consider a sample space that consists of two
More informationPart IA Probability. Definitions. Based on lectures by R. Weber Notes taken by Dexter Chua. Lent 2015
Part IA Probability Definitions Based on lectures by R. Weber Notes taken by Dexter Chua Lent 2015 These notes are not endorsed by the lecturers, and I have modified them (often significantly) after lectures.
More informationCommunication Theory II
Communication Theory II Lecture 8: Stochastic Processes Ahmed Elnakib, PhD Assistant Professor, Mansoura University, Egypt March 5 th, 2015 1 o Stochastic processes What is a stochastic process? Types:
More informationIntroduction to Probability and Stochastic Processes I
Introduction to Probability and Stochastic Processes I Lecture 3 Henrik Vie Christensen vie@control.auc.dk Department of Control Engineering Institute of Electronic Systems Aalborg University Denmark Slides
More informationBrief Review of Probability
Brief Review of Probability Nuno Vasconcelos (Ken Kreutz-Delgado) ECE Department, UCSD Probability Probability theory is a mathematical language to deal with processes or experiments that are non-deterministic
More informationVirtual Array Processing for Active Radar and Sonar Sensing
SCHARF AND PEZESHKI: VIRTUAL ARRAY PROCESSING FOR ACTIVE SENSING Virtual Array Processing for Active Radar and Sonar Sensing Louis L. Scharf and Ali Pezeshki Abstract In this paper, we describe how an
More informationDefinition of a Stochastic Process
Definition of a Stochastic Process Balu Santhanam Dept. of E.C.E., University of New Mexico Fax: 505 277 8298 bsanthan@unm.edu August 26, 2018 Balu Santhanam (UNM) August 26, 2018 1 / 20 Overview 1 Stochastic
More informationResidual Versus Suppressed-Carrier Coherent Communications
TDA Progress Report -7 November 5, 996 Residual Versus Suppressed-Carrier Coherent Communications M. K. Simon and S. Million Communications and Systems Research Section This article addresses the issue
More informationPerhaps the simplest way of modeling two (discrete) random variables is by means of a joint PMF, defined as follows.
Chapter 5 Two Random Variables In a practical engineering problem, there is almost always causal relationship between different events. Some relationships are determined by physical laws, e.g., voltage
More information13.42 READING 6: SPECTRUM OF A RANDOM PROCESS 1. STATIONARY AND ERGODIC RANDOM PROCESSES
13.42 READING 6: SPECTRUM OF A RANDOM PROCESS SPRING 24 c A. H. TECHET & M.S. TRIANTAFYLLOU 1. STATIONARY AND ERGODIC RANDOM PROCESSES Given the random process y(ζ, t) we assume that the expected value
More informationLecture 2. Spring Quarter Statistical Optics. Lecture 2. Characteristic Functions. Transformation of RVs. Sums of RVs
s of Spring Quarter 2018 ECE244a - Spring 2018 1 Function s of The characteristic function is the Fourier transform of the pdf (note Goodman and Papen have different notation) C x(ω) = e iωx = = f x(x)e
More informationG.PULLAIAH COLLEGE OF ENGINEERING & TECHNOLOGY DEPARTMENT OF ELECTRONICS & COMMUNICATION ENGINEERING PROBABILITY THEORY & STOCHASTIC PROCESSES
G.PULLAIAH COLLEGE OF ENGINEERING & TECHNOLOGY DEPARTMENT OF ELECTRONICS & COMMUNICATION ENGINEERING PROBABILITY THEORY & STOCHASTIC PROCESSES LECTURE NOTES ON PTSP (15A04304) B.TECH ECE II YEAR I SEMESTER
More informationApplications of the compensated compactness method on hyperbolic conservation systems
Applications of the compensated compactness method on hyperbolic conservation systems Yunguang Lu Department of Mathematics National University of Colombia e-mail:ylu@unal.edu.co ALAMMI 2009 In this talk,
More informationInformation geometry for bivariate distribution control
Information geometry for bivariate distribution control C.T.J.Dodson + Hong Wang Mathematics + Control Systems Centre, University of Manchester Institute of Science and Technology Optimal control of stochastic
More informationPROPERTIES OF THE CO-ENERGY FUNCTION FOR AC MACHINES WITH NON-LINEAR MAGNETIC CIRCUIT
PERIODICA POLYTECHNICA SER. EL. ENG. VOL. 45, NO. 3 4, PP. 223 233 (2001) PROPERTIES OF THE CO-ENERGY FUNCTION FOR AC MACHINES WITH NON-LINEAR MAGNETIC CIRCUIT Tadeusz J. SOBCZYK Institute on Electromechanical
More informationOptimum Passive Beamforming in Relation to Active-Passive Data Fusion
Optimum Passive Beamforming in Relation to Active-Passive Data Fusion Bryan A. Yocom Applied Research Laboratories The University of Texas at Austin Final Project EE381K-14 Multidimensional Digital Signal
More informationEXAMINATIONS OF THE ROYAL STATISTICAL SOCIETY
EXAMINATIONS OF THE ROYAL STATISTICAL SOCIETY GRADUATE DIPLOMA, 011 MODULE 3 : Stochastic processes and time series Time allowed: Three Hours Candidates should answer FIVE questions. All questions carry
More informationES 272 Assignment #2. in,3
ES 272 Assignment #2 Due: March 14th, 2014; 5pm sharp, in the dropbox outside MD 131 (Donhee Ham office) Instructor: Donhee Ham (copyright c 2014 by D. Ham) (Problem 1) The kt/c Noise (50pt) Imagine an
More informationPerturbation theory for the defocusing nonlinear Schrödinger equation
Perturbation theory for the defocusing nonlinear Schrödinger equation Theodoros P. Horikis University of Ioannina In collaboration with: M. J. Ablowitz, S. D. Nixon and D. J. Frantzeskakis Outline What
More informationECE276A: Sensing & Estimation in Robotics Lecture 10: Gaussian Mixture and Particle Filtering
ECE276A: Sensing & Estimation in Robotics Lecture 10: Gaussian Mixture and Particle Filtering Lecturer: Nikolay Atanasov: natanasov@ucsd.edu Teaching Assistants: Siwei Guo: s9guo@eng.ucsd.edu Anwesan Pal:
More informationMTH739U/P: Topics in Scientific Computing Autumn 2016 Week 6
MTH739U/P: Topics in Scientific Computing Autumn 16 Week 6 4.5 Generic algorithms for non-uniform variates We have seen that sampling from a uniform distribution in [, 1] is a relatively straightforward
More information5 Analog carrier modulation with noise
5 Analog carrier modulation with noise 5. Noisy receiver model Assume that the modulated signal x(t) is passed through an additive White Gaussian noise channel. A noisy receiver model is illustrated in
More informationSome asymptotic properties of solutions for Burgers equation in L p (R)
ARMA manuscript No. (will be inserted by the editor) Some asymptotic properties of solutions for Burgers equation in L p (R) PAULO R. ZINGANO Abstract We discuss time asymptotic properties of solutions
More information2. (a) What is gaussian random variable? Develop an equation for guassian distribution
Code No: R059210401 Set No. 1 II B.Tech I Semester Supplementary Examinations, February 2007 PROBABILITY THEORY AND STOCHASTIC PROCESS ( Common to Electronics & Communication Engineering, Electronics &
More informationLinear Dynamical Systems
Linear Dynamical Systems Sargur N. srihari@cedar.buffalo.edu Machine Learning Course: http://www.cedar.buffalo.edu/~srihari/cse574/index.html Two Models Described by Same Graph Latent variables Observations
More informationA proof of the existence of good nested lattices
A proof of the existence of good nested lattices Dinesh Krithivasan and S. Sandeep Pradhan July 24, 2007 1 Introduction We show the existence of a sequence of nested lattices (Λ (n) 1, Λ(n) ) with Λ (n)
More informationFor a stochastic process {Y t : t = 0, ±1, ±2, ±3, }, the mean function is defined by (2.2.1) ± 2..., γ t,
CHAPTER 2 FUNDAMENTAL CONCEPTS This chapter describes the fundamental concepts in the theory of time series models. In particular, we introduce the concepts of stochastic processes, mean and covariance
More informationRandomly Modulated Periodic Signals
Randomly Modulated Periodic Signals Melvin J. Hinich Applied Research Laboratories University of Texas at Austin hinich@mail.la.utexas.edu www.la.utexas.edu/~hinich Rotating Cylinder Data Fluid Nonlinearity
More informationCHAPTER 14. Based on the info about the scattering function we know that the multipath spread is T m =1ms, and the Doppler spread is B d =0.2 Hz.
CHAPTER 4 Problem 4. : Based on the info about the scattering function we know that the multipath spread is T m =ms, and the Doppler spread is B d =. Hz. (a) (i) T m = 3 sec (ii) B d =. Hz (iii) ( t) c
More informationImage Denoising using Uniform Curvelet Transform and Complex Gaussian Scale Mixture
EE 5359 Multimedia Processing Project Report Image Denoising using Uniform Curvelet Transform and Complex Gaussian Scale Mixture By An Vo ISTRUCTOR: Dr. K. R. Rao Summer 008 Image Denoising using Uniform
More informationEffect of Multipath Propagation on the Noise Performance of Zero Crossing Digital Phase Locked Loop
International Journal of Electronics and Communication Engineering. ISSN 0974-2166 Volume 9, Number 2 (2016), pp. 97-104 International Research Publication House http://www.irphouse.com Effect of Multipath
More informationPreliminary statistics
1 Preliminary statistics The solution of a geophysical inverse problem can be obtained by a combination of information from observed data, the theoretical relation between data and earth parameters (models),
More informationModern Navigation. Thomas Herring
12.215 Modern Navigation Thomas Herring Estimation methods Review of last class Restrict to basically linear estimation problems (also non-linear problems that are nearly linear) Restrict to parametric,
More informationTitle. Author(s)Tsai, Shang-Ho. Issue Date Doc URL. Type. Note. File Information. Equal Gain Beamforming in Rayleigh Fading Channels
Title Equal Gain Beamforming in Rayleigh Fading Channels Author(s)Tsai, Shang-Ho Proceedings : APSIPA ASC 29 : Asia-Pacific Signal Citationand Conference: 688-691 Issue Date 29-1-4 Doc URL http://hdl.handle.net/2115/39789
More informationECE531: Principles of Detection and Estimation Course Introduction
ECE531: Principles of Detection and Estimation Course Introduction D. Richard Brown III WPI 22-January-2009 WPI D. Richard Brown III 22-January-2009 1 / 37 Lecture 1 Major Topics 1. Web page. 2. Syllabus
More informationDigital Band-pass Modulation PROF. MICHAEL TSAI 2011/11/10
Digital Band-pass Modulation PROF. MICHAEL TSAI 211/11/1 Band-pass Signal Representation a t g t General form: 2πf c t + φ t g t = a t cos 2πf c t + φ t Envelope Phase Envelope is always non-negative,
More informationSOLUTIONS TO HOMEWORK ASSIGNMENT 1
ECE 559 Fall 007 Handout # 4 Septemer 0, 007 SOLUTIONS TO HOMEWORK ASSIGNMENT. Vanderkulk s Lemma. The complex random variale Z X + jy is zero mean and Gaussian ut not necessarily proper. Show that E expjνz
More informationA GENERALISED (M, N R ) MIMO RAYLEIGH CHANNEL MODEL FOR NON- ISOTROPIC SCATTERER DISTRIBUTIONS
A GENERALISED (M, N R MIMO RAYLEIGH CHANNEL MODEL FOR NON- ISOTROPIC SCATTERER DISTRIBUTIONS David B. Smith (1, Thushara D. Abhayapala (2, Tim Aubrey (3 (1 Faculty of Engineering (ICT Group, University
More informationLecture Notes 1 Probability and Random Variables. Conditional Probability and Independence. Functions of a Random Variable
Lecture Notes 1 Probability and Random Variables Probability Spaces Conditional Probability and Independence Random Variables Functions of a Random Variable Generation of a Random Variable Jointly Distributed
More informationChapter 5 Random Variables and Processes
Chapter 5 Random Variables and Processes Wireless Information Transmission System Lab. Institute of Communications Engineering National Sun Yat-sen University Table of Contents 5.1 Introduction 5. Probability
More informationA conjecture on sustained oscillations for a closed-loop heat equation
A conjecture on sustained oscillations for a closed-loop heat equation C.I. Byrnes, D.S. Gilliam Abstract The conjecture in this paper represents an initial step aimed toward understanding and shaping
More informationLecture Notes 1 Probability and Random Variables. Conditional Probability and Independence. Functions of a Random Variable
Lecture Notes 1 Probability and Random Variables Probability Spaces Conditional Probability and Independence Random Variables Functions of a Random Variable Generation of a Random Variable Jointly Distributed
More informationProbability and Distributions
Probability and Distributions What is a statistical model? A statistical model is a set of assumptions by which the hypothetical population distribution of data is inferred. It is typically postulated
More informationREVEAL. Receiver Exploiting Variability in Estimated Acoustic Levels Project Review 16 Sept 2008
REVEAL Receiver Exploiting Variability in Estimated Acoustic Levels Project Review 16 Sept 2008 Presented to Program Officers: Drs. John Tague and Keith Davidson Undersea Signal Processing Team, Office
More informationHomework 1 Due: Thursday 2/5/2015. Instructions: Turn in your homework in class on Thursday 2/5/2015
10-704 Homework 1 Due: Thursday 2/5/2015 Instructions: Turn in your homework in class on Thursday 2/5/2015 1. Information Theory Basics and Inequalities C&T 2.47, 2.29 (a) A deck of n cards in order 1,
More informationContinuous Wave Data Analysis: Fully Coherent Methods
Continuous Wave Data Analysis: Fully Coherent Methods John T. Whelan School of Gravitational Waves, Warsaw, 3 July 5 Contents Signal Model. GWs from rotating neutron star............ Exercise: JKS decomposition............
More informationwhere r n = dn+1 x(t)
Random Variables Overview Probability Random variables Transforms of pdfs Moments and cumulants Useful distributions Random vectors Linear transformations of random vectors The multivariate normal distribution
More informationProbabilistic Graphical Models
Probabilistic Graphical Models Brown University CSCI 2950-P, Spring 2013 Prof. Erik Sudderth Lecture 12: Gaussian Belief Propagation, State Space Models and Kalman Filters Guest Kalman Filter Lecture by
More informationOn the Multivariate Nakagami-m Distribution With Exponential Correlation
1240 IEEE TRANSACTIONS ON COMMUNICATIONS, VOL. 51, NO. 8, AUGUST 2003 On the Multivariate Nakagami-m Distribution With Exponential Correlation George K. Karagiannidis, Member, IEEE, Dimitris A. Zogas,
More informationMultiple Random Variables
Multiple Random Variables This Version: July 30, 2015 Multiple Random Variables 2 Now we consider models with more than one r.v. These are called multivariate models For instance: height and weight An
More informationSwitching Regime Estimation
Switching Regime Estimation Series de Tiempo BIrkbeck March 2013 Martin Sola (FE) Markov Switching models 01/13 1 / 52 The economy (the time series) often behaves very different in periods such as booms
More informationPROBABILITY AND RANDOM PROCESSESS
PROBABILITY AND RANDOM PROCESSESS SOLUTIONS TO UNIVERSITY QUESTION PAPER YEAR : JUNE 2014 CODE NO : 6074 /M PREPARED BY: D.B.V.RAVISANKAR ASSOCIATE PROFESSOR IT DEPARTMENT MVSR ENGINEERING COLLEGE, NADERGUL
More informationStochastic nonlinear Schrödinger equations and modulation of solitary waves
Stochastic nonlinear Schrödinger equations and modulation of solitary waves A. de Bouard CMAP, Ecole Polytechnique, France joint work with R. Fukuizumi (Sendai, Japan) Deterministic and stochastic front
More informationECE 541 Stochastic Signals and Systems Problem Set 11 Solution
ECE 54 Stochastic Signals and Systems Problem Set Solution Problem Solutions : Yates and Goodman,..4..7.3.3.4.3.8.3 and.8.0 Problem..4 Solution Since E[Y (t] R Y (0, we use Theorem.(a to evaluate R Y (τ
More informationECE 636: Systems identification
ECE 636: Systems identification Lectures 3 4 Random variables/signals (continued) Random/stochastic vectors Random signals and linear systems Random signals in the frequency domain υ ε x S z + y Experimental
More informationLab10: FM Spectra and VCO
Lab10: FM Spectra and VCO Prepared by: Keyur Desai Dept. of Electrical Engineering Michigan State University ECE458 Lab 10 What is FM? A type of analog modulation Remember a common strategy in analog modulation?
More informationIT IS well known that in mobile communications, a profound
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, VOL. 54, NO. 4, JULY 5 159 A Study on the Second Order Statistics of Nakagami-Hoyt Mobile Fading Channels Neji Youssef, Member, IEEE, Cheng-Xiang Wang, Student
More information