Relationship between Continuous-Time and Discrete-Time Systems in Stochastic Processes, Assignments of Cryptography and System Security

The connection between the continuous-time system and the discrete-time system in the context of stochastic processes. The text derives the relationship between the invariant probability vectors and the transition matrices of both systems, and discusses the physical interpretation of this result. Additionally, it covers the stability condition for the discrete-time markov chain.

Typology: Assignments

2019/2020

Uploaded on 06/08/2020

tiuw
tiuw 🇺🇸

4.7

(18)

286 documents

1 / 1

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Math 171 Homework #5
3.10
The continuous-time system obeys the eigenvector equation πA = 0. Ex-
panded out, this gives
X
k6=x
π(k)α(k, x)π(x)α(x) = 0
Solving for π(x) yields
π(x) = 1
α(x)X
k6=x
π(k)α(k, x)
Let the invariant probability vector for the discrete-time system be φ(x) and
the transition matrix be Bx,y =α(x, y)(x). Then φ(x) satisfies φB =φ:
φ(x) = X
k6=x
φ(k)α(k, x)
α(k)
Divide the above equation by α(x) to get
φ(x)
α(x)=1
α(x)X
k6=x
φ(k)
α(k)α(k, x)
Comparing this last equation with the equation for π(x) we see that π(x) =
φ(x)(x).
This has a nice physical interperetation. It says that the amount of time
spent in state x(π(x)) is the number of visits to x(φ(x)) times the length
of time to leave state x(1(x)).
3.12
We require that
X
n=1
µ1· · · µn
λ1· · · λn
<
Because λn= and µn=, we have
X
n=1
n!λn
n!µn<
X
n=1
λ
µn
<
which is true when λ < µ.

Partial preview of the text

Download Relationship between Continuous-Time and Discrete-Time Systems in Stochastic Processes and more Assignments Cryptography and System Security in PDF only on Docsity!

Math 171 Homework #

The continuous-time system obeys the eigenvector equation πA = 0. Ex- panded out, this gives ∑

k 6 =x

π(k)α(k, x) − π(x)α(x) = 0

Solving for π(x) yields

π(x) =

α(x)

k 6 =x

π(k)α(k, x)

Let the invariant probability vector for the discrete-time system be φ(x) and the transition matrix be Bx,y = α(x, y)/α(x). Then φ(x) satisfies φB = φ:

φ(x) =

k 6 =x

φ(k)

α(k, x) α(k)

Divide the above equation by α(x) to get

φ(x) α(x)

α(x)

k 6 =x

φ(k) α(k)

α(k, x)

Comparing this last equation with the equation for π(x) we see that π(x) = φ(x)/α(x).

This has a nice physical interperetation. It says that the amount of time spent in state x (π(x)) is the number of visits to x (φ(x)) times the length of time to leave state x (1/α(x)).

We require that

∑^ ∞

n=

μ 1 · · · μn λ 1 · · · λn

Because λn = nλ and μn = nμ, we have

∑^ ∞

n=

n!λn n!μn^

∑^ ∞

n=

( (^) λ μ

)n < ∞

which is true when λ < μ.