Index the queue-class pairs by . Let be the external Poisson process arrival rate, let be the routing probability after class completes service, and put for the exit probability. An open network has a transient routing matrix , equivalently spectral radius less than one. The effective throughputs solve the traffic equations for a multiclass queueing network:
At each queue use the ordered multiclass single-server queue model of part (a), with parameters . An external class- customer is inserted at position with rate . A customer of class in position completes at rate : it exits with probability or changes to class and is inserted at its destination with probability . Here is the destination population after the origin deletion, so this also defines routes returning to the same queue. At each node, either service rates are common to its classes or insertion and service fractions form a symmetric service discipline.
Let . The product-form stationary distribution of a multiclass queueing network is
To prove it by time reversal of a continuous-time Markov chain, interchange each node's insertion and service fractions, and use
Pairs with can be deleted. The traffic equations for a multiclass queueing network ensure . The ratio of the product weights for an insertion is ; for a deletion it is its reciprocal. Consequently every routing, entrance, and exit transition satisfies . The symmetric or common-rate hypothesis makes each node's total service rate identical forward and backward. Summing the traffic equations for a multiclass queueing network also gives . Thus the total transition rates match, proving global balance for a continuous-time Markov chain; each factor is normalized by part (a).
For the class counts the equilibrium law is
Under processor sharing these counts have explicit transition rates for , for , and for . Under a closed network, remove external entrances and exits and normalize the same product weights on the conserved customer populations, using a positive solution of the homogeneous traffic equations for a multiclass queueing network. That finite-state normalization does not require the open-network conditions .
A multiclass queue must specify its service discipline as well as its arrival and service rates. Here is an ordered-queue model that includes the usual class-independent exponential single-server queue. Let class- arrivals be independent Poisson processes of rates , and let a class- service requirement have the exponential distribution of rate . A state is a finite word of customer classes.
Choose service fractions with , and insertion probabilities with . The customer at position departs at rate ; a class- arrival is inserted at position at rate . All rates leading to the same word are added. For class-dependent , impose the symmetric service discipline . If all , arbitrary class-independent are permitted, including first come first served with insertion at the tail and service at the head.
Put and . For , the equilibrium word law is
For the symmetric service discipline, insertion and deletion satisfy detailed balance for a continuous-time Markov chain because
For the common-service-rate case, use time reversal of a continuous-time Markov chain instead: the reversed queue inserts with probabilities and serves with fractions . A forward insertion gives a reverse deletion rate , and a forward deletion gives a reverse insertion rate . Forward and reverse total rates both equal , proving global balance for a continuous-time Markov chain.
Summing the weights of all words of length gives , so the law is normalized. If counts class- customers and , the multinomial coefficient counting class orderings gives
For processor sharing, , the class-count vector itself is a Markov chain, with arrival rate and departure rate . Under first come first served, the ordered word is generally necessary. Class-dependent service rates combined with first come first served do not in general have the displayed simple equilibrium law; the discipline assumption matters.
There is now no separate rate- clockwise transition. The only transitions are
Both preserve modulo . The initial state therefore selects the closed communicating class , on which . The population is a birth-death process with birth rate and death rate .
Define and for . The detailed balance for a birth-death process weights are
Writing with , the product condition gives . Summing the resulting geometric series yields
This is also the law in part (b) conditioned on : the ratio of its weights at consecutive accessible populations is . The stability condition is , which uses the product of the periodic birth rates; it does not require for every .
Let denote the clock position and the queue population, including any customer in service. By the memoryless property of the exponential distribution, the nonzero transition rates are
Set . We claim that
where is the clock's stationary distribution from part (a). To check global balance for a continuous-time Markov chain, divide the incoming probability flux by . The clockwise contribution is . The arrival contribution, present only for , is . The service contribution is . Their sum is the outgoing rate .
The proposed probability distribution is normalized, so it is the equilibrium law. In particular, the clock position and queue population are independent random variables at any fixed equilibrium time, and the population has the geometric distribution of a stable M/M/1 queue. Independence of these two random variables does not assert independence of the two entire processes.
Write the clock states modulo . The detailed balance equations are . The product condition makes these equations consistent around the circle, so the stationary distribution is
Indeed, the rate of the time reversal of a continuous-time Markov chain along a clockwise edge is , while its anticlockwise rate is . Thus the clock is a reversible Markov chain.
The reversed clockwise transitions can be driven by a single independent rate- Poisson process, since their rate does not depend on the clock state. They correspond to the original anticlockwise transitions. Consequently the anticlockwise event times form a stationary Poisson process of rate . Its events before a given time are independent of the clock state at that time, which is the relevant quasireversibility property. If clockwise and anticlockwise transitions have the same destination, their direction is retained as an event mark when reversing the process.
To compute the first Tor of a cyclic quotient, start a free resolution of with , take a free module surjecting onto , and take surjecting onto the kernel of . Tensor right exactness identifies
By definition, the degree-one Tor functor is the kernel of modulo the image from . Quotienting first by that image consequently gives the natural identification
Equivalently the calculation gives the exact sequence
Thus its first term vanishes for every ideal exactly when each multiplication map is injective, proving
Together with the preceding implications, all four conditions are equivalent. The proof used explicit extension arguments, character duality and the definition of the Tor functor, rather than an unproved flatness or injectivity criterion.
For an ideal , the canonical multiplication map
has image . Consequently the asserted isomorphism is the natural surjection being injective. This is the flatness criterion using ideals, whose proof follows from the extension criterion above.
Under the character Tensor-hom adjunction, the dual map identifies with
Indeed, for , its restriction corresponds to . If is injective, this restriction is surjective, so the cogenerator criterion makes injective. Conversely, if every is injective, every restriction map in this display is surjective. The proved Baer criterion then makes injective. Thus
In particular, (1) also implies (3) directly by tensoring . No finite-generation condition on , and no Noetherian condition on , has been inserted.
Conversely, if the character module is injective, every restriction map in the preceding display is surjective. The character Tensor-hom adjunction then makes the dual of surjective. Since detects injections, that tensor map is injective for every inclusion . Hence is flat, and
Together these two implications prove the Lambek theorem, with both the duality and the injective-cogenerator facts justified.
A flat module is one for which tensoring preserves injections. This is equivalent to exactness of the tensor functor, since the tensor product of modules is always right exact: its universal balanced maps identify tensoring a cokernel with the cokernel of the tensored map.
Suppose is flat. For every inclusion , the map is injective. The detection property proved above therefore makes its character-dual map surjective. Under the character Tensor-hom adjunction, this is exactly
Every map from a submodule into thus extends, proving
Set and use the accelerated grid time . At population , the fluctuation increment is , , or zero. Its first and second conditional moments, multiplied by , are
Thus and , uniformly on every compact set. The centered conditional variance has the same limit, since the squared conditional mean contributes only .
Here is the diffusion approximation theorem being applied. For Markov chains sampled with time step , suppose the first and second conditional increment moments multiplied by converge uniformly on compact sets to continuous coefficients ; the large-jump conditional second moments satisfy the local Lindeberg condition; the initial laws converge; compact containment holds on finite time intervals; and the limiting martingale problem is well-posed and nonexplosive. Then the interpolated chains converge weakly, uniformly on finite time intervals, to that problem's continuous solution. The step versions converge in the Skorokhod J1 topology. Local moment bounds supply the stopped-process tightness, and the limiting test-function identities identify every subsequential limit by the martingale problem.
We verify the global condition rather than assume it. Before the stopping rule is triggered, , and implies . Therefore
For the Lyapunov function , the exact one-step generator is consequently
After stopping, the process is constant. Applying the elementary stopped supermartingale inequality for , and stopping also on first reaching , gives
This is compact containment. At zero population we use the natural absorbing convention ; the same calculation is valid there.
The invalid-probability threshold is at the positive root
Because , we have . The corresponding fluctuation level tends to infinity, so the stopping rule changes none of the compact-set moment limits. The initial fluctuation satisfies and hence tends to zero. All jumps have magnitude at most , so the Lindeberg condition is immediate. The limiting equation has globally Lipschitz coefficients and is well-posed and nonexplosive. The theorem therefore gives
This is the Ornstein-Uhlenbeck fluctuation limit at a stable population equilibrium. The limiting Ornstein-Uhlenbeck process has explicit solution
The PDF samples the linearly interpolated population at , so its displayed process is actually the step version. Using instead gives the continuous interpolation; their uniform distance is at most . Both therefore have the asserted continuous diffusion limit, with the step version interpreted in Skorokhod J1 topology.
Finally, if , the stopped fluctuation reaches a value greater than . The preceding Lyapunov function bound gives directly
This also shows explicitly why the artificial stopping rule is asymptotically irrelevant.
Use the local convention for the martingale problem, which is appropriate for arbitrary measurable coefficients. A continuous adapted process solves if the coefficient integrals are locally finite and, for every ,
is a local martingale. Here locally finite means almost surely for each . Suitable localization upgrades the identities to true martingale identities. With bounded coefficients and bounded test derivatives, this distinction disappears on finite horizons.
A Markov diffusion generated by is a time-homogeneous Markov process with continuous paths whose infinitesimal generator on the stated test domain is . Its transition operators satisfy the generator limit, and Dynkin formula gives precisely the preceding martingale problem identities. Some terminology uses L-diffusion for the test-function identities alone; the Markov property is an additional assertion unless uniqueness supplies it.
Suppose now . Use smooth compactly supported functions agreeing with and on , and stop before exiting this interval. Removing these stops shows that
is a continuous local martingale. Consequently is a continuous semimartingale. The martingale problem applied to the second cutoff function says that
is a local martingale. The Itô formula for also says that
is a local martingale. Subtracting, the continuous finite-variation process is a local martingale and starts at zero. The theorem that a continuous finite-variation local martingale is constant gives
Define
The integrand is predictable and is stochastically integrable, since its integral against the bracket is exactly . Thus is a continuous local martingale with and . The Lévy characterization of Brownian motion states that these conditions make a Brownian motion relative to the given filtration. Associativity of stochastic integrals recovers , and therefore
The original process is a weak solution of a stochastic differential equation, with its driving Brownian motion constructed on the original space.
Conversely, the Itô formula applied to any weak solution with locally integrable coefficients gives the martingale problem identities. A Markov diffusion with generator gives them by Dynkin formula. A well-posed martingale problem, with the usual measurable path-space setup, identifies a unique family of diffusion laws and yields the Strong Markov property. Without uniqueness, the test-function identities alone do not assert that every chosen solution is Markov. This specifies both the equivalence with weak equations and the role of uniqueness in identifying a diffusion.
Define the quadratic covariation of two continuous semimartingales by polarization:
The available quadratic variation convergence theorem and the polarization identity imply that the sums of products of simultaneous grid increments converge in uniform convergence on compacts in probability to . The quadratic covariation is a continuous finite-variation process, vanishing at time zero.
For a grid ending at , the elementary algebraic identity
telescopes. The first two sums converge by part (a), with the continuous processes as integrands; continuous paths are left-continuous and locally bounded. The last sum converges to the quadratic covariation by the preceding polarization argument. Finally, locally uniformly by continuity. Hence
This is the Itô product rule, or stochastic integration by parts. Uniqueness of the limit in uniform convergence on compacts in probability, followed by continuity, makes the identity hold simultaneously for all outside one null set. The proof used finite-grid algebra, part (a), and the permitted quadratic variation theorem; it did not use the Itô formula.
A left-continuous adapted process is a predictable process. Define the elementary predictable approximation
At every , its sample time approaches from the left, including when is itself a grid point. Thus pathwise by left continuity.
Use the continuous semimartingale decomposition , with a continuous local martingale and a continuous finite-variation process. Localize so that on a fixed horizon , the integrands are bounded by a deterministic constant, is bounded, and the total-variation process of at is bounded. This is possible using local boundedness and the finite pathwise clock and variation.
For the finite-variation integral, the dominated convergence theorem gives
almost surely. For the martingale integral, the Itô isometry and the Doob L2 maximal inequality give
Here the localized bounds justify dominated convergence also in expectation. Remove localization to conclude that the full elementary integrals converge in uniform convergence on compacts in probability to .
The sum of completed intervals differs from the full elementary integral at time by
Its maximum on tends to zero almost surely, by the local bound on and uniform continuity of . Therefore
This proves left-endpoint approximation of a continuous semimartingale integral, including the unfinished interval convention.
Put . The hitting times are finite almost surely by recurrence of planar Brownian motion, and increase with . At , the position is . By the Strong Markov property, Brownian scaling, and rotational invariance of planar Brownian motion, the future process
is a standard planar Brownian motion started at , independent of . The random rotation does not change its conditional law, since that law is rotationally invariant.
The first time this rescaled process reaches radius corresponds to the original time . Its continuous argument starts at zero and equals the increment of the original winding angle. Hence
For any ordered finite collection of levels, apply this assertion successively. Earlier winding values are measurable at the current hitting time, so all increments are independent and their laws depend only on the level differences. This proves the Lévy process property of winding at logarithmic radial passage levels.
One can also verify stochastic continuity and identify this process. The logarithmic-radius and argument martingales have the common clock
The same Cauchy-Riemann equations calculation as in part (a), applied locally to the logarithm and joined along the continuous argument, gives independent Brownian motions such that and . This clock diverges: if it had a finite limit, convergence of martingales with finite total quadratic variation would make converge to a finite value, forcing the clock integral itself to diverge.
Thus, with , we have . The Brownian first-passage Laplace transform is . It follows either from the Exponential martingale for Brownian motion stopped at , or from the one-dimensional exit equation. Conditioning on the independent process gives
So this is a symmetric Cauchy process, with level parameter . The formula implies stochastic continuity at zero and, by stationary increments, everywhere. Under the usual definition requiring càdlàg paths, use the right-continuous passage clock . It agrees with for each fixed level almost surely, and is the càdlàg modification. This distinction concerns simultaneous passage levels and does not alter any increment law.
The exponential map is a conformal bijection from the strip
to the wedge containing the starting point . The restriction makes it injective. A planar Brownian motion started at zero in this strip maps, after the conformal Brownian clock, to the required Brownian motion in the wedge.
Let be the first time exits . It is finite almost surely, and is a bounded martingale. The optional stopping theorem gives
The strip exit maps to the corresponding bounding ray. The exponential clock is finite at , because its integrand is continuous on the compact time interval . Thus it does not remove or reverse the exit event. We obtain
This is the Brownian wedge-exit probability obtained directly from the strip, using conformal invariance of planar Brownian motion.
The conformal invariance of planar Brownian motion is a time-change statement. If is a conformal bijection of planar domains and is planar Brownian motion started in , let be its exit time and define
Before exit this clock is strictly increasing. Its inverse makes a planar Brownian motion started at , up to its transformed lifetime . The image of the stopped trajectory is therefore invariant in law after this change of time. A holomorphic map with nonzero derivative gives the corresponding local statement even without global injectivity.
Here is the proof. Write and . The real and imaginary parts are harmonic. The Itô formula therefore gives the two continuous local martingales
stopped on compact subsets of . The Cauchy-Riemann equations imply
Thus both coordinate quadratic variations equal and their quadratic covariation is zero. After inverse time change, the covariance matrix of the two continuous local martingales is . The multidimensional Lévy characterization of multidimensional Brownian motion states that a continuous vector local martingale starting at zero with this covariance matrix is standard vector Brownian motion. Apply it after subtracting , and exhaust by compact subsets. This proves the theorem, with killing at the transformed exit lifetime. The clock is the conformal Brownian clock; omitting it would generally give an incorrect speed.
Write . The stopped Bessel process has the same law as the weighted stopped Brownian motion, so for every bounded measurable functional of a continuous path,
The Bessel process hits in finite time and never hits zero. Its minimum before that hitting time is strictly positive. Thus for sufficiently small , depending on the path, its stopped path equals exactly. Under Wiener measure, the events increase to , and on that event the stopped paths eventually equal . The remaining term is bounded by . The dominated convergence theorem therefore yields the precise limiting identity
The bounded stopped martingale gives
Hence the limiting identity says
This establishes Brownian conditioning by a stopped Bessel density.
The displayed density in the question requires a stopping convention. On the full, unstopped path space, the probability measure
is well-defined, but its paths continue as Brownian motion after . Its pushforward by the stopping map is the law of the stopped Bessel process. Equivalently, if denotes the law of Brownian motion stopped at , then the actual stopped laws satisfy
Literal absolute continuity of the stopped law with respect to unstopped Wiener measure is false: paths that become permanently constant at have probability one for the stopped law and zero under Wiener measure. The same density is also valid on the stopped sigma-algebra of the original coordinate space. Thus the corrected stopped-law formulation proves the intended conclusion without equating these different full-path measures.
Under Wiener measure, and
The definition of the stochastic exponential gives
On the other hand, the Itô formula applied to the logarithm gives precisely
Their initial values differ by , so
almost surely under Wiener measure. In particular,
For in , the radial function has
The multidimensional Itô formula, initially stopped away from zero, gives
The martingale term has quadratic variation , since . The Lévy characterization of Brownian motion says that a continuous local martingale starting at zero with quadratic variation is a standard Brownian motion. Consequently the radial equation is
On the drift is Lipschitz continuous. Extend it to a bounded globally Lipschitz function on the real line. The standard existence and pathwise uniqueness theorem for stochastic differential equations with globally Lipschitz coefficients applies to this extension, and gives uniqueness in law up to the first exit from . Both this radial process and the process in part (a) solve the same equation from . Thus
where each exit time is computed from its own coordinate path.
For completeness, the radial equation is valid globally: the three-dimensional Bessel process does not hit zero. Apply the Itô formula to on the annulus. Its drift is zero, and the stopped process is a bounded martingale. At the first exit,
so
The exit time is finite because one coordinate of the underlying Brownian motion reaches in finite time. Letting and then considering arbitrarily large proves that zero is never reached. The radial martingale therefore has quadratic variation on the whole time axis, completing the Lévy characterization of Brownian motion argument.
Use the following form of the Girsanov theorem. If is Brownian motion, is predictable, and
is a uniformly integrable martingale with , then the probability measure with terminal density makes a Brownian motion. The analogous finite-horizon statement uses the terminal density at that horizon.
Here take and . Before , , so the stochastic integral is well-defined. The assumed uniform integrability of the stopped stochastic exponential gives
Thus the proposed weighting really defines a probability measure, and the Girsanov theorem gives the Brownian motion
under that measure. In particular
All calculations are first made on this compact stopped interval. The integral involving is naturally defined on ; at its quadratic variation may diverge. The density will have a zero limiting extension there, as established below, rather than a finite stochastic logarithm at the hitting time.

Pinned article: Introduction to the OurBigBook Project

Welcome to the OurBigBook Project! Our goal is to create the perfect publishing platform for STEM subjects, and get university-level students to write the best free STEM tutorials ever.
Everyone is welcome to create an account and play with the site: ourbigbook.com/go/register. We belive that students themselves can write amazing tutorials, but teachers are welcome too. You can write about anything you want, it doesn't have to be STEM or even educational. Silly test content is very welcome and you won't be penalized in any way. Just keep it legal!
We have two killer features:
  1. topics: topics group articles by different users with the same title, e.g. here is the topic for the "Fundamental Theorem of Calculus" ourbigbook.com/go/topic/fundamental-theorem-of-calculus
    Articles of different users are sorted by upvote within each article page. This feature is a bit like:
    • a Wikipedia where each user can have their own version of each article
    • a Q&A website like Stack Overflow, where multiple people can give their views on a given topic, and the best ones are sorted by upvote. Except you don't need to wait for someone to ask first, and any topic goes, no matter how narrow or broad
    This feature makes it possible for readers to find better explanations of any topic created by other writers. And it allows writers to create an explanation in a place that readers might actually find it.
    Figure 1.
    Screenshot of the "Derivative" topic page
    . View it live at: ourbigbook.com/go/topic/derivative
  2. local editing: you can store all your personal knowledge base content locally in a plaintext markup format that can be edited locally and published either:
    This way you can be sure that even if OurBigBook.com were to go down one day (which we have no plans to do as it is quite cheap to host!), your content will still be perfectly readable as a static site.
    Figure 5. . You can also edit articles on the Web editor without installing anything locally.
    Video 3.
    Edit locally and publish demo
    . Source. This shows editing OurBigBook Markup and publishing it using the Visual Studio Code extension.
  3. https://raw.githubusercontent.com/ourbigbook/ourbigbook-media/master/feature/x/hilbert-space-arrow.png
  4. Infinitely deep tables of contents:
    Figure 6.
    Dynamic article tree with infinitely deep table of contents
    .
    Descendant pages can also show up as toplevel e.g.: ourbigbook.com/cirosantilli/chordate-subclade
All our software is open source and hosted at: github.com/ourbigbook/ourbigbook
Further documentation can be found at: docs.ourbigbook.com
Feel free to reach our to us for any help or suggestions: docs.ourbigbook.com/#contact