Jacobi-Anger Expansion

In physics and engineering, the way we describe a wave often depends on our perspective. A simple, straight-propagating plane wave can appear complex when viewed from a different coordinate system, creating a fundamental translation problem that arises in fields from optics to antenna design. How can we systematically express a plane wave in terms of circular waves? The answer lies in a powerful mathematical identity: the Jacobi-Anger expansion. This article serves as a guide to this essential tool. The first chapter, ​​Principles and Mechanisms​​, will unpack the mathematical foundation of the expansion, revealing how it acts as a generating function for the entire family of Bessel functions and allows for the elegant derivation of their properties. Following this, the chapter on ​​Applications and Interdisciplinary Connections​​ will demonstrate the expansion's remarkable utility in explaining real-world phenomena, from the sidebands of an FM radio signal to the quantum control of single atoms. We begin by exploring the core relationship between straight-line motion and circular waves that this expansion so beautifully describes.

Imagine you are standing in a vast, calm lake. You create a perfectly straight wave, a long, travelling ripple moving in one direction. A physicist might describe this with a simple, elegant expression, something like $e^{ikx}$. This is a ​​plane wave​​; its wavefronts are parallel straight lines. But now, another observer is sitting on a boat at some distance from you. From their perspective, your straight wave doesn't look so simple. They would naturally describe what they see in terms of circular ripples spreading out from their position. The question then becomes, how do you translate the simple language of a plane wave into the language of circular waves? How are these two descriptions related?

This is not just a curiosity. It is a fundamental problem that appears everywhere in science, from analyzing the scattering of light off a tiny cylinder to describing the vibrations of a drumhead, or even modeling the behavior of antennas radiating radio waves. The answer to this puzzle is a truly remarkable piece of mathematics known as the ​​Jacobi-Anger expansion​​. It’s more than just a formula; it's a key that unlocks a deep connection between straight-line motion and circular motion, and in doing so, it generates an entire family of indispensable mathematical tools: the Bessel functions.

Let's get a little more specific. Our plane wave, moving along the $x$-axis, can be written as $e^{ikx}$. If we set up a cylindrical coordinate system $(\rho, \phi, z)$, then the $x$-coordinate is simply $x = \rho\cos\phi$. So our plane wave becomes $e^{ik\rho\cos\phi}$. We want to express this as a sum of circular waves. What do circular waves look like?

Each fundamental circular wave pattern can be described by a term like $e^{im\phi}$. For $m=0$, this is just a constant value, a pure circular ripple. For $m=1$, it has one full cycle of phase as you go around the circle—think of a wave with one peak and one trough along the circumference. For $m=2$, it has two, and so on. These $e^{im\phi}$ terms are the "circular harmonics," the basic building blocks of any pattern on a circle.

Of course, the strength of each of these circular waves must also depend on the distance $\rho$ from the origin. This radial dependence is captured by a set of functions we'll call $J_m(k\rho)$, the ​​Bessel functions of the first kind​​. So, our grand idea is to write the plane wave as a weighted sum—a superposition—of these circular building blocks:

$e^{ik\rho\cos\phi} = \sum_{m=-\infty}^{\infty} c_m J_m(k\rho) e^{im\phi}$

Here, the coefficients $c_m$ tell us how much of each circular harmonic is needed to build our plane wave. The magic of the Jacobi-Anger expansion lies in discovering what these coefficients, and the functions $J_m$, truly are.

So what are these mysterious mixing coefficients, $c_m$? One might expect some horrendously complicated expression. The beauty of mathematics, however, is that often the most fundamental relationships are also the most elegant. Through a careful analysis of this expansion, one can show that these coefficients are astonishingly simple. They are just powers of the imaginary unit, $i$:

$c_m = i^m$

This means our expansion becomes a thing of profound simplicity and beauty:

$e^{iz\cos\phi} = \sum_{m=-\infty}^{\infty} i^m J_m(z) e^{im\phi}$

(We've simplified the notation a bit by letting $z = k\rho$). This identity is a cornerstone. It tells you the exact recipe for building a plane wave out of circular waves.

You will often see a "sister" version of this identity. By a clever change of the angle $\phi$, we can turn the $\cos\phi$ into a $\sin\phi$, and the little $i^m$ factors are absorbed, leaving an even cleaner form:

$e^{iz\sin\phi} = \sum_{m=-\infty}^{\infty} J_m(z) e^{im\phi}$

This form is particularly special. It's what mathematicians call a ​​generating function​​. Think of it like a mathematical vending machine. The function on the left, $e^{iz\sin\phi}$, is the machine itself. The terms on the right are the products it dispenses. If you want to know the Bessel function $J_m(z)$, all you have to do is find the $m$-th ​​Fourier coefficient​​ of the function $e^{iz\sin\phi}$. The entire, infinitely complex family of Bessel functions is encoded within one simple exponential function.

This expansion is more than just a definition; it's a "Rosetta Stone" that allows us to translate problems from one mathematical language to another. Specifically, it connects the world of complex exponentials (which are easy to manipulate) to the world of Bessel functions (which are often more opaque).

For instance, what if we needed to calculate a rather nasty-looking integral like this one from a problem in wave theory?

$I(z) = \int_0^{2\pi} e^{i(z\cos\theta - 4\theta)} d\theta$

This looks intimidating. But with our Rosetta Stone, we can see it for what it is. The term $e^{iz\cos\theta}$ is our "signal," and multiplying by $e^{-i4\theta}$ and integrating is just the standard mathematical procedure for finding the 4th Fourier coefficient. It's like listening to a complex chord and picking out a specific note. Using the expansion $e^{iz\cos\theta} = \sum_k i^k J_k(z) e^{ik\theta}$, the integral becomes:

$I(z) = \int_0^{2\pi} \left( \sum_{k=-\infty}^{\infty} i^k J_k(z) e^{ik\theta} \right) e^{-i4\theta} d\theta = \sum_{k=-\infty}^{\infty} i^k J_k(z) \int_0^{2\pi} e^{i(k-4)\theta} d\theta$

The integral $\int_0^{2\pi} e^{i(k-4)\theta} d\theta$ is zero unless the frequencies match perfectly, i.e., $k-4=0$. This property, called ​​orthogonality​​, makes all terms in the infinite sum vanish except for the one where $k=4$. The result is simply $i^4 J_4(z) \cdot (2\pi) = 2\pi J_4(z)$. A difficult integral is solved in two lines!

This "inversion" trick can be turned around. Instead of using the expansion to solve an integral, we can use the integral to define the Bessel function. This leads to a beautiful and profound integral representation:

$J_n(x) = \frac{1}{\pi} \int_0^\pi \cos(n\theta - x\sin\theta) d\theta$

This tells us that the Bessel function $J_n(x)$ is, in essence, the average value of a cosine function whose phase is being "wobbled" by a sinusoid. It's a snapshot of a cosmic dance between two coupled oscillations.

The true power of a great theoretical tool is not just in solving problems we already have, but in allowing us to discover new truths we didn't even know we were looking for. The Jacobi-Anger expansion is a veritable machine for generating new and surprising identities.

One way to use the machine is through ​​specialization​​. We can plug in a clever choice for the angle $\theta$ to make a complicated sum collapse into something simple. For example, consider the strange alternating sum of odd-order Bessel functions, $S = \sum_{k=0}^\infty (-1)^k J_{2k+1}(x)$. How could we possibly calculate that? We start with the imaginary part of the generating function:

$\sin(x \sin \theta) = 2 \sum_{k=0}^{\infty} J_{2k+1}(x) \sin((2k+1)\theta)$

We want the sum to have a factor of $(-1)^k$. Is there an angle $\theta$ where $\sin((2k+1)\theta)$ becomes $(-1)^k$? A moment's thought leads to $\theta = \frac{\pi}{2}$. For this angle, $\sin((2k+1)\frac{\pi}{2})$ becomes $1, -1, 1, -1, \dots$, which is exactly $(-1)^k$. Plugging $\theta = \pi/2$ into the identity, the left side becomes $\sin(x \sin(\pi/2)) = \sin(x)$. The right side becomes $2 \sum (-1)^k J_{2k+1}(x)$. And just like that, we've found our sum: $\sum_{k=0}^\infty (-1)^k J_{2k+1}(x) = \frac{1}{2}\sin(x)$. This powerful strategy can be used to solve a wide variety of similar series problems.

Another trick is to see what happens when we operate on the identity. For instance, what if we differentiate it? Let's take the identity $e^{iz\sin\theta} = \sum_n J_n(z) e^{in\theta}$ and differentiate both sides with respect to $\theta$:

$i z \cos\theta \cdot e^{iz\sin\theta} = \sum_{n=-\infty}^{\infty} i n J_n(z) e^{in\theta}$

This new identity must also be true for any $\theta$. What if we pick the simplest possible angle, $\theta=0$? The $\cos\theta$ becomes 1, and all the $e^{in\theta}$ terms on the right side also become 1. The equation implodes into a stunningly simple statement:

$z = \sum_{n=-\infty}^{\infty} n J_n(z)$

This is amazing! If you take all the Bessel functions, weight each by its order $n$, and add them all up, the result is simply the argument $z$. This reveals a deep, hidden structure within the family of Bessel functions.

In physics, some of the most profound laws are conservation laws—statements that a certain quantity (like energy or momentum) remains constant. There is a mathematical analogue to this in the world of Fourier series, called ​​Parseval's theorem​​. It relates the total "energy" of a signal (the integral of its squared magnitude) to the sum of the energies of its harmonic components.

Since our Jacobi-Anger expansion is a Fourier series, we can apply this powerful idea. Our "signal" is $f(\theta) = e^{ix\sin\theta}$ (for real $x$). The total energy is easy to calculate: $|f(\theta)|^2 = |e^{ix\sin\theta}|^2 = 1$. The average energy over one cycle is therefore simply 1.

Parseval's theorem says this must equal the sum of the squared magnitudes of the Fourier coefficients. The coefficients are $J_n(x)$. Therefore, we arrive at a fundamental conservation law for Bessel functions:

$\sum_{n=-\infty}^{\infty} [J_n(x)]^2 = 1$

No matter what the value of $x$ is, the sum of the squares of all these infinitely many, wildly oscillating functions is always, exactly, 1. This simple fact is a powerful constraint used in many areas of physics and engineering. By applying the same theorem to the derivative of the signal, we can derive other sum rules, like $\sum n^2 [J_n(x)]^2 = \frac{x^2}{2}$.

We can push this idea even further. Parseval's theorem also relates the "cross-correlation" of two different signals to their Fourier coefficients. Let's take two signals, $f(\theta) = e^{ia\cos\theta}$ and $g(\theta) = e^{ib\cos\theta}$. Their Fourier coefficients are $c_n = i^n J_n(a)$ and $d_n = i^n J_n(b)$, respectively. The theorem tells us that $\frac{1}{2\pi}\int f(\theta)\overline{g(\theta)}d\theta = \sum c_n \overline{d_n}$. Working this out leads to one of the most elegant of all Bessel function identities, ​​Neumann's addition formula​​:

$\sum_{n=-\infty}^{\infty} J_n(a) J_n(b) = J_0(a-b)$

The correlation between the entire family of Bessel functions for argument $a$ and the family for argument $b$ collapses to the single, simplest Bessel function, $J_0$, evaluated at the difference $a-b$. This is a statement of profound structural coherence.

The story doesn't end here. The principles we've uncovered are part of a much larger, interconnected web of mathematical ideas.

For example, the Bessel functions we've been discussing, $J_n(x)$, typically describe wave-like phenomena. What about processes like heat diffusion or other phenomena that decay exponentially? It turns out there is a parallel family of functions, the ​​modified Bessel functions​​ $I_n(x)$, that govern these situations. They have their own Jacobi-Anger expansion, $e^{A\cos\theta} = \sum_{n=-\infty}^{\infty} I_n(A) e^{in\theta}$. All the machinery we've developed can be applied here as well. Using Parseval's theorem, we can immediately find the corresponding "conservation law" for these functions: $\sum_{n=-\infty}^{\infty} [I_n(A)]^2 = I_0(2A)$. The structure is analogous, a beautiful example of how a single powerful idea echoes through different branches of mathematics.

And what about our restriction to integer orders $n$? Is that fundamental? Not at all. The Jacobi-Anger expansion can be generalized to handle Bessel functions of any order $\nu$, integer or not. This generalized identity, $\sum_{n=-\infty}^\infty J_{n+\nu}(x) e^{in\theta} = e^{ix\sin\theta} e^{-i\nu\theta}$, shows the underlying structure is even more robust and flexible than we first imagined.

From a simple question about plane and circular waves, we have been led on a journey that generated a whole family of functions, revealed their hidden symmetries, and gave us a powerful toolkit for solving problems. The Jacobi-Anger expansion is a testament to the inherent beauty and unity of physics and mathematics, showing how a single elegant idea can illuminate a vast and intricate landscape.

Now that we have acquainted ourselves with the machinery of the Jacobi-Anger expansion, we can embark on a journey to see it in action. You might be tempted to view this expansion as just another mathematical curiosity, a clever trick for manipulating symbols. But nothing could be further from the truth. This relationship is a profound statement about the nature of waves and oscillations. It is a universal key that unlocks the secrets of any system where a simple sinusoidal process is modulated by another.

The expansion acts like a mathematical prism. It takes a seemingly complicated single expression—a sine wave inside another wave—and decomposes it into an infinite, but beautifully ordered, series of simple components. Each of these components, weighted by a mysterious and wonderful function called the Bessel function, corresponds to a distinct physical phenomenon. By turning this key, we will find the same underlying harmony in the broadcast of a radio station, the shimmering colors from a diffraction grating, the precise control of single atoms, and the very structure of crystalline materials. Let's look around and see where nature has used this elegant piece of mathematics.

Perhaps the most direct and classic application of the Jacobi-Anger expansion is in the world of telecommunications, specifically in understanding Frequency Modulation, or FM, radio. You have a "carrier" wave, a pure, high-frequency tone of the form $\cos(\omega_c t)$. To encode a signal, say a simple musical note $\sin(\omega_m t)$, we don't just add it; we let it modulate the phase of the carrier. The resulting signal looks like $\cos(\omega_c t + \beta\sin(\omega_m t))$.

At first glance, this looks like a hopelessly complex jumble. How can we possibly determine the frequencies present in this signal? This is where the magic happens. By representing the cosine with complex exponentials, we find the term $e^{i\beta\sin(\omega_m t)}$. And this is precisely the form that the Jacobi-Anger expansion deciphers for us! It tells us that this single, phase-modulated term is actually equivalent to a sum of an infinite number of pure frequencies:

What does this mean physically? It means our original carrier wave at frequency $\omega_c$ has been transformed into an entire "comb" of frequencies. We have the original carrier at $\omega_c$ (the $k=0$ term), and an infinite series of "sidebands" at frequencies $\omega_c \pm \omega_m$, $\omega_c \pm 2\omega_m$, $\omega_c \pm 3\omega_m$, and so on. The amplitude of each of these sidebands is not random; it is given precisely by the Bessel function $J_k(\beta)$, where $\beta$ is the modulation index—a measure of how strongly we are modulating the phase. The power in the carrier is proportional to $J_0^2(\beta)$, the power in the first pair of sidebands is proportional to $J_1^2(\beta)$, and so on. This isn't just theory; it's what every FM radio receiver must be designed to handle. A seemingly simple phase wiggle blossoms into a rich spectrum of frequencies, all orchestrated by the family of Bessel functions.

The same principle extends beyond analog radio into the digital world. In digital signal processing, one can encounter a signal whose frequency-domain representation (its Fourier Transform) has a purely sinusoidal phase, like $X(e^{i\omega}) = e^{i\alpha \sin(\omega)}$. What does the signal itself, back in the time domain, look like? The Jacobi-Anger expansion provides the answer directly. The inverse Fourier transform, which involves integrating over this expression, elegantly picks out one term from the series for each point in time. The resulting time-domain signal $x[n]$ turns out to be nothing other than the Bessel function itself, $J_{-n}(\alpha)$. This beautiful duality shows the deep and symmetric connection the expansion forges between the time and frequency domains.

The world of optics provides some of the most stunning visual demonstrations of the Jacobi-Anger expansion. Imagine taking a flat, transparent piece of plastic and etching its thickness with a gentle, sinusoidal wave pattern. This is known as a sinusoidal phase grating. When a beam of laser light passes through it, the light's phase is modulated spatially, creating a wavefront described by an expression like $e^{im \sin(Kx)}$, where $m$ is the depth of the modulation and $K$ is its spatial frequency.

What do you see on a screen placed far behind this grating? You don’t see a distorted blur. Instead, you see a sharp, central spot of light, flanked by a series of dimmer, equally spaced spots. These are the diffracted orders. The Jacobi-Anger expansion describes this phenomenon perfectly. The transmitted wave is decomposed into a sum of simple plane waves, each traveling in a slightly different direction. Each term in the expansion corresponds to one of the spots you see. The intensity of the central, undiffracted beam (the 0th order) is proportional to $J_0^2(m)$, the intensity of the first pair of spots (the ±1st orders) is proportional to $J_1^2(m)$, and so on for all the orders. By simply adjusting the depth of the etched grooves, we can control how much light is channeled into each diffracted beam, effectively "painting with phase" and seeing the Bessel functions come to life.

We can play a similar game in time instead of space. In a device like a Twyman-Green interferometer, two beams of light are recombined, and their interference depends on the difference in their path lengths. If we mount one of the mirrors on a transducer and make it vibrate back and forth sinusoidally, we are modulating the phase of one beam in time. A photodetector watching the output doesn't just see a flickering light; it registers a signal containing distinct frequencies that are integer multiples (harmonics) of the mirror's vibration frequency. The amplitude of the second harmonic is proportional to $J_2$, the fourth to $J_4$, and so on. This effect is not a mere curiosity; it is the basis for heterodyne interferometry, a technique that allows for astoundingly precise measurements of vibrations and surface features by analyzing the strength of these Bessel-governed harmonics.

An even more subtle effect appears when we modulate polarization. If you take a beam of horizontally polarized light and pass it through a half-wave plate that is being rapidly rotated back and forth, the polarization of the outgoing light is constantly changing. If you measure this light with a slow detector, what do you see? You might guess you'd just get unpolarized light. The Jacobi-Anger expansion allows us to calculate the exact time-averaged state of polarization. It turns out the light becomes partially depolarized, and the remaining degree of polarization is given by $|J_0(4\alpha)|$, where $\alpha$ is the amplitude of the angular oscillation. A perfectly coherent, deterministic modulation can, after time-averaging, produce an effect that mimics random depolarization, a phenomenon quantified with perfect precision by a Bessel function.

The reach of the Jacobi-Anger expansion extends down to the quantum level and into the structure of matter itself. Consider the challenge of controlling a single atom, a fundamental task in building a quantum computer. Such an atom can be modeled as a two-level system (a "qubit"). We "talk" to it using electromagnetic fields, like lasers or microwaves. If we modulate the phase of our driving field sinusoidally, we are no longer driving the atom with a single frequency, but with the entire comb of sidebands we saw in the FM radio example.

This gives us a remarkable form of control. Suppose the main frequency of our laser is far from what the atom wants to absorb. By adding phase modulation, we can create a sideband that is perfectly in tune with the atom's transition frequency. The atom will now respond to this sideband! The strength of this new, engineered interaction—the "effective Rabi frequency"—is directly proportional to the corresponding Bessel function, $J_n(\beta)$. This technique effectively creates tunable, artificial transitions, giving physicists a versatile tool for the delicate choreography of quantum states.

Finally, let us look at the solid-state world. A perfect crystal is a perfectly periodic array of atoms. When we scatter X-rays from it, we get sharp diffraction spots known as Bragg peaks. But what if the crystal has a subtle, wave-like distortion running through it—a sinusoidal ripple in the atomic positions? This is called an incommensurate modulation. The Jacobi-Anger expansion once again provides the key to understanding the diffraction pattern from such a structure. The scattering amplitude calculation involves a sum over atoms with positions like $x_n' = na + u \sin(qna)$. The term $e^{-iK x_n'}$ contains our familiar friend, which can be expanded into Bessel functions. The result is that in addition to the main Bragg peaks, the diffraction pattern will be adorned with a series of smaller "satellite" peaks. The intensity of the $m$-th satellite peak is proportional to $J_m^2$, where the argument of the Bessel function depends on the scattering vector and the amplitude of the atomic displacement. Crystallographers use this very principle to detect and measure these subtle structural waves in real materials, mapping the intricate dances of atoms in the solid state.

From the macro-world of radio waves and optics to the micro-world of atoms and crystals, the Jacobi-Anger expansion reveals a common thread. It is a testament to the fact that the universe often uses the same mathematical ideas in the most disparate of settings. A sinusoidal wiggle, whether in the phase of a light wave, the position of an atom, or the timing of a signal, invariably gives birth to a rich spectrum of harmonics, whose strengths are forever governed by the elegant and ubiquitous Bessel functions.