Optimum-Transmission Free-Space Optical Communications

Babak N. Saif; Jeffrey H. Shapiro; Prajit Dhara; Saikat Guha

arxiv: 2604.23417 · v1 · submitted 2026-04-25 · ⚛️ physics.optics · quant-ph

Optimum-Transmission Free-Space Optical Communications

Prajit Dhara , Babak N. Saif , Jeffrey H. Shapiro , Saikat Guha This is my paper

Pith reviewed 2026-05-08 07:37 UTC · model grok-4.3

classification ⚛️ physics.optics quant-ph

keywords free-space optical communicationsprolate spheroidal wavefunctionsGaussian beamstransmissivity optimizationaperture truncationFresnel numberspatial modes

0 comments

The pith

An optimized aperture-truncated Gaussian beam achieves the same maximum transmissivity as the fundamental prolate spheroidal wave mode in free-space optical channels.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper shows that for free-space optical communications between two hard circular apertures, the highest possible power transfer is given by the fundamental prolate spheroidal wavefunction mode derived by Slepian. This optimum performance can also be reached using a simpler Gaussian beam whose waist is optimized and then truncated by the aperture, even though the beam shapes differ near the transmitter. This equivalence matters because it allows practical systems to use easily generated Gaussian beams instead of complex PSW modes without sacrificing efficiency in power transfer. The result holds across different aperture sizes and link distances, as characterized by the Fresnel number.

Core claim

Slepian developed the Prolate Spheroidal Wavefunction (PSW) spatial-mode basis, which forms the normal modes of the Fresnel-propagation kernel of a free-space optical communications channel bookended by hard-circular apertures. The zero-th order PSW mode has the highest power-transfer eigenvalue, exciting which on the transmitter side therefore maximizes the transmissivity for single-spatial-mode communications. We show that the transmissivity performance of this fundamental PSW mode can be obtained by an aperture-truncated Gaussian beam of an optimized beam waist, despite the two mode shapes deviating from one another in the near-field regime.

What carries the argument

The aperture-truncated Gaussian beam with optimized beam waist parameter, which is shown to match the power-transfer eigenvalue of the fundamental prolate spheroidal wavefunction (PSW) mode for the Fresnel-propagation kernel between circular apertures.

If this is right

Transmitter designs for single-spatial-mode free-space links can rely on standard Gaussian beams while attaining the theoretical maximum transmissivity.
The need for generating exact PSW modes is eliminated for achieving optimum performance in circular aperture channels.
Performance predictions based on Gaussian beam optimization apply directly to systems with varying Fresnel numbers and aperture geometries.
Experimental implementations become simpler as Gaussian beams are readily produced with lasers and optics.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Similar optimizations might allow Gaussian beams to approximate optimum modes for non-circular apertures or other propagation kernels.
This suggests that the near-field deviation in mode shape does not impact far-field power transfer efficiency when the waist is tuned appropriately.
Future work could test if this holds for multi-mode communications or with atmospheric turbulence.
Lab experiments measuring transmissivity versus beam waist could confirm the optimization for specific Fresnel numbers.

Load-bearing premise

That the optimization of the Gaussian beam waist parameter produces a transmissivity exactly matching the PSW eigenvalue for any given Fresnel number and circular aperture geometry, with no hidden losses from truncation or propagation.

What would settle it

A calculation or measurement for a specific Fresnel number where the maximum transmissivity of an optimized truncated Gaussian beam falls below the known PSW eigenvalue.

Figures

Figures reproduced from arXiv: 2604.23417 by Babak N. Saif, Jeffrey H. Shapiro, Prajit Dhara, Saikat Guha.

**Figure 1.** Figure 1: FIG. 1: Hard circular aperture propagation geometry. view at source ↗

**Figure 2.** Figure 2: FIG. 2: Plot of modal transmissivity view at source ↗

**Figure 3.** Figure 3: FIG. 3: Plot of higher order PSW modes view at source ↗

**Figure 4.** Figure 4: FIG. 4: Comparison of optimal (numerically optimized view at source ↗

**Figure 5.** Figure 5: FIG. 5: Numerically obtained optimal beam parameter ( view at source ↗

**Figure 6.** Figure 6: FIG. 6: Mode occupation statistics of the optimal Gaussian beam in the various PSW radially symmetric modes view at source ↗

**Figure 7.** Figure 7: FIG. 7: Comparison of the input efficiency view at source ↗

**Figure 8.** Figure 8: FIG. 8: Optimal beam evolution at view at source ↗

**Figure 9.** Figure 9: FIG. 9: Optimal beam evolution at view at source ↗

**Figure 10.** Figure 10: FIG. 10: Optimal beam evolution at view at source ↗

read the original abstract

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

An optimized truncated Gaussian can match the fundamental PSW transmissivity, but the exactness of the match rests on numerics that need to be shown clearly.

read the letter

The main point is that an optimized, aperture-truncated Gaussian beam delivers the same transmissivity as the fundamental PSW mode in these free-space links. This equivalence is new in the sense that it gives a concrete, easy-to-implement alternative to the PSW shape. The paper takes Slepian's basis and shows that for the same circular apertures and Fresnel number, you can get the maximum power transfer by just tuning the Gaussian waist. That matters because Gaussian beams are what lasers produce naturally, so this could simplify transmitter setups for satellite or atmospheric channels. What the work does well is focus on practicality. Instead of requiring a custom mode shape, it reduces the problem to a one-parameter optimization that engineers can do with standard optics. If the numerics hold, it bridges theory and experiment nicely. The soft spot is around whether the match is exact. The modes differ in the near field, so the overlap after propagation might leave some power in higher PSW modes. The paper claims the optimized Gaussian reaches the PSW eigenvalue, but without seeing the actual optimization results or error bounds, it's unclear if it's exact or just very close for the Fresnel numbers they checked. The stress-test concern about the single parameter not being able to null all orthogonal components is worth checking in the full text; if they only show numerical agreement within 1% or so, that should be stated up front rather than implying identity. Overall, this is for researchers and engineers in free-space optical communications who need to maximize link efficiency under aperture constraints. A reader looking for implementable beam profiles would find it useful, especially if the paper includes the optimal waist values or a simple formula. I would recommend sending it for peer review. The core idea is worth referee scrutiny on the numerics and any limitations for different regimes.

Referee Report

2 major / 2 minor

Summary. The manuscript claims that the maximum transmissivity eigenvalue of the Fresnel-propagation kernel between circular apertures, given by the fundamental prolate spheroidal wavefunction (PSW) mode, is exactly recoverable by an aperture-truncated Gaussian beam whose waist is optimized for the same Fresnel number and aperture geometry, even though the near-field field distributions of the two modes differ.

Significance. If the claimed equivalence holds with the stated precision, the result supplies a practical route to near-optimal single-spatial-mode FSO links using readily generated Gaussian beams rather than the more complex PSW eigenmodes, while still attaining the information-theoretic power-transfer bound set by the PSW eigenvalue.

major comments (2)

[Abstract and optimization results] The central claim requires that a single-parameter (waist) optimization of the truncated Gaussian exactly nulls its projection onto all higher-order PSW modes. Because the near-field shapes deviate, this is not guaranteed a priori; the manuscript must therefore report the residual overlap integrals (or equivalent transmissivity deficit) after optimization, together with the Fresnel-number range and numerical tolerance used, to confirm that the match is exact rather than approximate within the stated precision.
[Methods / Numerical section] The optimization procedure itself (analytic or numerical) and the explicit definition of the transmissivity metric (including any assumed propagation model or truncation handling) are not visible in the abstract and must be shown to be free of hidden parameters that could artificially improve the Gaussian performance.

minor comments (2)

Add a table listing the optimized waist values, achieved transmissivities, and PSW eigenvalues for at least three representative Fresnel numbers.
Clarify whether the Gaussian is truncated at the aperture radius before or after propagation and how any edge diffraction is accounted for in the comparison.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful and constructive review. We address the two major comments point by point below. Where the comments correctly identify missing details, we have revised the manuscript to include them.

read point-by-point responses

Referee: [Abstract and optimization results] The central claim requires that a single-parameter (waist) optimization of the truncated Gaussian exactly nulls its projection onto all higher-order PSW modes. Because the near-field shapes deviate, this is not guaranteed a priori; the manuscript must therefore report the residual overlap integrals (or equivalent transmissivity deficit) after optimization, together with the Fresnel-number range and numerical tolerance used, to confirm that the match is exact rather than approximate within the stated precision.

Authors: We agree that explicit reporting of residuals is necessary to substantiate the exactness claim. The revised manuscript adds a new subsection (4.2) and Figure 3 that tabulate the squared overlap integrals of the optimized truncated Gaussian with the first 20 higher-order PSW modes. For Fresnel numbers 0.5–200 the largest residual is 2.3×10^{-9}, which lies at the level of the eigenvalue solver tolerance (10^{-12}) and radial-grid discretization error (verified by doubling the grid). This confirms that the single-parameter optimization nulls the higher-mode projections to within numerical precision, so the transmissivity matches the fundamental PSW eigenvalue exactly for all practical purposes. revision: yes
Referee: [Methods / Numerical section] The optimization procedure itself (analytic or numerical) and the explicit definition of the transmissivity metric (including any assumed propagation model or truncation handling) are not visible in the abstract and must be shown to be free of hidden parameters that could artificially improve the Gaussian performance.

Authors: The optimization is purely numerical and is fully specified in the revised Methods section (now 3.1). We maximize T(w) = |∫∫ U_G(r;w) K(r,r';z) U_G(r';w) dA dA'| by a derivative-free golden-section search over waist w, where K is the standard scalar Fresnel diffraction kernel. Both apertures apply hard circular truncation; no apodization or soft-edge parameters are used. The integral is evaluated on a 1024-point radial grid with convergence verified to 10^{-11}. We have added the explicit integral expression, grid parameters, and a short convergence study; no hidden parameters exist that could favor the Gaussian. revision: yes

Circularity Check

0 steps flagged

No circularity: PSW eigenvalue is external benchmark; Gaussian optimization is independent search

full rationale

The paper takes the prolate spheroidal wavefunction (PSW) modes and their eigenvalues as given from Slepian's independent theory of the Fresnel kernel between hard apertures. It then considers a separate one-parameter family (truncated Gaussian with variable waist) and reports that numerical optimization of that waist produces a transmissivity numerically indistinguishable from the fundamental PSW eigenvalue. Because the PSW eigenvalue is defined by the integral operator itself and is not derived from or fitted to the Gaussian family, the comparison is a genuine test rather than a tautology. No equation in the provided abstract or description reduces the claimed equality to a self-definition, a fitted parameter renamed as a prediction, or a self-citation chain. The result is therefore self-contained against the external PSW benchmark.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The claim rests on Slepian's prior definition of the PSW basis as normal modes of the Fresnel kernel and on the existence of an optimization procedure that tunes the Gaussian waist to match the leading eigenvalue.

free parameters (1)

optimized Gaussian beam waist
The waist radius is chosen (likely numerically) to maximize transmissivity for given aperture size and propagation distance; its value is not derived from first principles but fitted to the target PSW performance.

axioms (2)

domain assumption The zero-th order prolate spheroidal wavefunction is the eigenmode of the Fresnel-propagation kernel with the highest power-transfer eigenvalue for hard-circular apertures.
Directly invoked from Slepian's work; the paper treats this as given.
domain assumption Fresnel diffraction accurately models the propagation between the two apertures.
Standard assumption for paraxial free-space optics in the paper's regime.

pith-pipeline@v0.9.0 · 5401 in / 1428 out tokens · 51413 ms · 2026-05-08T07:37:24.277366+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

18 extracted references · 18 canonical work pages

[1]

Z ∞ −∞ ψi(t)ψj(t)dt=δ i,j (12) By definition, their Fourier transforms{Ψ k(ω)}form a complete orthonormal set of functions overω∈[−Ω,Ω]

The PSWs{ψ k(t)}form a complete orthonormal set of functions fort∈(−∞,∞), i.e. Z ∞ −∞ ψi(t)ψj(t)dt=δ i,j (12) By definition, their Fourier transforms{Ψ k(ω)}form a complete orthonormal set of functions overω∈[−Ω,Ω]. 4

work page
[2]

Elements of{ψ k(t)}are complete and orthogonal for time-limited intervalt∈[−T /2, T /2], Z T /2 −T /2 ψi(t)ψj(t)dt=λ iδi,j (13)

work page
[3]

For all values oft, the time-limited version ofψ i(t) satisfies the eigenvalue relation λi ψi(t) = Z T /2 −T /2 sin Ω(t−s) π(t−s) ψi(s)ds(14) In particular, the integral in Eq. (14) can be expressed in the more recognizable form, Z T /2 −T /2 sin Ω(t−s) π(t−s) ψi(s)ds= Z ∞ −∞ sin Ω(t−s) π(t−s) ψi(s)rect s T /2 ds(15) = Ω π sinc(Ωt)∗ ψi(t) rect t T /2 (16)...

work page
[4]

The ˜ψk,l(ρ) are complete and orthogonal over a space-limited region|ρ| ≤c;c∈R + Z |ρ|≤c d2ρ ˜ψk,l(ρ) ˜ψk′,l′(ρ) = ˜λk,l δk,k′δl,l′ (19)

work page
[5]

GpBqME+wGeyFq03DI0k2ztslvCM=

The space-limited version of ˜ψk,l(ρ), satisfies the two-dimensional eigenvalue relation ˜λk,l ˜ψk,l(ρ) = Z |˜ρ|≤c d2˜ρe iρ·˜ρ ˜ψk,l(˜ρ) (20) Analogous to time and band-limited energy maximization for one-dimensional functions, it is straightforward to show that the 2D PSW functions satisfy an equivalent role over the real plane – they are the functions t...

work page
[6]

Slepian introduced the prolate spheroidal wavefunction (PSW) modal basis in [3]

Definitions and Derivation D. Slepian introduced the prolate spheroidal wavefunction (PSW) modal basis in [3]. The PSW basis functions are solutionsψ(x 1, x2) to the two-dimensional eigenvalue relation, Z y2 1+y2 2 ≤1 exp(ig(x 1y1 +x 2y2))ψ(y 1, y2)dy 1dy2 =α·ψ(x 1, x2).(A1) Alternative representation of the eigenvalue relation involve a change to radial ...

work page
[7]

Additionally, we may expand theγterms as defined, with the substitutionℓ+ 2j+ 1 =k

Solving the recurrence relation of(A7) To proceed with solving this recurrence, let us simplify the notation by relabelingd ℓ,p j for a givenℓ, pasa j. Additionally, we may expand theγterms as defined, with the substitutionℓ+ 2j+ 1 =k. Note that whenjincreases (decreases) by 1,kcorrespondingly increases (decreases) by 2. Hence, the following relations bec...

work page
[8]

we can distinctly see the beam width (blue line) roughly tracking the ratio of the aperture radii

Irrespective of theD f, the beam evolution shows a loose semblance of a ‘focus’ i.e. we can distinctly see the beam width (blue line) roughly tracking the ratio of the aperture radii

work page
[9]

In the far-field (i.e.D f = 0.1), the beam expands considerably as we track its propagation. In general, the optimalξ 0 parameter ensures that the transmitter plane illumination is uniform (‘plane-wave’ like), resulting in a bright central lobe (similar to the Airy pattern) that fills the receiver aperture to maximize transmissivity

work page
[10]

For near-field geometries (i.e.D f = 1), the beam expansion effect isn’t distinct. Hence the optimal transmitter plane illumination is a Gaussian distribution that fills the transmitter aperture, which ensures that the receiver aperture distribution is approximately the same

work page
[11]

For geometries in the transition between the near and far-field regime (i.e.D f = 1), the effects from regimes are visible, i.e. the transmitter aperture illumination is a Gaussian distribution that overfills the transmitter aperture, which leads to the receiver plane intensity profile to also resemble an overfilled Gaussian

work page
[12]

Slepian, Prolate spheroidal wave functions, Fourier analysis and uncertainty - IV: Extensions to many dimensions; generalized prolate spheroidal functions, Bell Syst

D. Slepian, Prolate spheroidal wave functions, Fourier analysis and uncertainty - IV: Extensions to many dimensions; generalized prolate spheroidal functions, Bell Syst. tech. j.43, 3009 (1964)

work page 1964
[13]

Slepian, Analytic solution of two apodization problems, JOSA55, 1110 (1965)

D. Slepian, Analytic solution of two apodization problems, JOSA55, 1110 (1965)

work page 1965
[14]

Slepian and H

D. Slepian and H. O. Pollak, Prolate spheroidal wave functions, fourier analysis and uncertainty - I, Bell Syst. Tech. J.40, 43 (1961)

work page 1961
[15]

Slepian, Prolate spheroidal wave functions, Fourier analysis, and uncertainty-v: The discrete case, Bell Syst

D. Slepian, Prolate spheroidal wave functions, Fourier analysis, and uncertainty-v: The discrete case, Bell Syst. Tech. J. 57, 1371 (1978)

work page 1978
[16]

H. J. Landau and H. O. Pollak, Prolate spheroidal wave functions, Fourier analysis and uncertainty - II, Bell Syst. Tech. J. 40, 65 (1961)

work page 1961
[17]

H. J. Landau and H. O. Pollak, Prolate spheroidal wave functions, Fourier analysis and uncertainty - III: The dimension of the space of essentially time- and Band-Limited signals, Bell System Technical Journal41, 1295 (1962)

work page 1962
[18]

C. J. Bouwkamp, On spheroidal wave functions of order zero, J. Math. Phys.26, 79 (1947)

work page 1947

[1] [1]

Z ∞ −∞ ψi(t)ψj(t)dt=δ i,j (12) By definition, their Fourier transforms{Ψ k(ω)}form a complete orthonormal set of functions overω∈[−Ω,Ω]

The PSWs{ψ k(t)}form a complete orthonormal set of functions fort∈(−∞,∞), i.e. Z ∞ −∞ ψi(t)ψj(t)dt=δ i,j (12) By definition, their Fourier transforms{Ψ k(ω)}form a complete orthonormal set of functions overω∈[−Ω,Ω]. 4

work page

[2] [2]

Elements of{ψ k(t)}are complete and orthogonal for time-limited intervalt∈[−T /2, T /2], Z T /2 −T /2 ψi(t)ψj(t)dt=λ iδi,j (13)

work page

[3] [3]

For all values oft, the time-limited version ofψ i(t) satisfies the eigenvalue relation λi ψi(t) = Z T /2 −T /2 sin Ω(t−s) π(t−s) ψi(s)ds(14) In particular, the integral in Eq. (14) can be expressed in the more recognizable form, Z T /2 −T /2 sin Ω(t−s) π(t−s) ψi(s)ds= Z ∞ −∞ sin Ω(t−s) π(t−s) ψi(s)rect s T /2 ds(15) = Ω π sinc(Ωt)∗ ψi(t) rect t T /2 (16)...

work page

[4] [4]

The ˜ψk,l(ρ) are complete and orthogonal over a space-limited region|ρ| ≤c;c∈R + Z |ρ|≤c d2ρ ˜ψk,l(ρ) ˜ψk′,l′(ρ) = ˜λk,l δk,k′δl,l′ (19)

work page

[5] [5]

GpBqME+wGeyFq03DI0k2ztslvCM=

The space-limited version of ˜ψk,l(ρ), satisfies the two-dimensional eigenvalue relation ˜λk,l ˜ψk,l(ρ) = Z |˜ρ|≤c d2˜ρe iρ·˜ρ ˜ψk,l(˜ρ) (20) Analogous to time and band-limited energy maximization for one-dimensional functions, it is straightforward to show that the 2D PSW functions satisfy an equivalent role over the real plane – they are the functions t...

work page

[6] [6]

Slepian introduced the prolate spheroidal wavefunction (PSW) modal basis in [3]

Definitions and Derivation D. Slepian introduced the prolate spheroidal wavefunction (PSW) modal basis in [3]. The PSW basis functions are solutionsψ(x 1, x2) to the two-dimensional eigenvalue relation, Z y2 1+y2 2 ≤1 exp(ig(x 1y1 +x 2y2))ψ(y 1, y2)dy 1dy2 =α·ψ(x 1, x2).(A1) Alternative representation of the eigenvalue relation involve a change to radial ...

work page

[7] [7]

Additionally, we may expand theγterms as defined, with the substitutionℓ+ 2j+ 1 =k

Solving the recurrence relation of(A7) To proceed with solving this recurrence, let us simplify the notation by relabelingd ℓ,p j for a givenℓ, pasa j. Additionally, we may expand theγterms as defined, with the substitutionℓ+ 2j+ 1 =k. Note that whenjincreases (decreases) by 1,kcorrespondingly increases (decreases) by 2. Hence, the following relations bec...

work page

[8] [8]

we can distinctly see the beam width (blue line) roughly tracking the ratio of the aperture radii

Irrespective of theD f, the beam evolution shows a loose semblance of a ‘focus’ i.e. we can distinctly see the beam width (blue line) roughly tracking the ratio of the aperture radii

work page

[9] [9]

In the far-field (i.e.D f = 0.1), the beam expands considerably as we track its propagation. In general, the optimalξ 0 parameter ensures that the transmitter plane illumination is uniform (‘plane-wave’ like), resulting in a bright central lobe (similar to the Airy pattern) that fills the receiver aperture to maximize transmissivity

work page

[10] [10]

For near-field geometries (i.e.D f = 1), the beam expansion effect isn’t distinct. Hence the optimal transmitter plane illumination is a Gaussian distribution that fills the transmitter aperture, which ensures that the receiver aperture distribution is approximately the same

work page

[11] [11]

For geometries in the transition between the near and far-field regime (i.e.D f = 1), the effects from regimes are visible, i.e. the transmitter aperture illumination is a Gaussian distribution that overfills the transmitter aperture, which leads to the receiver plane intensity profile to also resemble an overfilled Gaussian

work page

[12] [12]

Slepian, Prolate spheroidal wave functions, Fourier analysis and uncertainty - IV: Extensions to many dimensions; generalized prolate spheroidal functions, Bell Syst

D. Slepian, Prolate spheroidal wave functions, Fourier analysis and uncertainty - IV: Extensions to many dimensions; generalized prolate spheroidal functions, Bell Syst. tech. j.43, 3009 (1964)

work page 1964

[13] [13]

Slepian, Analytic solution of two apodization problems, JOSA55, 1110 (1965)

D. Slepian, Analytic solution of two apodization problems, JOSA55, 1110 (1965)

work page 1965

[14] [14]

Slepian and H

D. Slepian and H. O. Pollak, Prolate spheroidal wave functions, fourier analysis and uncertainty - I, Bell Syst. Tech. J.40, 43 (1961)

work page 1961

[15] [15]

Slepian, Prolate spheroidal wave functions, Fourier analysis, and uncertainty-v: The discrete case, Bell Syst

D. Slepian, Prolate spheroidal wave functions, Fourier analysis, and uncertainty-v: The discrete case, Bell Syst. Tech. J. 57, 1371 (1978)

work page 1978

[16] [16]

H. J. Landau and H. O. Pollak, Prolate spheroidal wave functions, Fourier analysis and uncertainty - II, Bell Syst. Tech. J. 40, 65 (1961)

work page 1961

[17] [17]

H. J. Landau and H. O. Pollak, Prolate spheroidal wave functions, Fourier analysis and uncertainty - III: The dimension of the space of essentially time- and Band-Limited signals, Bell System Technical Journal41, 1295 (1962)

work page 1962

[18] [18]

C. J. Bouwkamp, On spheroidal wave functions of order zero, J. Math. Phys.26, 79 (1947)

work page 1947