Radial Integration and Ball Volume in Continuous Dimension

Andreu Ballús Santacana · 2026

Read Radial Integration and Ball Volume in Continuous Dimension by Andreu Ballús Santacana (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-04351.txt.

Introduction

Andreu Ballus Santacana proves a rigidity result about integration in dimensions that need not be whole numbers. He asks which averaging rules on positive inputs obey a scaling law of a given degree and match the standard Gaussian normalisation.

Only one family of measures qualifies, built from the Mellin transform and the gamma function, and Santacana shows this structure is forced by the assumptions rather than chosen. A change of variables turns the scaling law into plain translation invariance, which pins the measure down. As a payoff, the familiar formula for the volume of a ball in any dimension falls out directly.

Catalog record

CREATORAndreu Ballús Santacana
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE41 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2605.04351

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Radial Integration and Ball Volume in Continuous Dimension Andreu Ballús Santacana Abstract We prove that on the space C c ​ ( ℝ > 0 ) C_{c}(\mathbb{R}_{>0}) of compactly supported continuous functions on the positive half-line, the only positive linear functionals satisfying scaling covariance of degree x / 2 x/2 and Gaussian normalization to π x / 2 \pi^{x/2} are those represented by the Mellin–Gamma measure d ​ μ x ​ ( u ) = π x / 2 Γ ⁡ ( x / 2 ) ​ u x / 2 − 1 ​ d ​ u , x > 0 . d\mu_{x}(u)\;=\;\frac{\pi^{x/2}}{\Gamma(x/2)}\,u^{x/2-1}\,du,\qquad x>0. Among all positive Radon measures on ( 0 , ∞ ) (0,\infty) , the Mellin–Gamma family is the unique family compatible with both axioms; the classification is therefore a rigidity statement, not a derivation of a known formula. The proof reduces, via the logarithmic change of variable u = e t u=e^{t} , to the Haar uniqueness theorem on ( ℝ , + ) (\mathbb{R},+) : scaling covariance forces translation-invariance of an associated measure on ℝ \mathbb{R} , and Gaussian normalization fixes the multiplicative constant through the Mellin transform of e − u e^{-u} . The Euclidean ball-volume formula V ⁡ ( x ) = π x / 2 / Γ ⁡ ( x / 2 + 1 ) V(x)=\pi^{x/2}/\Gamma(x/2+1) follows by integrating μ x \mu_{x} over the unit interval in the squared-radius variable. The induced dimension-shift structure factors through two multiplicative 1 1 -cocycles for the dimension-shift relation x ↦ x + 2 ​ r x\mapsto x+2r on ℝ > 0 \mathbb{R}_{>0} : the radial-integration transport R ⁡ ( x , r ) R(x,r) and the ball-volume transport T ⁡ ( x , r ) T(x,r) . They differ by the multiplicative coboundary of β ⁡ ( x ) = x \beta(x)=x , and T T is independently characterized by a shifted Bohr–Mollerup theorem. The coboundary β ⁡ ( x ) \beta(x) is the categorical-dimension functional of the standard object in Deligne’s interpolation category Rep ⁡ ( O t ) \mathrm{Rep}(O_{t}) . 1 Introduction The volume of the Euclidean unit ball in \mathbb{R}^{n} is V_{n}\;=\;\operatorname{vol}\bigl\{y\in\mathbb{R}^{n}:|y|\leq 1\bigr\}\;=\;\frac{\pi^{n/2}}{\Gamma(n/2+1)}. (1) Replacing n by a real parameter x>0 gives the standard continuation V(x)\;=\;\frac{\pi^{x/2}}{\Gamma(x/2+1)}. (2) Agreement of ( 2 ) with the integer-dimensional values does not, by itself, determine this continuation. For any entire function h of subexponential growth, \frac{\pi^{z/2}}{\Gamma(z/2+1)}\;+\;\sin(\pi z)\,h(z) agrees with V_{n} at every n\in\mathbb{Z}_{\geq 1} but differs from ( 2 ) for general non-integer z . Hadamard’s 1894 interpolation [ 7 ] of 1/\Gamma(z+1) by an entire function exhibits a concrete second analytic candidate. The question that selects ( 2 ) from among such candidates is therefore not interpolation but the imposition of additional structure on the candidate continuations. This paper gives one such selection and proves that it is rigid. We work directly with the radial integration operator obtained from polar coordinates in the squared-radius variable u=r^{2} . For each x>0 we consider positive linear functionals \mathcal{I}_{x} on C_{c}(\mathbb{R}_{>0}) and impose two axioms: scaling covariance of degree x/2 , and Gaussian normalization to \pi^{x/2} . The classification theorem (Theorem 4.1 ) states that the unique family \{\mathcal{I}_{x}\}_{x>0} satisfying both axioms is given by \mathcal{I}_{x}(\phi)\;=\;\frac{\pi^{x/2}}{\Gamma(x/2)}\int_{0}^{\infty}\phi(u)\,u^{x/2-1}\,du. The Mellin–Gamma form is therefore not posited but forced, given the framework of positive functionals on C_{c}(\mathbb{R}_{>0}) and the two axioms; no assumption of holomorphy, of a closed-form ansatz for V(x) , or of properties of \Gamma enters the proof. The Euclidean ball-volume formula ( 2 ) is recovered as \mu_{x}((0,1)) , where \mu_{x} is the Radon measure representing \mathcal{I}_{x} . Three observations motivate the axioms. First, the classical polar-coordinate formula on \mathbb{R}^{n} rewrites Euclidean integration in terms of a radial density r^{n-1} together with a spherical surface measure. Bui and Randles [ 2 ] extend this to general homogeneous settings, replacing n-1 by a homogeneous order and producing an associated Radon surface measure on the unit level set; the continuous-dimension analogue we work with is the simplest instance of this pattern. Second, the Gamma function is the Mellin transform of the Gaussian kernel e^{-u} , in the sense that \int_{0}^{\infty}e^{-u}u^{s-1}\,du=\Gamma(s) [ 6 , §4] ; this identity fixes the multiplicative constant in the classification. Third, \Gamma is uniquely characterized on (0,\infty) by the recurrence \Gamma(x+1)=x\Gamma(x) , the normalization \Gamma(1)=1 , and log-convexity [ 1 , 8 , 9 ] ; we use a shifted form of this characterization to give a second, independent rigidity statement for the dimension-shift transport in Section 7 . The classification produces a measure-valued family \{\mu_{x}\}_{x>0} . The induced dimension-shift map factors through two distinct one-parameter cocycles: • the radial-integration transport R(x,r)=\pi^{r}\,\Gamma(x/2)/\Gamma(x/2+r) , controlling shifts of the radial measure \mu_{x} ; • the ball-volume transport T(x,r)=\pi^{r}\,\Gamma(x/2+1)/\Gamma(x/2+r+1) , controlling shifts of the scalar V(x) . The two transports satisfy T(x,r)=\frac{x}{x+2r}\,R(x,r) . In Section 6 we show that R and T are multiplicative 1 -cocycles for the dimension-shift relation x\mapsto x+2r on \mathbb{R}_{>0} , and that the factor \frac{x+2r}{x} relating them is the multiplicative coboundary of the linear function \beta(x)=x . Categorical interpretation. The function \beta(x)=x admits an interpretation in Deligne’s interpolation categories. The framework for the orthogonal series is recalled in [ 4 , §§9–10] , and we use the Brauer-category presentation of [ 3 , §§2.1–2.2] , in which \mathrm{Rep}(O_{t}) is the Karoubi envelope of the additive envelope of the Brauer category \mathcal{B}(t) . The standard generating object \mathsf{X}_{t} has categorical (quantum) dimension \dim_{\mathrm{cat}}(\mathsf{X}_{t})=t ; hence \beta(x)=\dim_{\mathrm{cat}}(\mathsf{X}_{x}) , and the coboundary distinguishing R from T is the categorical-dimension functional applied to the standard object. This is recorded as Proposition 6.6 . The interpretation is external to the analytic content: the construction of \mathcal{M} in Sections 3 – 4 does not refer to \mathrm{Rep}(O_{t}) , and the proposition is included only to identify the coboundary with an independently defined categorical functional. Outline. Section 2 fixes the polar-coordinate setup and the change of variable u=r^{2} . Section 3 states the two axioms. Section 4 proves the classification theorem. Section 5 obtains the ball-volume formula. Section 6 develops the cocycle structure of R and T and identifies the coboundary with the categorical-dimension functional. Section 7 gives the shifted Bohr–Mollerup characterization of T . Section 8 embeds V(x) in a wider system of Mellin observables. Section 9 extends the mechanism to the generalized homogeneous polar setting. Section 10 compares the three routes by which the closed form for V(x) can be obtained. 2 Preliminaries For integer n\geq 1 and a measurable \phi:[0,\infty)\to\mathbb{R} for which the integral is well-defined, \int_{\mathbb{R}^{n}}\phi(|y|^{2})\,dy\;=\;\omega_{n-1}\int_{0}^{\infty}\phi(r^{2})\,r^{n-1}\,dr, (3) where \omega_{n-1}=2\pi^{n/2}/\Gamma(n/2) is the surface area of the unit sphere S^{n-1}\subset\mathbb{R}^{n} . Substituting u=r^{2} , du=2r\,dr , and absorbing the factor of 2 , ( 3 ) becomes \int_{\mathbb{R}^{n}}\phi(|y|^{2})\,dy\;=\;\frac{\pi^{n/2}}{\Gamma(n/2)}\int_{0}^{\infty}\phi(u)\,u^{n/2-1}\,du. (4) We write this radial-integration coefficient as C(n)\;:=\;\frac{\pi^{n/2}}{\Gamma(n/2)}. Setting \phi=\mathbf{1}_{(0,1)} in ( 4 ) recovers ( 1 ): V_{n}\;=\;C(n)\int_{0}^{1}u^{n/2-1}\,du\;=\;\frac{2\,C(n)}{n}\;=\;\frac{\pi^{n/2}}{\Gamma(n/2+1)}. Hence, at integer dimensions, V(n)\;=\;\frac{2}{n}\,C(n). (5) This elementary relation between radial-integration normalization and ball-volume normalization will reappear in continuous dimension as the algebraic origin of the gap between the two transports of Section 6 . For real x>0 , define C(x)\;:=\;\frac{\pi^{x/2}}{\Gamma(x/2)},\qquad V(x)\;:=\;\frac{\pi^{x/2}}{\Gamma(x/2+1)}. (6) Using \Gamma(x/2+1)=(x/2)\Gamma(x/2) , we have V(x)=(2/x)\,C(x) , the continuous-dimension version of ( 5 ). The reader should not at this

Read the complete text