Jacob’s ladders, E. C. Titchmarsh’s hypothesis (1934) and new ζ \zeta -equivalents of the Fermat-Wiles theorem or connections between Fermat’s rationals and the Gram’s sequence
Jan Moser · 2025
Read Jacob’s ladders, E. C. Titchmarsh’s hypothesis (1934) and new ζ \zeta -equivalents of the Fermat-Wiles theorem or connections between Fermat’s rationals and the Gram’s sequence by Jan Moser (2025) 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-2506-01706.txt.
Introduction
Jan Moser, a solo author, connects number theory results in a 2025 math paper. The paper builds on his past proof of a 1934 idea by E C Titchmarsh and on two formulas he derived from it in 1991. From those formulas, he obtains three new forms of the Fermat-Wiles theorem, stated through the zeta function. The title also points to links between Fermat's rational numbers and the Gram sequence.
Catalog record
| CREATOR | Jan Moser |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 24 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-01706
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Jacob’s ladders, E. C. Titchmarsh’s hypothesis (1934) and new ζ \zeta -equivalents of the Fermat-Wiles theorem or connections between Fermat’s rationals and the Gram’s sequence Jan Moser Abstract . In connection of our proof (1980) of the Titchmarsh’s hypothesis (1934), we have obtained two asymptotic formulae (1991). In this paper we obtain three new ζ \zeta -equivalents of the Fermat-Wiles theorem based on the mentioned asymptotic formulae. 1. Introduction This paper is a continuation of the series of our ten papers about the points of contact between the Riemann’s zeta-function and the Fermat-Wiles theorem. 1.1. Let us remind the definition of the Riemann’s zeta function (1.1) \zeta(s)=\prod_{p}\frac{1}{1-\frac{1}{p^{s}}},\ s=\sigma+it,\ \sigma>1, where p runs over the set of all primes, and analytic continuation of this function on all complex s except s=1 . Of course, this definition is based on the Euler’s fundamental identity (1.2) \prod_{p}\frac{1}{1-\frac{1}{p^{x}}}=\sum_{n=1}^{\infty}\frac{1}{n^{x}},\ x>1. Riemann defined also the following real-valued function (1.3) Z(t)=e^{i\vartheta(t)}\zeta\left(\frac{1}{2}+it\right) where 1 1 1 See [ 14 ] , (35), (44), (62); [ 16 ] , p. 98. (1.4) \begin{split}&\vartheta(t)=-\frac{t}{2}\ln\pi+\im\Gamma\left(\frac{1}{4}+i\frac{t}{2}\right)=\\ &\frac{t}{2}\ln\frac{t}{2\pi}-\frac{t}{2}-\frac{\pi}{8}+\mathcal{O}\left(\frac{1}{t}\right).\end{split} Remark 1 . It follows from ( 1.3 ), that the properties of the basic signal ( 1.3 ) generated by the function \zeta\left(\frac{1}{2}+it\right) are connected with the distribution of the primes. This fact is to be regarded as a pleasant circumstance from the point of view of the Pythagorean philosophy of the Universe. 1.2. Let us remind the following facts: (A) Titchmarsh presented in 1934 the hypothesis 2 2 2 [ 16 ] , p. 105. : there is A>0 such that (1.5) \sum_{\nu=M+1}^{N}Z^{2}(t_{\nu})Z^{2}(t_{\nu+1})=\mathcal{O}(N\ln^{A}N), where, if t_{N}=T , (1.6) N\sim\frac{1}{2\pi}T\ln T, and M is sufficiently big fixed number, and next, the sequence \{t_{\nu}\} (the Gram’s sequence) is defined by the condition 3 3 3 Comp. [ 16 ] , p. 99. (1.7) \vartheta(t_{\nu})=\pi\nu,\ \nu=1,2,\dots Remark 2 . Namely, E. C. Titchmarsh finished his paper with this sentence: ”…but there are additional complications, and the conjecture not been verified.” (B) We have proved this Titchmarsh’s hypothesis in 1980 with A=4 , see [ 3 ] . (C) Next, in 1991, we have derived the following asymptotic formula (1.8) \sum_{T\leq t_{\nu}\leq 2T}Z^{2}(t_{\nu})Z^{2}(t_{\nu+1})=\frac{3}{4\pi^{5}}T\ln^{5}T+\mathcal{O}(T\ln^{4}T) (see [ 4 ] , (2.4), (2.6) and (2.10)). Remark 3 . That means, we have obtained the final result: the order of the Titchmarsh’s sum is (1.9) N\ln^{4}N, comp. ( 1.6 ) and ( 1.8 ). 1.3. In this paper we obtain, for example, the functional (cross-bred) based on our asymptotic formula ( 1.8 ). Namely, it is true that: (1.10) \begin{split}&\lim_{\tau\to\infty}\frac{1}{\tau}\left\{\int_{\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}x\tau}^{[\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}x\tau]^{1}}|\zeta(\sigma+it)|^{2}{\rm d}t\right\}^{5}\times\\ &\left\{\int_{\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}x\tau}^{[\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}x\tau]^{1}}\left|\zeta\left(\frac{1}{2}+it\right)\right|^{2}{\rm d}t\right\}^{-5}\times\\ &\sum_{t_{\nu}\geq\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}x\tau}^{\leq\frac{8\pi^{5}}{3\zeta^{5}(2\sigma)}x\tau}\left|\zeta\left(\frac{1}{2}+it_{\nu}\right)\right|^{2}\left|\zeta\left(\frac{1}{2}+it_{\nu+1}\right)\right|^{2}=x,\ x>0\end{split} for every fixed \sigma\geq\frac{1}{2}+\epsilon and every fixed small \epsilon>0 . 1.4. Next, in the special case of the Fermat’s rationals (1.11) x\to\frac{x^{n}+y^{n}}{z^{n}},\ x,y,z,n\in\mathbb{N},\ n\geq 3, it follows from the equation ( 1.10 ) that \zeta -condition (1.12) \begin{split}&\lim_{\tau\to\infty}\frac{1}{\tau}\left\{\int_{\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau}^{[\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau]^{1}}|\zeta(\sigma+it)|^{2}{\rm d}t\right\}^{5}\times\\ &\left\{\int_{\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau}^{[\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau]^{1}}\left|\zeta\left(\frac{1}{2}+it\right)\right|^{2}{\rm d}t\right\}^{-5}\times\\ &\sum_{t_{\nu}\geq\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau}^{\leq\frac{8\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau}\left|\zeta\left(\frac{1}{2}+it_{\nu}\right)\right|^{2}\left|\zeta\left(\frac{1}{2}+it_{\nu+1}\right)\right|^{2}\not=1\end{split} on the class of all Fermat’s rationals ( 1.11 ) represents new \zeta -equivalent of the Fermat-Wiles theorem for every fixed \sigma\geq\frac{1}{2}+\epsilon and every small fixed \epsilon>0 . 1.5. Further, in this paper we prove the following result (1.13) \frac{\int_{\overset{r-1}{T}}^{\overset{r}{T}}|\zeta\left(\frac{1}{2}+it\right)|^{2}{\rm d}t}{\int_{\overset{r-1}{T}}^{\overset{r}{T}}|S_{1}(t)|^{2l}{\rm d}t}=\frac{1}{\bar{c}(l)}\ln\overset{r-1}{T}+\mathcal{O}(1),\ r=1,\dots,k,\ T\to\infty for every fixed k,l\in\mathbb{N} , where (1.14) S_{1}(t)=\frac{1}{\pi}\int_{0}^{t}\arg\left\{\zeta\left(\frac{1}{2}+iu\right)\right\}{\rm d}u. Next, on the basis of the formula ( 1.13 ), we obtain the following functional (cross-bred) (1.15) \begin{split}&\lim_{\tau\to\infty}\frac{1}{\tau}\left\{\int_{\frac{4\pi^{5}}{3\{\bar{c}(l)\}^{5}}x\tau}^{[\frac{4\pi^{5}}{3\{\bar{c}(l)\}^{5}}x\tau]^{1}}|S_{1}(t)|^{2l}{\rm d}t\right\}^{5}\times\\ &\left\{\int_{\frac{4\pi^{5}}{3\{\bar{c}(l)\}^{5}}x\tau}^{[\frac{4\pi^{5}}{3\{\bar{c}(l)\}^{5}}x\tau]^{1}}\left|\zeta\left(\frac{1}{2}+it\right)\right|^{2}{\rm d}t\right\}^{-5}\times\\ &\sum_{t_{\nu}\geq\frac{4\pi^{5}}{3\{\bar{c}(l)\}^{5}}x\tau}^{\leq\frac{8\pi^{5}}{3\{\bar{c}(l)\}^{5}}x\tau}\left|\zeta\left(\frac{1}{2}+it_{\nu}\right)\right|^{2}\left|\zeta\left(\frac{1}{2}+it_{\nu+1}\right)\right|^{2}=x,\ x>0\end{split} for every fixed l\in\mathbb{N} . Remark 4 . Of course, next \zeta -equivalent of the Fermat-Wiles theorem follows from ( 1.15 ) as an analogue of the eq. ( 1.12 ). 2. Jacob’s ladders: notions and basic geometrical properties 2.1. In this paper we use the following notions of our works [ 5 ] – [ 9 ] : (a) Jacob’s ladder \varphi_{1}(T) , (b) direct iterations of Jacob’s ladders \begin{split}&\varphi_{1}^{0}(t)=t,\ \varphi_{1}^{1}(t)=\varphi_{1}(t),\ \varphi_{1}^{2}(t)=\varphi_{1}(\varphi_{1}(t)),\dots,\\ &\varphi_{1}^{k}(t)=\varphi_{1}(\varphi_{1}^{k-1}(t))\end{split} for every fixed natural number k , (c) reverse iterations of Jacob’s ladders (2.1) \begin{split}&\varphi_{1}^{-1}(T)=\overset{1}{T},\ \varphi_{1}^{-2}(T)=\varphi_{1}^{-1}(\overset{1}{T})=\overset{2}{T},\dots,\\ &\varphi_{1}^{-r}(T)=\varphi_{1}^{-1}(\overset{r-1}{T})=\overset{r}{T},\ r=1,\dots,k,\end{split} where, for example, (2.2) \varphi_{1}(\overset{r}{T})=\overset{r-1}{T} for every fixed k\in\mathbb{N} and every sufficiently big T>0 . We also use the properties of the reverse iterations listed below. (2.3) \overset{r}{T}-\overset{r-1}{T}\sim(1-c)\pi(\overset{r}{T});\ \pi(\overset{r}{T})\sim\frac{\overset{r}{T}}{\ln\overset{r}{T}},\ r=1,\dots,k,\ T\to\infty, (2.4) \overset{0}{T}=T<\overset{1}{T}(T)<\overset{2}{T}(T)<\dots<\overset{k}{T}(T), and (2.5) T\sim\overset{1}{T}\sim\overset{2}{T}\sim\dots\sim\overset{k}{T},\ T\to\infty. Remark 5 . The asymptotic behaviour of the points \{T,\overset{1}{T},\dots,\overset{k}{T}\} is as follows: at T\to\infty these points recede unboundedly each from other and all together are receding to infinity. Hence, the set of these points behaves at T\to\infty as one-dimensional Friedmann-Hubble expanding Universe. 2.2. Let us remind that we have proved 4 4 4 See [ 9 ] , (3.4). the existence of almost linear increments (2.6) \begin{split}&\int_{\overset{r-1}{T}}^{\overset{r}{T}}\left|\zeta\left(\frac{1}{2}+it\right)\right|^{2}{\rm d}t\sim(1-c)\overset{r-1}{T},\\ &r=1,\dots,k,\ T\to\infty,\ \overset{r}{T}=\overset{r}{T}(T)=\varphi_{1}^{-r}(T)\end{split} for the Hardy-Littlewood integral (1918) (2.7) J(T)=\int_{0}^{T}\left|\zeta\left(\frac{1}{2}+it\right)\right|^{2}{\rm d}t. For completeness, we give here some basic geometrical properties related to Jacob’s ladders. These are generated by the sequence (2.8) T\to\left\{\overset{r}{T}(T)\right\}_{r=1}^{k} of reverse iterations of the Jacob’s ladders for every sufficiently big T>0 and every fixed k\in\mathbb{N} . Property 1 . The sequence ( 2.8 ) defines a partition of the segment [T,\overset{k}{T}] as follows (2.9) |[T,\overset{k}{T}]|=\sum_{r=1}^{k}|[\overset{r-1}{T},\overset{r}{T}]| on the asymptotically equidistant parts (2.10) \begin{split}&\overset{r}{T}-\overset{r-1}{T}\sim\overset{r+1}{T}-\overset{r}{T},\\ &r=1,\dots,k-1,\ T\to\infty.\end{split} Property 2 . Simultaneously with the Property 1, the sequence ( 2.8 ) defines the partition of the integral (2.11) \int_{T}^{\overset{k}{T}}\left|\zeta\left(\frac{1}{2}+it\right)\right|^{2}{\rm d}t into the parts (2.12) \int_{T}^{\overset{k}{T}}\left|\zeta\left(\frac{1}{2}+it\right)\right|^{2}{\rm d}t=\sum_{r=1}^{k}\int_{\overset{r-1}{T}}^{\overset{r}{T}}\left|\zeta\left(\frac{1}{2}+it\right)\right|^{2}{\rm d}t, that are asymptotically equal (2.13) \int_{\overset{r-1}{T}}^{\overset{r}{T}}\left|\zeta\left(\frac{1}{2}+it\right)\right|^{2}{\rm d}t\sim\int_{\overset{r}{T}}^{\overset{r+1}{T}}\left|\zeta\left(\frac{1}{2}+it\right)\right|^{2}{\rm d}t,\ T\to\infty. It is clear, that ( 2.10 ) follows from ( 2.3 ) and ( 2.5 ) since (2.14) \overset{r}{T}-\overset{r-1}{T}\sim(1-c)\frac{\overset{r}{T}}{\ln\overset{r}{T}}\sim(1-c)\frac{T}{\ln T},\ r=1,\dots,k, while our eq. ( 2.13 ) follows from ( 2.6 ) and ( 2.5 ). 3. The equivalent of the Fermat-Wiles theorem generated by our first asymptotic formula (1991) 3.1. We use the following \zeta -formulas: (A) The quotient formula 5 5 5 See [ 11 ] , (3.11). (3.1) \begin{split}&\frac{\int_{\overset{r-1}{T}}^{\overset{r}{T}}|\zeta\left(\frac{1}{2}+it\right)|^{2}{\rm d}t}{\int_{\overset{r-1}{T}}^{\overset{r}{T}}|\zeta(\sigma+it)|^{2}{\rm d}t}=\frac{1}{\zeta(2\sigma)}\ln\overset{r-1}{T}+\mathcal{O}(1),\\ &r=1,\dots,k,\ T\to\infty\end{split} for every fixed k\in\mathbb{N} , \sigma\geq\frac{1}{2}+\epsilon , that is based on the Hardy-Littlewood formula (1922), comp. [ 11 ] , (3.1), and our almost linear formula, comp. [ 9 ] , (3.6). (B) The asymptotic formula ( 1.8 ), ( 1.9 ) (3.2) \begin{split}&\sum_{t_{\nu}\geq T}^{t_{\nu}\leq 2T}\left|\zeta\left(\frac{1}{2}+it_{\nu}\right)\right|^{2}\left|\zeta\left(\frac{1}{2}+it_{\nu+1}\right)\right|^{2}=\\ &\frac{3}{4\pi^{5}}T\ln^{5}T\left\{1+\mathcal{O}\left(\frac{1}{\ln T}\right)\right\}\end{split} for the Titchmarsh’s sum. Since (see ( 3.1 ), r=1 ) (3.3) \ln T=\zeta(2\sigma)\frac{\int_{T}^{\overset{1}{T}}|\zeta\left(\frac{1}{2}+it\right)|^{2}{\rm d}t}{\int_{T}^{\overset{1}{T}}|\zeta(\sigma+it)|^{2}{\rm d}t}\left\{1+\mathcal{O}\left(\frac{1}{\ln T}\right)\right\}, then it follows from ( 3.2 ) that (3.4) \begin{split}&\left\{\frac{\int_{T}^{\overset{1}{T}}|\zeta\left(\frac{1}{2}+it\right)|^{2}{\rm d}t}{\int_{T}^{\overset{1}{T}}|\zeta(\sigma+it)|^{2}{\rm d}t}\right\}^{5}\times\\ &\sum_{t_{\nu}\geq T}^{t_{\nu}\leq 2T}\left|\zeta\left(\frac{1}{2}+it_{\nu}\right)\right|^{2}\left|\zeta\left(\frac{1}{2}+it_{\nu+1}\right)\right|^{2}=\\ &\frac{3\zeta^{5}(2\sigma)}{4\pi^{5}}T\left\{1+\mathcal{O}\left(\frac{1}{\ln T}\right)\right\}.\end{split} Now, the substitution (3.5) T=\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}x\tau,\ x>0 into the eq. ( 3.4 ) gives the following functional (as the cross-breed of our asymptotic formula ( 1.3 )). Lemma 1 . (3.6) \begin{split}&\lim_{\tau\to\infty}\frac{1}{\tau}\left\{\int_{\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}x\tau}^{[\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}x\tau]^{1}}|\zeta(\sigma+it)|^{2}{\rm d}t\right\}^{5}\times\\ &\left\{\int_{\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}x\tau}^{[\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}x\tau]^{1}}\left|\zeta\left(\frac{1}{2}+it\right)\right|^{2}{\rm d}t\right\}^{-5}\times\\ &\sum_{t_{\nu}\geq\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}x\tau}^{\leq\frac{8\pi^{5}}{3\zeta^{5}(2\sigma)}x\tau}\left|\zeta\left(\frac{1}{2}+it_{\nu}\right)\right|^{2}\left|\zeta\left(\frac{1}{2}+it_{\nu+1}\right)\right|^{2}=x\end{split} for every fixed x>0,\ \sigma\geq\frac{1}{2}+\epsilon. 3.2. Now, in the special case 6 6 6 See ( 1.11 ). x\to\frac{x^{n}+y^{n}}{z^{n}} wee obtain the following Lemma from ( 3.6 ). Lemma 2 . (3.7) \begin{split}&\lim_{\tau\to\infty}\frac{1}{\tau}\left\{\int_{\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau}^{[\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau]^{1}}|\zeta(\sigma+it)|^{2}{\rm d}t\right\}^{5}\times\\ &\left\{\int_{\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau}^{[\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau]^{1}}\left|\zeta\left(\frac{1}{2}+it\right)\right|^{2}{\rm d}t\right\}^{-5}\times\\ &\sum_{t_{\nu}\geq\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau}^{\leq\frac{8\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau}\left|\zeta\left(\frac{1}{2}+it_{\nu}\right)\right|^{2}\left|\zeta\left(\frac{1}{2}+it_{\nu+1}\right)\right|^{2}=\frac{x^{n}+y^{n}}{z^{n}}\end{split} for every fixed Fermat’s rational and every fixed \sigma\geq\frac{1}{2}+\epsilon . Consequently, we have the following result Theorem 1 . The \zeta -condition (3.8) \begin{split}&\lim_{\tau\to\infty}\frac{1}{\tau}\left\{\int_{\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau}^{[\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau]^{1}}|\zeta(\sigma+it)|^{2}{\rm d}t\right\}^{5}\times\\ &\left\{\int_{\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau}^{[\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau]^{1}}\left|\zeta\left(\frac{1}{2}+it\right)\right|^{2}{\rm d}t\right\}^{-5}\times\\ &\sum_{t_{\nu}\geq\frac{4\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau}^{\leq\frac{8\pi^{5}}{3\zeta^{5}(2\sigma)}\frac{x^{n}+y^{n}}{z^{n}}\tau}\left|\zeta\left(\frac{1}{2}+it_{\nu}\right)\right|^{2}\left|\zeta\left(\frac{1}{2}+it_{\nu+1}\right)\right|^{2}\not=1\end{split} on the class of all Fermat’s rationals represents the next \zeta -equivalent of the Fermat-Wiles theorem for every fixed \sigma\geq\frac{1}{2}+\epsilon . 4. The second equivalent of the Fermat-Wiles theorem generated by our first asymptotic formula (1991) 4.1. Let us remind the following Selberg’s formula, see [ 15 ] , p. 130: (4.1) \begin{split}&\int_{T}^{T+H}|S_{1}(t)|^{2l}{\rm d}t=\bar{c}(l)H+\mathcal{O}\left(\frac{H}{\ln T}\right),\\ &T^{a}\leq H\leq T,\ \frac{1}{2}<a\leq 1,\ \bar{c}(l)>0,\ l\in\mathbb{N}.\end{split} We use in this section: (A) The formula ( 4.1 ) in the case (4.2) T=\overset{r-1}{T},\ H=\overset{r}{T}-\overset{r-1}{T}, that gives 7 7 7 See ( 2.5