--- title: 'Sidon Sets: Uniqueness and Applications' url: https://www.emergentmind.com/topics/sidon type: topic --- # Sidon Sets: Uniqueness and Applications Sidon denotes a cluster of uniqueness and interpolation phenomena that originated with additive \(B_2\) sequences and now span additive combinatorics, finite-field geometry, coding theory, harmonic analysis, and operator spaces. In its classical form, a Sidon set is a subset \(A\) of an abelian group such that \(a+b=c+d\) implies \(\{a,b\}=\{c,d\}\); modern variants include multiplicative Sidon sets in \(\mathbb N\), Sidon subsets of \(\mathbb F_2^t\), Sidon sets in dual groups of compact abelian groups, completely Sidon sets in discrete groups, and the \(q\)-analogues called Sidon spaces [1802.10511][2605.02064][2606.06669][1706.03844][2303.17306]. ## 1. Core meanings and definitions The unifying theme is uniqueness of representation under a binary or multilinear operation. In additive combinatorics, a Sidon set \(A\) in an abelian group satisfies \[ a+b=c+d \Rightarrow \{a,b\}=\{c,d\}. \] In \(\mathbb F_2^t\), the characteristic-\(2\) formulation is equivalent to requiring that the sum of any four distinct elements is never \(0\). In multiplicative number theory, a multiplicative Sidon set \(A\subseteq \mathbb N\) satisfies \[ ab=cd \Rightarrow \{a,b\}=\{c,d\}. \] In harmonic analysis on a compact abelian group \(G\) with dual group \(\Gamma\), a set \(\Lambda\subset \Gamma\setminus\{0\}\) is Sidon if there exists \(S_\Lambda<\infty\) such that \[ \sum_{\gamma\in A}|a_\gamma|\le S_\Lambda\left\|\sum_{\gamma\in A}a_\gamma\,\gamma\right\|_{L^\infty(G)} \] for every finite \(A\subseteq \Lambda\). In the \(q\)-analogue, an \(\mathbb F_q\)-subspace \(U\le \mathbb F_{q^n}\) is a Sidon space if for nonzero \(a,b,c,d\in U\), \[ ab=cd \Rightarrow a/c=d/b\in\mathbb F_q \] [1802.10511][2411.12911][2606.06669][2605.02064][2303.17306]. | Setting | Defining uniqueness property | Typical reformulation | |---|---|---| | Additive groups | \(a+b=c+d\Rightarrow \{a,b\}=\{c,d\}\) | pairwise sums are distinct | | \(\mathbb F_2^t\) | no four distinct elements sum to \(0\) | \(a+b=c+d\Rightarrow \{a,b\}=\{c,d\}\) | | Multiplicative subsets of \(\mathbb N\) | \(ab=cd\Rightarrow \{a,b\}=\{c,d\}\) | unordered products are distinct | | Dual groups \(\Gamma\) | \(\ell^1\)-\(L^\infty\) comparability on \(\Lambda\) | bounded interpolation on \(\Lambda\) | | Sidon spaces | \(ab=cd\Rightarrow a/c=d/b\in\mathbb F_q\) | \(\dim_{\mathbb F_q}(U\cap \alpha U)\le 1\) | A further abstraction replaces pair sums by a fixed linear form \(\varphi(x_1,\dots,x_h)=\sum_{i=1}^h c_i x_i\). A set \(A\) is then \(\varphi\)-Sidon if \(\varphi\) is injective on \(A^h\). In an infinite vector space, infinite \(\varphi\)-Sidon sets exist if and only if the coefficient set \(\{c_1,\dots,c_h\}\) has distinct subset sums, and the integer growth bounds become \(A(t)\ll t^{1/h}\) in general and \(A(t)\gg t^{1/(2h-1)}\) for explicit constructions [2101.01034]. ## 2. Additive combinatorics in integers and finite groups Classical Sidon theory is organized by the near-square-root bound. For a finite abelian group \(G\) of order \(n\), a Sidon set \(S\) of size \(k\) satisfies \(k(k-1)\le n-1\), so dense Sidon means \(k\approx \sqrt n\). A geometric synthesis asserts that all known dense Sidon subsets of abelian groups arise from projective planes through a common construction, and the desarguesian classification recovers the Singer, Bose, Spence, and Hughes–Cilleruelo families; the same framework also yields many examples from nondesarguesian planes and motivates the conjecture that all dense Sidon sets arise in this way [2107.05744]. Several extremal variants are now well developed. For Sidon systems of \(k\)-subsets of \([N]\), the maximum size \(F_k(N)\) satisfies \[ F_k(N)\le \binom{N-1}{k-1}+N-k, \] with \(F_2(N)=2N-3\), \(F_3(N)\ge N^2/2-O(N)\), and \(F_k(N)=\Omega_k(N^{k-1})\); in the random model \(\mathcal S(N,k,p)\), the threshold for being Sidon is \(p\asymp N^{-(2k+1)/4}\) [1802.10511]. For arbitrary \(n\)-element subsets of \(\mathbb Z\), if \(H(n)\) denotes the minimum possible size of the largest Sidon subset, then \[ H(n)\ge \left(\frac{1}{3\sqrt 3}+o(1)\right)\sqrt n, \] and the same lower bound holds uniformly for \(n\)-point subsets of \(\mathbb R^N\) [2605.03181]. The basis problem is a separate line of inquiry. Erdős asked for an infinite Sidon sequence that is an asymptotic basis of order \(3\). Modularly, for all sufficiently large \(N\), there exists a Sidon set in \(\mathbb Z_N\) that is a basis of order \(3\); in the integers, there exists an infinite \(B_2[2]\) sequence that is an asymptotic basis of order \(3\), and for every \(\varepsilon>0\) there exists a Sidon sequence that is an asymptotic basis of order \(3+\varepsilon\) in the sense that every sufficiently large integer is a sum of four elements of the sequence, one of them at most \(n^\varepsilon\) [1304.5351]. A complementary density result shows that there exists a Sidon set \(A\subset \mathbb N\) with \[ d(A+A+A)>0.064, \] so the triple sumset of a genuine Sidon sequence can already have positive lower density even though the order-\(3\) basis question remains open [2103.10349]. ## 3. Sidon sets in \(\mathbb F_2^t\), coding theory, and APN constructions Over \(\mathbb F_2^t\), Sidon sets are tightly linked to binary linear codes. If \(M\subseteq \mathbb F_2^t\setminus\{0\}\) is represented as the column set of a parity-check matrix \(H_M\), then the associated code \(C_M\) satisfies: \(M\) is sum-free iff \(d(C_M)\ge 4\), and \(M\) is sum-free Sidon iff \(d(C_M)\ge 5\). Consequently, the nonexistence of a binary \([n,n-t,5]\) code is equivalent to the upper bound \(s_{\max}(t)\le n\) for Sidon sets in \(\mathbb F_2^t\). The trivial estimate \[ \binom{|S|}{2}\le 2^t-1 \] is sharpened to \(s_{\max}(t)\le 2^{(t+1)/2}-2\) for odd \(t\ge 7\), and to an explicit even-\(t\) bound of the form \[ s_{\max}(t)\le \left\lfloor \sqrt{2^{t+1}+0.5}\right\rfloor-\lambda_{a,b,\varepsilon} \] for \(t\ge 6\). The same paper classifies maximal Sidon sets up to affine equivalence for \(t\le 6\), proves that every maximal Sidon set in \(t=7\) has size \(12\), and shows that the possible sizes in \(t=8\) are \(15,16,18\) [2304.07906]. A second major theme identifies Sidon sets with graphs of cryptographically strong functions. For \(F:\mathbb F_{2^m}\to \mathbb F_{2^m}\), the graph \[ G(F)=\{(x,F(x)):x\in \mathbb F_{2^m}\}\subseteq \mathbb F_2^{2m} \] is Sidon if and only if \(F\) is APN. Intersecting \(G(F)\) with a hyperplane corresponding to a maximizing Walsh coefficient produces a Sidon set in \(\mathbb F_2^{2m-1}\) of size \[ 2^{m-1}+\mathrm{Lin}(F)/2. \] Using 8-bit APN functions of linearity \(128\), this yields Sidon sets of size \(192\) in \(\mathbb F_2^{15}\), improving the previous best-known size \(152\), and it produces a binary \([191,176,5]\) code. The same analysis proves the general APN bound \(\mathrm{Lin}(F)\le 2^m-4\) [2411.12911]. The cryptographic analogy has recently been pushed further. For Sidon sets \(S\subseteq \mathbb F_2^n\), minimal linearity is characterized exactly by the \(k\)-cover property: every \(p\in \mathbb F_2^n\setminus S\) has exactly \(k\) representations as \(x+y+z=p\) with \(\{x,y,z\}\subseteq S\). This classification is recast through the Cayley graph of the Boolean function \(\gamma_S\), yielding strongly regular graphs; in particular, the unique rank-\(3\) strongly regular graph with parameters \((2048,276,44,36)\) is realized as such a Cayley graph, and the best-known lower bound for the largest Sidon set in \(\mathbb F_2^{4t+1}\) is improved by \(1\) for all \(t\ge 4\) [2501.11184]. Related work on polynomial graphs in \(\mathbb F_q\times \mathbb F_q\) shows that planar polynomials make \((x,P(x))\) a maximum Sidon set, while various monomial and cubic candidates over \(\mathbb F_p\) are excluded by exact multiplicity criteria [2111.08886]. ## 4. Harmonic-analytic, probabilistic, and operator-space formulations In harmonic analysis, Sidon sets are interpolation sets for Fourier coefficients. For a compact abelian group \(G\) with dual \(\Gamma\), a subset \(\Lambda\subset \Gamma\setminus\{0\}\) is Sidon precisely when bounded data on \(\Lambda\) extend to Fourier transforms of bounded continuous functions or bounded measures with uniform control. In dual groups of bounded torsion, this interpolation property is equivalent to a purely combinatorial decomposition statement: \[ \Lambda \text{ is Sidon } \iff \Lambda \text{ is a finite union of quasi-independent sets.} \] The final missing prime-power case is proved by combining Pisier’s proportional quasi-independence theorem, Bourgain’s projection theorem, a local dimension bound for reduced signed relations, and a Rado–Horn support-partition argument; in prime-power groups, the number of quasi-independent pieces is bounded by \[ k\le \left\lceil \frac{s\log_2(p)}{\delta}\right\rceil \] [2606.06669]. For torsion-free dual groups, Sidon sets admit an even sharper proportional structure. A Sidon set is proportionally \(n\)-degree independent for every \(n\), and for every \(\varepsilon>0\) one can extract proportionally large subsets with Sidon constant at most \(1+\varepsilon\), the minimum possible value. The argument combines Pisier’s proportional characterization, higher-order independence, and Riesz-product interpolation; in torsion groups such uniform small constants can fail [1808.03128]. The same vocabulary extends beyond characters. Any uniformly bounded \(\psi_2(C)\)-orthonormal system is \(2\)-fold tensor Sidon, improving Bourgain–Lewko’s \(5\)-fold result, and a bounded system is randomly Sidon if and only if it is \(4\)-fold tensor Sidon, equivalently \(k\)-fold tensor Sidon for some or all \(k\ge 4\). Pisier’s proof uses Talagrand’s majorizing measure theorem, Gaussian domination, and a tensor decomposition \[ \sum_{n=1}^N \varphi_n^1\otimes \varphi_n^2=t+r, \qquad \|t\|_\wedge\le w(\varepsilon),\quad \|r\|_\vee\le \varepsilon \] [1602.02430]. A noncommutative version replaces classical Banach-space Sidonicity by operator-space structure. A subset \(\Lambda\) of a discrete group \(G\) is completely Sidon if its span in \(C^*(G)\) is completely isomorphic to \(\ell_1(\Lambda)\) with maximal operator space structure. Completely Sidon sets are stable under finite unions, bounded Hermitian functions on symmetric completely Sidon sets extend to positive definite functions, and \(\Lambda\) is completely Sidon if and only if the dual operator space \((C_\Lambda)^*\) is exact [1706.03844]. ## 5. Sidon spaces and \(q\)-analogues Sidon spaces are the \(q\)-analogues of Sidon sets. An \(\mathbb F_q\)-subspace \(U\le \mathbb F_{q^n}\) is Sidon exactly when \[ ab=cd \Rightarrow a/c=d/b\in \mathbb F_q \] for nonzero \(a,b,c,d\in U\), or equivalently when \[ \dim_{\mathbb F_q}(U\cap \alpha U)\le 1 \qquad (\alpha\in \mathbb F_{q^n}\setminus \mathbb F_q). \] Roth, Raviv, and Tamo showed that this is equivalent to saying that the cyclic orbit \(\mathrm{Orb}(U)=\{\alpha U:\alpha\in \mathbb F_{q^n}^\times\}\) has full size \((q^n-1)/(q-1)\) and minimum subspace distance \(2k-2\), making Sidon spaces precisely the optimal one-orbit cyclic subspace codes of that type [2303.17306]. A large class of constructions lives inside sums of two multiplicative cosets of a fixed subfield. Writing \(F_0=\mathbb F_{q^k}\subseteq \mathbb F_{q^n}\), one considers \[ V_{U,\gamma}=\{u+v\gamma:(u,v)\in U\}\subseteq F_0+\gamma F_0, \] or, in graph form, \[ V_{f,\gamma}=\{u+f(u)\gamma:u\in F_0\} \] for a linearized polynomial \(f\). The paper “Constructions and equivalence of Sidon spaces” characterizes when \(V_{U,\gamma}\) is Sidon through an intrinsic “Sidon space property” for \(U\subseteq F_0^2\), proves that scattered polynomials yield Sidon space polynomials, establishes a sharp monomial criterion in degree-\(2\) extensions, gives a direct-sum construction, and formulates semilinear equivalence in terms of \(\Gamma L\)-actions and \(GL(2,F_0)\)-orbits [2303.17306]. The generalized theory replaces pairwise multiplicative uniqueness by \(r\)-fold multiplicative uniqueness. An \(r\)-Sidon space is a subspace \(V\) such that equality of two \(r\)-fold products of nonzero elements forces equality of the multisets of one-dimensional \(\mathbb F_q\)-subspaces. If \(V_r\) denotes the \(r\)-span, then \[ \dim_{\mathbb F_q}(V_r)\le \min\left\{n,\binom{k+r-1}{r}\right\}, \] and equality implies that \(V\) is \(r\)-Sidon. The same framework gives lower bounds on \(\dim(V_s)\), prime-degree strengthenings, constructions from \(B_r\)-sets and irreducible polynomials, and a scattered-polynomial construction \[ V_{f,y}=\{u+f(u)y:u\in \mathbb F_{q^k}\} \] which is \(r\)-Sidon whenever \(f\) is scattered and \(y\) generates an extension of degree at least \(r+1\) [2312.12245]. These spaces are motivated not only by combinatorics but also by random linear network coding, because one-orbit cyclic subspace codes derived from Sidon spaces achieve large size together with minimum distance \(2k-2\) [2312.12245]. ## 6. Multiplicative, geometric, and current developments The multiplicative analogue asks for subsets of \(\{1,\dots,n\}\) with unique unordered products. If \(g(n)\) denotes the least \(L\) such that some multiplicative Sidon set \(A\subseteq\{1,\dots,n\}\) meets every interval \([x,x+L]\subseteq[1,n]\), then primes show that multiplicative Sidon sets can be large, while Erdős proved that the maximum size is \[ \pi(n)+\Theta\!\big(n^{3/4}(\log n)^{3/2}\big). \] Recent work answers Sárközy’s square-root question by proving \[ g(n)\le \lfloor \sqrt n\rfloor \qquad (n\in\mathbb N), \] via the explicit congruence class construction \(A=\{a\le n:a\equiv 1\pmod q\}\) with \(q=\lfloor \sqrt n\rfloor\), and then improves this asymptotically to \[ g(n)\ll_\varepsilon n^{\rho+\varepsilon},\qquad \rho=\frac{13-\sqrt{69}}{10}<0.47, \] using primes in short intervals, a private-prime Sidon criterion, and a weighted Hall matching argument. A notable auxiliary feature is that the square-root proof was autonomously discovered and formally verified in Lean by Aristotle, and the power-saving argument was formalized in Lean under its analytic inputs [2605.02064]. A subsequent refinement lowers the exponent dramatically: \[ g(n)\ll_\varepsilon n^{10/33+\varepsilon}. \] The proof introduces a Lovász Local Lemma selection principle for choosing one element from each of many large local sets while preserving multiplicative Sidonicity, then combines it with exceptional-set bounds for primes in short intervals. Under Lindelöf- or RH-type estimates one further gets the conditional bound \(g(n)\ll_\varepsilon n^{1/4+\varepsilon}\) [2606.07428]. This suggests that the exact order of \(g(n)\), currently bracketed between logarithmic lower bounds and subpower upper bounds, remains one of the central open problems in the multiplicative theory. Sidon sequences also admit a geometric incarnation in CAT(0) geometry. Given a finite Sidon sequence \(a_0