Generalized Bilinear Bochner-Riesz Operator

• The generalized bilinear Bochner-Riesz operator is a bilinear Fourier multiplier that extends classical Bochner-Riesz means with innovations such as logarithmic and fractional modifications.
• It utilizes techniques like dyadic decomposition, Fourier series expansion, and sparse domination to establish dimension-free boundedness and critical L² × L² → L¹ estimates.
• The framework underpins maximal and square-function variants that influence spectral multipliers and endpoint estimates in both Euclidean and non-Euclidean settings.

A generalized bilinear Bochner-Riesz operator is a bilinear (or multilinear) Fourier multiplier whose defining symbol is a nontrivial function of the Euclidean norms, generalizing the classical Bochner-Riesz means to bilinear or multilinear contexts. These operators naturally interpolate between linear Bochner-Riesz means, bilinear ball multipliers, and more exotic variants involving logarithmic or fractional power modifications. Their boundedness properties, smoothness thresholds, and maximal (as well as square-function) variants constitute a rich area in harmonic analysis, with deep connections to multilinear Calderón-Zygmund theory, spectral multipliers on Lie groups, and the geometry of non-Euclidean spaces.

1. Definitions and Variants

The generalized bilinear Bochner-Riesz operator is defined via a Fourier multiplier: $$T_\sigma(f,g)(x) = \iint_{\mathbb{R}^n \times \mathbb{R}^n} \sigma(\xi,\eta)\, \widehat{f}(\xi)\, \widehat{g}(\eta)\, e^{2\pi i x\cdot(\xi+\eta)}\, d\xi\,d\eta$$ Where the symbol $\sigma$ is typically radial in $(|\xi|,|\eta|)$ or a function of $|\xi|^2 + |\eta|^2$; standard special cases include:

• Classical (standard) bilinear Bochner-Riesz: $$\sigma^\lambda(\xi,\eta) = (1 - |\xi|^2 - |\eta|^2)^\lambda_+$$
• Modified/logarithmic variant: $$\sigma^{\lambda,\gamma}(\xi,\eta) = \frac{(1 - |\xi|^2 - |\eta|^2)_+^\lambda}{(1 - \log(1 - [|\xi|^2 + |\eta|^2]))^\gamma}$$
• Generalized power structure: $$\mathcal{B}_{\delta,\lambda}(f,g)(x) = \iint (1 - (|\xi|^2 + |\eta|^2)^\lambda)_+^\delta \,\widehat{f}(\xi)\,\widehat{g}(\eta) e^{2\pi i x\cdot(\xi+\eta)} d\xi d\eta$$

Maximal and square-function variants include

• Maximal operator: $T^*(f,g)(x) = \sup_{R>0} |T_{\sigma_R}(f,g)(x)|$
• Stein-type square function: $$\mathcal{G}^\delta(f,g)(x) = \left(\int_0^\infty |B_{\delta+1,R}(f,g)(x)|^2 R\,dR\right)^{1/2}$$

These objects generalize further to spectral multipliers associated with sub-Laplacians or differential operators on Lie groups and degenerate spaces (Honzík et al., 14 Jan 2026, Bagchi et al., 24 May 2025, Bagchi et al., 6 Apr 2025).

2. Main Boundedness Theorems

The central analytic question is to determine, for fixed exponents $1\leq p_1, p_2 \leq \infty$, when the operator $T_\sigma$ is bounded from $L^{p_1}\times L^{p_2}$ into $L^p$ with $1/p = 1/p_1 + 1/p_2$. The principal results are:

• For standard Bochner-Riesz with $\lambda>0$:
  • $T_{\sigma^\lambda}$ is bounded $L^2 \times L^2 \to L^1$ as soon as $\lambda > 0$ (Bernicot et al., 2012, Honzík et al., 14 Jan 2026).
  • For multilinear case ($m$-linear), $\lambda > \frac{m}{2}-1$ suffices for boundedness $L^2\times\cdots\times L^2 \to L^{2/m}$ (Honzík et al., 14 Jan 2026).
• For generalized (logarithmic) Bochner-Riesz:
  • $$T_{\sigma^{0,\gamma}}: L^2 \times L^2 \rightarrow L^1$$ whenever $\gamma > 1$ (Honzík et al., 14 Jan 2026).
• Critical Sobolev-type condition:
  • If $$\sup_j \|\sigma(2^j\cdot)\Phi\|_{L^2_{1/2+\epsilon}(\mathbb{R}^{2n})} < \infty$$ for some cutoff $\Phi$, then $T_\sigma:L^2 \times L^2 \to L^1$ uniformly in $n$ (Honzík et al., 14 Jan 2026, Choudhary et al., 2021).
• Generalized degree variant:
  • For $\mathcal{B}_{\delta,\lambda}^*(f,g)$, $L^{p_1}\times L^{p_2}\to L^p$ boundedness holds whenever $\delta > \alpha^+(p_1, p_2) + 1/2$ where $\alpha^+(p_1,p_2)$ is an explicit critical index, generalizing the classical Bochner-Riesz exponents (Choudhary et al., 2021).

The sharpness of the $s > 1/2$ threshold for radial multipliers is established, with explicit counterexamples at $s = 1/2$ (Honzík et al., 14 Jan 2026).

3. Symbol Smoothness: Dimension-Free and Sobolev Conditions

A distinct feature is the identification of precise regularity requirements on the bilinear (or multilinear) multiplier symbol for dimension-free boundedness: $$\sup_{j\in\mathbb{Z}} \|\sigma(2^j \cdot)\Phi\|_{L^2_{1/2 + \varepsilon}(\mathbb{R}^{2n})} < \infty \implies T_\sigma: L^2 \times L^2 \to L^1$$ This replaces high-order Sobolev requirements from previous works with a threshold independent of $n$, and crucially leverages radiality of $\sigma$. The main technical tools involve dyadic decompositions and Fourier series expansions for shell-localized pieces of $\sigma$, culminating in a reduction to summing products of linear $L^2$-bounded multipliers (Honzík et al., 14 Jan 2026).

For other ranges $L^{p_1}\times L^{p_2}\to L^p$ with $p_i\neq 2$, this method does not reach presumed optimality, indicating the necessity for new approaches.

4. Methodologies and Proof Strategies

The foundational strategies across recent research include:

• Littlewood-Paley and dyadic decompositions: Partition both frequency and space to localize the action of $\sigma$, permitting precise estimates on each piece.
• Product decomposition via Fourier series: Radiality of the symbol permits expansions into products of linear multipliers, reducing the multilinear analysis to (generalized) Coifman-Meyer theory (Honzík et al., 14 Jan 2026).
• Hardy-Littlewood maximal and square-function controls: Off-diagonal or nonlocal terms are managed with classical maximal function bounds; diagonal pieces employ $L^2$ square function estimates or orthogonality.
• Multilinear interpolation: Real and complex interpolation fill out the full triangle of exponents, using endpoint estimates at "corner" tuples.
• Sparse domination techniques: Endpoint and weak-type inequalities are handled by (bi-)sparse form controls (Choudhary et al., 2021).

The combination of these tools enables a sharp dimensional-independent $L^2\times L^2 \to L^1$ theory for a wide class of bilinear multipliers and their maximal analogues.

5. Maximal and Square-Function Variants

Maximal versions

$T^*(f,g)(x) = \sup_{R>0} |T_{\sigma_R}(f,g)(x)|$

and bilinear Stein-type square functions

$$\mathcal{G}^\delta(f,g)(x) = \left( \int_0^\infty |B_{\delta+1,R}(f,g)(x)|^2 R\,dR \right)^{1/2}$$

are central for fine convergence results and further regularity.

Key facts include:

• For maximal operators generated by symbols satisfying the Sobolev-type decay, the same $L^2\times L^2 \to L^1$ boundedness holds, with no dimensional dependence (Honzík et al., 14 Jan 2026, Choudhary et al., 2021).
• Maximal generalized Bochner-Riesz operators, as in $\sup_R |\mathcal B_{\delta,\lambda,R}(f,g)|$, are controlled by square-function estimates and Hardy-Littlewood maximal inequalities; the critical index for $\delta$ in $L^{p_1}\times L^{p_2}\to L^p$ theory is increased by $1/2$ over the square-function threshold (Choudhary et al., 2021).

These results establish a hierarchy:

• Square functions require $\delta > \alpha^+(p_1,p_2)$,
• Maximal functions require $\delta > \alpha^+(p_1,p_2) + 1/2$.

6. Applications, Optimality, and Open Problems

Applications of the generalized bilinear Bochner-Riesz framework include:

• Sharp maximal function control for dimension-free classes of radial multipliers, including bilinear spherical maximal and fractional Schrödinger multipliers.
• Spectral multipliers for subelliptic operators: Generalizations to operators on Métivier and Grushin groups, with analogous thresholds replacing Euclidean dimension by the topological dimension (Bagchi et al., 6 Apr 2025, Bagchi et al., 24 May 2025).
• Endpoint counterexamples: For radial multipliers, boundedness at the $s = 1/2$ threshold fails for $L^2 \times L^2 \to L^1$ (Honzík et al., 14 Jan 2026).

Open questions and conjectures:

• Determination of the precise critical indices for $L^{p_1}\times L^{p_2}\to L^p$ for the full range of exponents, especially for $p_i\ne 2$.
• Extension of dimension-free criteria to higher-order multilinear settings and other group symmetries.
• Endpoint and weak-type boundedness, particularly at the critical index and on $L^1$.

7. References and Connections

Key references include:

• Bagchi–Molla–Singh on Métivier groups (Bagchi et al., 6 Apr 2025).
• Bernicot–Grafakos–Song–Yan on the classical bilinear Bochner-Riesz problem (Bernicot et al., 2012).
• Sharp $L^2\times L^2\to L^1$ and symbol regularity theory (Honzík et al., 14 Jan 2026, Choudhary et al., 2021).
• Weighted and endpoint analysis in critical regimes (Jotsaroop et al., 2020, Choudhary et al., 2022).

The generalized bilinear Bochner-Riesz operator theory stands at the interface of harmonic analysis, PDE spectral theory, and the structure of non-Euclidean and nilpotent Lie groups, encapsulating a broad generalization of classical summability and maximal function theory in modern Fourier analysis.