Fourier Restriction: From Linear Restriction to Multilinear Restriction

This dissertation studies the Fourier restriction, which is to find the range of the constants p, q such that the L^q norm on a chosen subset of the Fourier domain is bounded above by the L^p norm in a spacial domain, up to some constant that is independent of the function. We discuss linear restriction, including Hausdorff-Young’s inequality, A proof of the restriction estimate on curves, and further discussions on the restriction problem on the sphere and paraboloid via the Stein-Tomas argument. We then discuss bilinear restriction, where the estimate on 2-dimensional case is proved by the reverse square function estimate and the bilinear interaction of transverse wave packets. The result is further used to verify the restriction conjecture on the 2-dimensional paraboloid. We discuss about multi-linear restriction in the final section, focusing on a short proof of a close result of the multilinear restriction estimate from I. Bejenaru.

💡 Research Summary

This dissertation provides a comprehensive treatment of the Fourier restriction problem, progressing from the classical linear theory to modern bilinear and multilinear extensions. The first part revisits the Fourier transform’s basic boundedness (‖ĥ‖∞ ≤ ‖h‖₁), Plancherel’s identity, and the Riesz‑Thorin interpolation theorem, culminating in the Hausdorff‑Young inequality (‖ĥ‖{L^q} ≤ C‖h‖{L^p} for 1 ≤ p ≤ 2, q = p′). Using scaling arguments, Knapp examples, and Khinchine’s inequality, the author establishes that on the whole space ℝⁿ the restriction operator R_{ℝⁿ}: f ↦ ĥ|_{ℝⁿ} is bounded from L^p to L^q if and only if 1 ≤ p ≤ 2 and q ≥ p′.

The second chapter extends these ideas to hypersurfaces, focusing on the sphere S^{n‑1} and the truncated paraboloid P^{n‑1}. By employing the Stein–Tomas method, the dissertation derives sufficient conditions for restriction estimates on these manifolds, showing that the admissible range expands compared with the whole space. In two dimensions, the author proves a sharp result for non‑degenerate curves: the restriction estimate holds precisely when p > 4 and 3/p + 1/q ≤ 1, using a careful change of variables, Jacobian estimates, Hölder’s inequality, and the Hardy–Littlewood–Sobolev inequality.

Chapter three is devoted to bilinear restriction. The central theorem is an L²‑based bilinear estimate of the form ‖E(f)·E(g)‖{L^q} ≲ ‖f‖{L^p}‖g‖_{L^p}, where E denotes the extension operator associated with a surface. The proof hinges on a reverse square‑function estimate and the analysis of transverse wave packets. By arranging wave packets so that their supports are essentially orthogonal, the author controls the interaction terms and obtains the desired bound. This bilinear machinery is then applied to the two‑dimensional paraboloid, yielding a complete verification of the restriction conjecture in that setting—an achievement that surpasses what can be obtained by linear methods alone.

The final chapter addresses multilinear restriction. Building on recent work by I. Bejenaru, the dissertation presents a concise proof of a near‑optimal multilinear estimate: for m ≥ 3 functions f₁,…,f_m, one has ‖∏{j=1}^m E_j f_j‖{L^{r}} ≲ ∏{j=1}^m ‖f_j‖{L^2} with r = 2/(m‑1). The argument uses a multilinear Kakeya‑type estimate and a transversality condition ensuring that the associated wave packets intersect minimally. By extending the reverse square‑function technique to the multilinear setting, the author demonstrates that multilinear restriction enjoys a larger admissible (p,q) region than its linear counterpart, a fact that has deep implications for decoupling theory, number theory, and nonlinear PDEs.

Overall, the dissertation not only consolidates classical linear restriction results but also showcases how bilinear and multilinear perspectives can resolve longstanding conjectures (e.g., the 2‑D paraboloid case) and push the frontier of restriction theory in higher dimensions. The methods introduced—reverse square‑function estimates, transverse wave‑packet analysis, and multilinear Kakeya techniques—are poised to become standard tools in future research on decoupling, dispersive equations, and related areas.