Phase Retrieval in C4 Requires Exactly Eleven Measurements

{ "core_finding": "The paper proves that the minimum number of intensity measurements required for phase retrieval in complex four-dimensional space is exactly eleven. This resolves a long-standing open problem by demonstrating that no set of ten vectors can uniquely distinguish all pure states in this dimension.", "caveats": "The result relies on the specific rank-one structure of the measurement operators and the topological properties of complex projective space, meaning it does not directly generalize to higher-rank measurements or different signal spaces.", "markdown": "## Research Question\n\nThe phase retrieval problem asks for the minimal number of measurements $N$ required to uniquely reconstruct a signal $x \\in \\mathbb{C}^M$ from intensity measurements $|\\langle a_j, x \\rangle|^2$. While it was known that $N=11$ is sufficient for $\\mathbb{C}^4$, it remained unclear whether a smaller set of 10 vectors could achieve the same property. This paper addresses whether $N=10$ is sufficient for phase retrieval in $\\mathbb{C}^4$, a question with direct implications for pure state quantum tomography and the number of orthonormal bases required to distinguish quantum states.\n\n## Approach\n\nThe author employs tools from differential topology, specifically characteristic classes and cohomology groups, to analyze the geometry of the intensity map. By assuming a ten-vector frame possesses the phase retrieval property, the author constructs a smooth embedding of the complex projective space $\\mathbb{CP}^3$ into $\\mathbb{R}^9$. The proof identifies a topological obstruction: the rank-one nature of the measurements forces a geometric splitting of the normal bundle over a projective hyperplane $\\mathbb{CP}^2$ that is incompatible with the first Pontryagin class of the bundle. This contradiction proves that no such ten-vector frame can exist.\n\n## Main Findings\n\nThe study establishes that the minimum number of measurements for phase retrieval in $\\mathbb{C}^4$ is exactly 11. This result confirms that three orthonormal bases are insufficient to distinguish all pure states in $\\mathbb{C}^4$, as such a configuration would imply the existence of a ten-vector phase-retrievable frame. Consequently, the author concludes that exactly four orthonormal bases are necessary and sufficient for this task, providing a definitive answer to a problem previously posed in the quantum information literature.\n\n## Why It Matters\n\nThis work settles a fundamental question in both harmonic analysis and quantum information theory. By establishing the exact lower bound for phase retrieval in $\\mathbb{C}^4$, it clarifies the limits of measurement efficiency in quantum state tomography. The topological approach used here also demonstrates the power of characteristic classes in solving problems that were previously resistant to purely algebraic or probabilistic methods.\n\n## Key Terms and Definitions\n\n- Phase Retrieval — The process of recovering an unknown signal from its intensity measurements, which are invariant under a global phase shift.\n- Parseval Frame — A set of vectors $\{b_j\}$ such that the identity operator can be expressed as the sum of the outer products of these vectors, ensuring a normalized measurement structure.\n- Pure State Quantum Tomography — The reconstruction of a quantum state represented by a rank-one density operator from a set of measurement outcomes.\n- Pontryagin Class — A characteristic class associated with a real vector bundle that provides topological invariants, used here to detect obstructions to the existence of specific embeddings.\n- Informationally Complete — A property of a set of measurements that allows for the unique identification of any quantum state within a given space." }