## Abstract

In an excellent paper written in 1983 Van Hove et al. [IEEE Trans. Acoust. Speech Signal Process. ASSP-31, 1286 (1983)] proved that a real polynomial P(z), free of zeros in z = 1, is determined uniquely by its magnitude A(9) and the sign of its real part u(0) for z = e^{j0}. Van Hove et al. also indicated a possible extension to real polynomials in more than one variable and suggested an iterative procedure for the recovery of P(z). The restriction to polynomi¬als (or rational functions), real or otherwise, is not crucial. In fact, on due reflection, it is seen that the problem continues to make sense for functions H(z) = H(z_{1}, z_{2},…, z_{n}) in n variables that are holomorphic in the open polydisc U^{n} = {z: z_{l} < 1, z_{2} < 1,…, z_{n} < 1}, continuous in the closure U^{n} = [z: z_{1} < 1, z_{2} < 1,…, z_{n} < 1), and free of zeros on its distinguished boundary T^{n} = {z: z_{1} = 1, z_{2} = 1,…, z_{n} = 1}. The collection A(U^{n}) of functions holomorphic in Un and continuous in Un is Rudin’s polydisc algebra [W. Rudin, Function Theory in Polydiscs (Benjamin, New York, 1969)]. Our main result can be stated as follows. Let if (z) belong to A(U^{n}) and be free of ze¬ros in P1. For real θ = (θ_{l}, θ_{2},…,θ_{n}), write, in terms of real and imaginary parts, H(exp(jθ_{1}), exp(jθ_{2}),…, exp(jθ_{n})) = H(e^{j}^{θ}) = u(θ) + jv(θ), and let A(θ) = H(e^{j}^{θ}); then H(z) is determined uniquely by the functions A(0) and sign u(0), together with its value at one point z_{0} exp(jθ_{0}) ϵ T^{n} for which u(θ_{0}) ≠ 0. We also clarify the meaning and the scope of this theorem by the use of specific examples, but reconstruction techniques are not discussed.

Original language | English (US) |
---|---|

Pages (from-to) | 859-862 |

Number of pages | 4 |

Journal | Journal of the Optical Society of America A: Optics and Image Science, and Vision |

Volume | 6 |

Issue number | 6 |

DOIs | |

State | Published - Jun 1989 |

## ASJC Scopus subject areas

- Electronic, Optical and Magnetic Materials
- Atomic and Molecular Physics, and Optics
- Computer Vision and Pattern Recognition

## Fingerprint

Dive into the research topics of 'A uniqueness characterization in terms of signed magnitude for functions in the polydisc algebra A(U^{n})'. Together they form a unique fingerprint.