Answers are grounded in this paper and cite their sources.
The uniqueness theorem for ambiguity functions states that ff waveformsu(t)andv(t)have the same ambiguity function, i.e.,χᵤ(τ, Δ) = X_{υ , Δ), thenu(t)andv(t)are identical except for a rotation, i.e.,v(t) = eiλu(t), where a real constant. Through the artifice of treating the even and odd parts of the waveforms, denotede(t)ando(t), respectively, correlative results have been obtained for the real and imaginary parts of ambiguity functions. Thus, ifRe \{χᵤ(τ, Δ)\} = Re \{χυ(τ, Δ)\}, theneυ(t) = ei λ eeᵤ(t)andoυ(t) = ei λ ooᵤ(t). FromRe \{χᵤ(τ, Δ)\}, the waveform classu(t) = eiλ [eᵤ(t) + eⁱᵏoᵤ(t)]may be constructed, but because of the arbitrary rotation,eⁱᵏ, a uniqueχᵤ-function is not determinable, in general. An important exception to this statement is the case whenχᵤ(τ, Δ)is real, andRe \{χᵤ\} = χᵤdetermines a unique waveform (within a rotation) and this waveform can only be even or odd. IfIm \{χᵤ(τ, Δ)\} = Im \{χυ(τ, Δ)\theneυ(t) =aei γeᵤ(t)andoυ(t) = 1/aeⁱʳoᵤ(t). IfIm \{χᵤ(τ, Δ)\}is given, { and}u(t)is known to have unit energy, then within rotations of the formei λ, only two possible waveform choices are possible foru(t). If it also is known which ofeᵤ(t)andoᵤ(t)has the greater energy, the functionIm \{χᵤ(τ, Δ)\}uniquely determinesu(t)(within a rotation) and the completeχᵤ-function. The results on magnitude/phase relationships include a formula which enables one to compute the squared magnitude of an ambiguity function as an ordinary two-dimensional correlation function. Self-reciprocal two-dimensional Fourier transforms are demonstrated for the product of the squared-magnitude function and either of the first partial derivatives of the phase function.
No takes yet. Share an insight, caveat, or question.
C. Stutt (1964) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: