Calibration of vector modulators

Claim 1

A method for quadrature calibration of a vector modulator, the method being characterized by the steps of: generating a periodic first local oscillator signal A in,1 of predetermined frequency f c and a second local oscillator signal A in,2 that is equal to the first local oscillator signal shifted in phase by approximately 90 degrees, where these two signals are independently adjustable in phase; providing independent, adjustable first and second carrier leakage compensation dc signals of adjustable amplitudes A clc,1 and A clc,2 , respectively; providing independent first and second offset dc signals of adjustable amplitudes A off,1 and A off,2 , respectively; providing independent first and second modulation signals of adjustable amplitudes A mod,1 from data input signal I m , and A mod,2 from data input signal Q m , respectively; adding the first carrier leakage compensation signal A clc,1 and the first offset signal A off,1 to the first modulation signal A mod,1 to produce a first compensated modulation input signal A comp,1 ; adding the second carrier leakage compensation signal A clc,2 and the second offset signal A off,2 to the second modulation signal A mod,2 to produce a second compensated modulation input signal A comp,2 ; multiplying the first local oscillator signal by the first compensated modulation input signal to produce a first modulation output signal of maximum amplitude A out,1 = A in,1 ·A comp,1 ; multiplying the second local oscillator signal by the second compensated modulation input signal to produce a second modulation output signal of maximum amplitude A out,2 = A in,2 ·A comp,2 ; adding the signals A out,1 and A out,2 together to form a vector sum output signal A out ; expressing the vector sum output signal A out in terms of two-dimensional coordinates (A out,1 , A out,2 ), for A clc,1 , A clc,2 , A off,1 and A off,2 each set equal to a respective selected signal value and for the following choices of first and second modulation input amplitudes (A out,1,q , A out,2,q )(q = 1, 2, 3, 4, 5, 6, 7, 8) q=1 A mod,1 = r 1,1 I A mod,2 = r 1,2 Q , q=2 A mod,1 = r 3,1 I A mod,2 = r 1,2 Q , q=3 A mod,1 = r 3,1 I A mod,2 = r 3,2 Q , q=4 A mod,1 = r 1,1 I A mod,2 = r 3,2 Q , q=5 A mod,1 = r 2,1 I A mod,2 = r 1,2 Q , q=6 A mod,1 = r 3,1 I A mod,2 = r 2,2 Q , q=7 A mod,1 = r 2,1 I A mod,2 = r 3,2 Q , q=8 A mod,1 = r 1,1 I A mod,2 = r 2,2 Q , where I and Q are predetermined, non-Zero signal values, and where r 1,1 , r 2,1 and r 3,1 are predetermined values of a first attenuation parameter Atten₁ satisfying 0 ≦ r 1,1 < r 2,1 < r 3,1 and r 1,2 , r 2,2 and r 3,2 are predetermined values of a second attenuation parameter Atten₂ satisfying 0 ≦ r 1,2 < r 2,2 < r 3,2 , and representing these eight coordinate pairs on a two-dimensional graph to form a quadrilateral with four consecutive vertices (A out,1,q , A out,2,q ) (q=1,2,3,4) and four intermediate boundary points for the quadrilateral (A out,1,q , A out,2,q ) (q=5,6,7,8); for a vertex point ((A v,1 ,A v,2 ) corresponding to the output signal A out being zero in the plane of the quadrilateral, determining the distances L qv between this vertex and each of the points (A out,1,q , A out,2,q ) (q=1,2,4,5,8); determining the lengths L₁₂ and L₁₄ of the sides of the quadrilateral extending from (A out,1,1, , A out,2,1, ) to (A out,1,2, , A out,2,2, ) and from (A out,1,1, , A out,2,1 ) to (A out,1,4, , A out,2,4 ), respectively, on the graph and the angle ψ₂₁₄ between the intersection of these two quadrilateral sides; determining the value of the quadrature angular error Δψ according to the relation ΔV = ψ₂₁₄ - ψ₀ , where ψ₀ is ± 90 degrees; adjusting the phase shift between the first local oscillator signal and the second local oscillator signal to reduce the magnitude of the quadrature angular error Δψ to a minimum; adjusting at least one of the attenuation parameter values Atten₁ and Atten₂ so that the lengths L₁₂ and L₁₄ are of ratio R when the input signals have equivalent values, and fixing these attenuation parameter values; with the intermediate signals A mod,1 and A mod,2 temporarily set equal to zero, with offset signals A off,1 and A off,2 temporarily set to zero, and with the quadrature angle ψ₂₁₄ set equal to ψ₀, adjusting the first and second carrier leakage compensation signals A clc,1 and A clc,2 so that the magnitude of the total output signal A out,1 + A out,2 is minimized, and fixing the compensation signals A clc,1 and A clc,2 at these values; adjusting the first and second offset signals A off,1 and A off,2 so that the amplitude of the total output signal A out,1 + A out,2 is approximately the same when the data input signals cause A mod,1 and A mod,2 to be both at a minimum as when the data input signals cause A mod,1 and A mod,2 to be both at a maximum, and so that the amplitude of the total output signal A out,1 + A out,2 is approximately the same when the data input causes A mod,1 to be at a maximum and A mod,2 to be at a minimum as when the data input signals cause A mod,1 to be at a minimum and A mod,2 to be at a maximum, and fixing the offset signals A off,1 and A off,2 at these values; whereby the vector modulator is calibrated to approximately minimize quadrature phase angle error, to approximately minimize I to Q channel gain ratio error, and to approximately minimize carrier leakage in the output signals of the vector modulator.

Claim 10

A method of calibrating a vector modulator having a signal approximately in phase to an input uncalibrated RF source (13) and a signal approximately 90° out-of-phase, or in quadrature, to the input RF source (13), the in-phase signal is modified by a first modulation signal I m and a first offset signal (37) and a first leakage signal (39) through a first mixer (19), the quadrature-phase signal is modified by a second modulation signal Q m and a second offset signal (47) and a second leakage signal (49) through a second mixer (23), the modified in-phase signal is combined with the modified quadrature-phase signal as the output signal, the output signal is measured by a calibrated scalar detector (27), comprising the following steps: (a) measuring the phase between the in-phase signal and the quadrature-phase signal of the uncalibrated RF source (13) using a calibrated scalar detector (27); (b) measuring and adjusting the phase between the in-phase signal and the quadrature-phase signal till the phase is as close to 90° as possible; (c) selectively introducing attenuation into the two paths from the inputs of the modulation signals I m and Q m to the calibrated scalar detector (27) to equalize the attenuation of each path; (d) selectively introducing leakage signal (39,49) to the IF ports of the first and second mixers (19,23) to minimize the output signal when the first and second offset signals (37,47) and the first and second modulation signals are nulled; (e) selectively introducing offset signal (37,47) to equalize the output signal (i) when the second modulation signal is maximally positive and maximally negative with the first modulation signal nulled and (ii) when the first modulation signal is maximally positive and maximally negative with the second modulation signal nulled; and (f) setting the phase between the in-phase signal and the quadrature-phase signal to any predetermined value.