Near-field test method, system, readable storage medium and computer

A near-field test method, comprising steps of:

• S101, in a selected coordinate system, controlling a mover to cause a random relative motion between a device under test (DUT) and at least one probe to generate multiple random test points that satisfy the near field so as to determine one or more postures of the probe and electromagnetic field coefficients of the probe corresponding to the postures of the probe respectively;
• S102, obtaining measured values of electromagnetic field signals collected by the at least one probe so as to obtain a measured value set , ={mi(j):
• 1≤i≤N, 1≤j≤K}, where mi(j) represents the measured values of the electromagnetic field signals when the probe is at a j posture and at the test point pi;
• S103, calculating electromagnetic field coefficients v of the DUT:
• wherein a sparse feature of the electromagnetic field coefficients v of the DUT to be determined is a convex function which is represented by f(v);
• according to Lorenz reciprocity theorem, there is a random linear relationship Av= between the electromagnetic field coefficients v of the DUT and the measured value set , the random linear relationship Av= is used as a constraint or a penalty function in a domain optimization;
• combining the convex function f(v) and the random linear relationship Av= into an algorithm of convex optimization;
• when Av= is used as the constraint, the algorithm of convex optimization is written as Expression (1):
• Minimize f(v) subject to Av= Expression (1)
• when Av= is used as the penalty function, the algorithm of convex optimization is written as Expression (2):
• Minimize f(v)+P(Av-) Expression (2)
• where A is a random matrix, and elements of A are determined by positions of the random test points and the electromagnetic field coefficients of the probe at each posture;
• P(.) is a convex function, which increases with an increase of modulus of the vector Av-, Expressions (1) and (2) are solved by the algorithm of convex optimization, and a variable v is calculated and is determined as the electromagnetic field coefficients of the DUT;
• and
• S104, obtaining, according to the electromagnetic field coefficients v of the DUT, a far-field pattern of the DUT or an electric field and/or a magnetic field at any point outside the DUT.

A computer comprising:

• one or more processors;
• a memory;
• and
• one or more computer programs;
• wherein the processors and the memory are connected by a bus;
• the one or more computer programs are stored on the memory and configured to be executed by the one or more processors, when the processors execute the computer programs, perform a near-field test method, comprising steps of:
• S101, in a selected coordinate system, controlling a mover to cause a random relative motion between a device under test (DUT) and at least one probe to generate multiple random test points that satisfy the near field so as to determine one or more postures of the probe and electromagnetic field coefficients of the probe corresponding to the postures of the probe respectively;
• S102, obtaining measured values of electromagnetic field signals collected by the at least one probe so as to obtain a measured value set , ={mi(j):
• 1≤i≤N, 1≤j≤K}, where mi(j) represents the measured values of the electromagnetic field signals when the probe is at aj posture and at the test point pi;
• S103, calculating electromagnetic field coefficients v of the DUT:
• wherein a sparse feature of the electromagnetic field coefficients v of the DUT to be determined is a convex function which is represented by f(v);
• according to Lorenz reciprocity theorem, there is a random linear relationship Av= between the electromagnetic field coefficients v of the DUT and the measured value set , which is used as a constraint or a penalty function in a domain optimization, the convex function f(v) and the random linear relationship Av= are combined into an algorithm of convex optimization, where A is a random matrix, and elements of A are determined by positions of the random test points and the electromagnetic field coefficients of the probe at each posture;
• when Av= is used as the constraint, the algorithm of convex optimization is written as Expression (1):
• Minimize f(v) subject to Av= Expression (1)
• when Av= is used as the penalty function, the algorithm of convex optimization is written as Expression (2):
• Minimize f(v)+P(Av-) Expression (2)
• where P(.) is a convex function, which increases with an increase of modulus of the vector Av-, Expressions (1) and (2) are solved by the algorithm of convex optimization, and a variable v is calculated and is determined as the electromagnetic field coefficients v of the DUT;
• and
• S104, obtaining, according to the electromagnetic field coefficients v of the DUT, a far-field pattern of the DUT or an electric field and/or a magnetic field at any point outside the DUT.