Method and apparatus of simulating semiconductor circuit

Claim 2

A simulation apparatus for simulating a characteristic of a semiconductor circuit, comprising: a distribution information input means for receiving distribution information, wherein the distribution information includes a distribution profile of a variation of a device parameter or a process parameter of the semiconductor circuit and is adaptive to an actual distribution and wherein said distribution information input means can receive distribution information having both Gaussian and non-Gaussian distribution profiles; a random number generating means for generating a random number on the basis of a probability corresponding to the distribution profile of the distribution information received by the distribution information input means; a characteristic calculation means for calculating a dominant formula by using the random number to obtain a characteristic of the semiconductor circuit; and an output means for outputting the resultant characteristic of the semiconductor circuit.

Claim 3

A simulation apparatus according to claim 1, wherein the distribution profile of the distribution information is expressed by a numerical data series or a histogram.

Claim 4

A simulation apparatus according to claim 1, wherein said simulation apparatus comprises a digital computer.

Claim 7

A method according to claim 4, wherein the distribution profile of the distribution information is expressed by a numerical data series or a histogram.

Claim 8

A method according to claim 4, wherein the step of generating of the random number is carried out by a Monte Carlo method.

Claim 9

A simulation apparatus for simulating a characteristic of a semiconductor circuit, comprising: a distribution information input means for receiving distribution information, the distribution information including a profile, center and deviance of a distribution of a parameter variation of the semiconductor circuit, wherein the parameter variation corresponds to at least a device parameter or process parameter; a sampling means for sampling two points corresponding to each parameter, the two points being either a center point of the distribution and a point that is on the profile and spaced apart from the center point by a distance defined by a predetermined distribution ratio or two points that are on the profile, one on each side of the center point, and spaced apart from the center point by distances defined by a predetermined distribution ratio; a characteristic calculation means for calculating a dominant formula of the semiconductor circuit at the two sampled points to obtain two characteristics of the semiconductor circuit at the two points; a distribution approximation term calculation means for calculating an approximation term and deriving via a moment development approach a formula expressing the characteristic of the semiconductor circuit, wherein the formula is expressed as an average of powers of a value corresponding to the characteristic to be obtained; a distribution characteristic calculation means for calculating a variance or standard deviation of the characteristic of the semiconductor circuit over a full range of the parameter by using the approximation term from the distribution approximation term calculating means; and an output means for outputting the result of the distribution characteristic calculations.

Claim 10

A simulation apparatus for simulating a characteristic of a semiconductor circuit, comprising: a distribution information input means for receiving distribution information, the distribution information including a profile, center and deviance of a distribution of a parameter variation of the semiconductor circuit, wherein the parameter variation corresponds to at least a device parameter or process parameter; a sampling means for sampling two points corresponding to each parameter, the two points being either a center point of the distribution and a point that is on the profile and spaced apart from the center point by a distance defined by a predetermined distribution ratio or two points that are on the profile, one on each side of the center point, and spaced apart from the center point by distances defined by a predetermined distribution ratio; a characteristic calculation means for calculating a dominant formula of the semiconductor circuit at the two sampled points to obtain two characteristics of the semiconductor circuit at the two points; a distribution approximation term calculation means for calculating an approximation term and deriving via a moment development approach a formula expressing the characteristic of the semiconductor circuit, wherein the formula is expressed as an average of powers of a value corresponding to the characteristic to be obtained; a distribution characteristic calculation means for calculating a variance or standard deviation of the characteristic of the semiconductor circuit over a full range of the parameter by using the approximation term from the distribution approximation term calculating means; and an output means for outputting the result of the distribution characteristic calculations.

Claim 11

A simulation method for simulating a characteristic of a semiconductor circuit, including the steps of: receiving distribution information, the distribution information including a profile, center and deviance of a distribution of a parameter variation of the semiconductor circuit, wherein the parameter variation corresponds to at least a device parameter or process parameter: sampling two points corresponding to each parameter, the two points being either a center point of the distribution and a point that is on the profile and spaced apart from the center point by a distance defined by a predetermined distribution ratio or two points that are on the profile, one on each side of the center point, and spaced apart from the center point by distances defined by a predetermined distribution ratio; calculating a dominant formula of the semiconductor circuit at the two sampled points to obtain two characteristics of the semiconductor circuit at the two points; calculating an approximation term and deriving via a moment development approach a formula expressing the characteristic of the semiconductor circuit, wherein the formula is expressed as an average of powers of a value corresponding to the characteristic to be obtained; calculating a variance or standard deviation of the characteristic of the semiconductor circuit over a full range of the parameter by using the calculated approximation term; and outputting the result of the distribution characteristic calculation.

Claim 12

A simulation method for simulating a characteristic of a semiconductor circuit, including the steps of: receiving distribution information, the distribution information including a profile, center and deviance of a distribution of a parameter variation of the semiconductor circuit, wherein the parameter variation corresponds to at least a device parameter or process parameter: sampling two points corresponding to each parameter, the two points being either a center point of the distribution and a point that is on the profile and spaced apart from the center point by a distance defined by a predetermined distribution ratio or two points that are on the profile, one on each side of the center point, and spaced apart from the center point by distances defined by a predetermined distribution ratio; calculating a dominant formula of the semiconductor circuit at the two sampled points to obtain two characteristics of the semiconductor circuit at the two points; calculating an approximation term and deriving via a moment development approach a formula expressing the characteristic of the semiconductor circuit, wherein the formula is expressed as an average of powers of a value corresponding to the characteristic to be obtained; calculating a variance or standard deviation of the characteristic of the semiconductor circuit over a full range of the parameter by using the calculated approximation term; and outputting the result of the distribution characteristic calculation.

Claim 13

A recording medium storing software for simulating a characteristic of a semiconductor circuit, including the steps of: receiving a distribution information, wherein the distribution information includes a distribution profile of a variation of a device parameter or a process parameter of the semiconductor circuit and is adaptive to an actual distribution and wherein said distribution information can have both Gaussian and non-Gaussian distribution profiles; generating a random number on the basis of a probability corresponding to the distribution profile of the distribution information received; calculating a dominant formula by using the random number to obtain a characteristic of the semiconductor circuit and outputting the resultant characteristic of the semiconductor circuit; and outputting the resultant characteristic of the semiconductor circuit.

Claim 14

A recording medium storing software for simulating a characteristic of a semiconductor circuit, including the steps of: receiving a distribution information, wherein the distribution information includes a distribution profile of a variation of a device parameter or a process parameter of the semiconductor circuit and is adaptive to an actual distribution and wherein said distribution information can have both Gaussian and non-Gaussian distribution profiles; generating a random number on the basis of a probability corresponding to the distribution profile of the distribution information received; calculating a dominant formula by using the random number to obtain a characteristic of the semiconductor circuit and outputting the resultant characteristic of the semiconductor circuit; and outputting the resultant characteristic of the semiconductor circuit.

Claim 15

The recording medium of claim 9, wherein the distribution profile of the distribution information is expressed by a numerical data series or a histogram.

Claim 16

The recording medium of claim 9, wherein the generating of the random number is carried out by a Monte Carlo method.

Claim 17

A recording medium storing software for simulating a characteristic of a semiconductor circuit, including the steps of: receiving distribution information, the distribution information including a profile, center and deviance of a distribution of a parameter variation of the semiconductor circuit, wherein the parameter variation corresponds to at least a device parameter and/or process parameter; sampling two points corresponding to each parameter, the two points being either a center point of the distribution and a point that is on the profile and spaced apart from the center point by a distance defined by a predetermined distribution ratio or two points that are on the profile, one on each side of the center point, and spaced apart from the center point by distances defined by a predetermined distribution ratio; calculating a dominant formula of the semiconductor circuit at the two sampled points to obtain two characteristics of the semiconductor circuit at the two points; calculating an approximation term and deriving via a moment development approach a formula expressing the characteristic of the semiconductor circuit, wherein the formula is expressed as an average of powers of a value corresponding to the characteristic to be obtained; calculating a variance and/or standard deviation of the characteristic of the semiconductor circuit over a full range of the parameter by using the approximation term from the approximation term calculating step; and outputting the result of the distribution characteristic calculation.

Claim 18

A recording medium storing software for simulating a characteristic of a semiconductor circuit, including the steps of: receiving distribution information, the distribution information including a profile, center and deviance of a distribution of a parameter variation of the semiconductor circuit, wherein the parameter variation corresponds to at least a device parameter and/or process parameter; sampling two points corresponding to each parameter, the two points being either a center point of the distribution and a point that is on the profile and spaced apart from the center point by a distance defined by a predetermined distribution ratio or two points that are on the profile, one on each side of the center point, and spaced apart from the center point by distances defined by a predetermined distribution ratio; calculating a dominant formula of the semiconductor circuit at the two sampled points to obtain two characteristics of the semiconductor circuit at the two points; calculating an approximation term and deriving via a moment development approach a formula expressing the characteristic of the semiconductor circuit, wherein the formula is expressed as an average of powers of a value corresponding to the characteristic to be obtained; calculating a variance and/or standard deviation of the characteristic of the semiconductor circuit over a full range of the parameter by using the approximation term from the approximation term calculating step; and outputting the result of the distribution characteristic calculation.