Input Nyquist sampling suffices to identify and compensate nonlinear systems

Tsimbinos, J.; Lever, K.V.

doi:10.1109/78.720389

Cited by 30 publications

(15 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3. As suggested by (12), given and , the sampling frequency increases with decreasing . Therefore, the upper bound of (i.e., ) for a stage constraint is reached when approaches its lower bound .…”

Section: Bandpass Sampling Requirements For Quadratically Nonlinmentioning

confidence: 82%

“…In addition, we notice that the obtained ranges of from all partitions for a given value of are disjoint. This is due to the fact that, given fixed and in (12), disjoint ranges of yield disjoint ranges of . Moreover, one can recognize that the range for partition with in Table VII is the range of for Nyquist sampling.…”

Section: Computer Simulationmentioning

confidence: 95%

“…Note that (2) and (12) have been used in deriving (25). Equivalently, (25) can be written as (26) We see that the different ranges of in (26) lead to different values of , which, in turn, yield different constraints on in (24).…”

Section: (25)mentioning

confidence: 99%

“…By using (8), one can obtain the stage 6 constraint on as follows: (17) where . (18) Again, (2) and (12) have been used in deriving (18). By substituting (18) into (17), one obtains the ranges of for stages , in Table IV.…”

Section: Bandpass Sampling Requirements For Cubically Nonlinear Smentioning

confidence: 99%

“…(2) and (12) have been used in deriving (16). By substituting (16) into (15), one obtains the ranges of for stages , in Table IV.…”

Section: Bandpass Sampling Requirements For Cubically Nonlinear Smentioning

confidence: 99%

See 4 more Smart Citations

A Universal Formula for the Complete Bandpass Sampling Requirements of Nonlinear Systems

Tseng

2009

IEEE Trans. Signal Process.

View full text Add to dashboard Cite

A nonlinear system driven by a bandpass input signal may produce an output signal that occupies multiple frequency bands. This makes bandpass sampling the output signal of the nonlinear system without causing aliasing a challenging task. Although the bandpass sampling theory for linear systems is well developed, its counterpart for nonlinear systems is relatively immature in the sense that complete bandpass sampling requirements have only been developed up to third-order nonlinear systems. In this paper, a novel method is used to derived the complete bandpass sampling requirements for nonlinear systems of an arbitrary order. The strategy used by this method is to build the constraints on the sampling frequency for th-order nonlinear systems upon those for ( 1)th-order nonlinear systems. This process makes the derivation easily extendable to nonlinear systems of any order. The derivation method also gives an insight into the mechanism for formularizing the constraints on the bandpass sampling frequency. As a result, a universal formula for the complete bandpass sampling requirements for nonlinear systems of any order is derived. This formula consolidates the bandpass sampling theory for nonlinear systems and facilitates the selection of the bandpass sampling frequency for nonlinear systems driven by bandpass input signals.

show abstract

Section: Bandpass Sampling Requirements For Quadratically Nonlinmentioning

confidence: 82%

Section: Computer Simulationmentioning

confidence: 95%

Section: (25)mentioning

confidence: 99%

Section: Bandpass Sampling Requirements For Cubically Nonlinear Smentioning

confidence: 99%

“…(2) and (12) have been used in deriving (16). By substituting (16) into (15), one obtains the ranges of for stages , in Table IV.…”

Section: Bandpass Sampling Requirements For Cubically Nonlinear Smentioning

confidence: 99%