The Volterra series expansion of the response of a non-linear system is described, along with its counterpart in the frequency domain. Cross-correlation methods for identifying the kernel functions which occur in this expansion are reviewed, with particular emphasis on techniques for obtaining the linear approximant to a non-linear system. Some recent work which appears to be unrelated to the Volterra approach is also discussed.
Get full access to this article
View all access options for this article.
References
1.
GodfreyK R: ‘The theory of the correlation method of dynamic analysis and its applications to industrial processes and nuclear power plant’, Measurement and Control1969, 2, pp T65–72.
2.
NikiforukP NGuptaM M: ‘A bibliography on the properties, generation and control system applications of shift register sequences’, Int J Control, 1969, Vol 9, No 2, pp 217–234.
3.
VolterraV: ‘Theory of Functionals’. (Blackie & Son, London, 1930).
4.
VolterraV: ‘Theory of Functionals and of Integral and Integro-Differential Equations’. (Dover Publications Inc, New York, 1959).
5.
WienerN: ‘Non-linear Problems in Random Theory’. (Chapman and Hall, 1958).
6.
FlakeR H: ‘Volterra series representation of non-linear systems’. IEEE Trans, Applications and Industry, 1963, Vol 81, pp 330–335.
LubbockJ KBandalV S: ‘Multi-dimensional Laplace Transforms for solution of non-linear equations’. Proc IEE, Vol 116, No 12, Dec 1969, pp. 2075–2082.
9.
GeorgeD A: ‘Continuous non-linear systems’. Report No 355, Research Laboratory of Electronics, MIT, 1959.
10.
ReddyD CJaganN C: ‘Multi-dimensional transforms: new technique for the association of variables’, Electron Lett1971, 7, pp. 278–279.
11.
KadriF L: ‘Non-linear plant identification by cross-correlation’. IFAC Symposium on digital simulation of continuous processes, Gyor, Hungary, Sept 1971.
12.
KadriF L: ‘The cross-correlation function characteristics of non-linear systems subjected to pseudo-random sequences’. M Eng Thesis, UWIST, Cardiff, 1971.
13.
BruleJ D: ‘A note on the Vandermonde determinant’. IEEE Trans, 1964, AC-9, pp 314–315.
14.
EconomakosE: ‘Identification of a group of internal signals of zero-memory non-linear systems’, Electron Lett1971, 7, pp. 99–100.
15.
GoodmanT P: ‘Determination of the characteristics of multi-input and non-linear systems from normal operating records’. Trans ASME, 1957, pp 567–575.
16.
GardinerA B: ‘Elimination of the effect of non-linearities on process cross-correlations’, Electron Lett1966, 2, pp. 164–165.
17.
GardinerA B: ‘Identification of the linear channel of a non-linear process using pseudo-random cross-correlation techniques’. IEE Colloquium on pseudo-random sequences applied to control systems. London, 1967.
18.
GardinerA B: ‘Determination of the linear output signal of a process containing single-valued non-linearities’, Electron Lett1968, 4, pp. 224–226.
19.
GodfreyK R: ‘Three level m-sequences’, Electron Lett1966, 2, pp. 241–243.
20.
GyftopoulosE PHooperR J: ‘Signals for transfer function measurements in non-linear systems’. Proc of Conf Noise Analysis in Nuclear Systems, Gainsville, Fla, 1965, pp 335–345.
21.
ReamN: ‘Non-linear identification using inverse repeat m-sequences’, Proc IEE, Vol 117, No 1, Jan 1970, pp 213–218.
22.
BrownR FGoodwinG C: ‘Simplified method of eliminating the effect of non-linearities on process cross-correlations’, Electron Lett1966, 2, pp. 453–454.
23.
SimpsonH R: ‘Statistical Properties of a class of pseudo-random sequences’, Proc IEE, Vol 113, No 12, Dec 1966, pp. 2075–2081.
24.
SandbergAStarkL: ‘Wiener G Function Analysis as an approach to non-linear characteristics of human pupil light reflex’. Presbyterian-St Luke's Hospital Medical Bulletin, Vol 5, 1966, pp 89–105.
25.
EykhoffP: ‘Process parameter estimation’. Progress in Control Engineering, 1964, 2, pp 161–208. (Heywood Scientific book).
26.
LeeY WSchetzenM: ‘Measurement of the kernels of a non-linear system by cross-correlation’. Quarterly Progress Report No 60, Research Laboratory of Electronics MIT, Jan 1961, pp. 118–130.
27.
LeeY WSchetzenM: ‘Measurement of the Wiener kernels of a non-linear system by cross-correlation’, Int J Control1965, 2, pp. 237–254.
28.
BarkerH AObidegwuS NPradisthayonT: ‘Performance of anti-symmetric pseudo-random signals in the measurement of second-order Volterra kernels by cross-correlation’. Proc IEE, Vol 119, No 3, March 1972, pp. 353–362.
29.
BrownR F: ‘Modification of the prbn cross-correlation method to correct for the effect of non-linearity’ (Electron Lett, 1969, 5, pp 594–595.
30.
PowerH MSimpsonR J: ‘Application of almost periodic functions to system identification’, Electron Lett1970, 6, pp. 269–270.
31.
BesicovitchA S: ‘Almost periodic functions’. (Cambridge University Press, 1938).
32.
SimpsonR JPowerH M: ‘Generalisation of a method for system identification using almost periodic functions’, Electron Lett1970, 6, pp. 803–804.
33.
BrownR F: ‘Effects of drift and nonlinearity upon pseudo-random binary cross-correlation estimates’. 2nd IFAC Symposium on Identification Processes, Czechoslovakia, June 1970, Paper 9.3.
34.
BrownR F: ‘The effect of drift upon the estimation of impulse responses using pseudo-random binary noise’. PhD Thesis, University of New South Wales, 1970.
35.
BrownR F: ‘Review and comparison of drift correction schemes for periodic cross-correlation’, Electron Lett1969, 5, pp. 179–181.
36.
StarkL: ‘The pupillary control system: Its non-linear adaptive and stochastic engineering design characteristics’. Automatica, Vol 5, 1969, pp. 655–676.
37.
BarrettJ F: ‘The use of Volterra series to find region of stability of a non-linear differential equation’, Int J Control1965, 1, pp. 209–216.
38.
BansalV S: ‘Study of boundedness of the response of nuclear reactor using Volterra series’, Proc IEE1969, 116, pp. 2072–2074.
