International Journal of Engineering Research and General Science Volume 2, Issue 4, June-July, 2014

ISSN 2091-2730

Comparative Study of type-1 and Type-2 Fuzzy Systems

Assistant proessor, GIMET Amritsar

E-mail- gmetneetugupta@[Link]

Abstract

TYPE-2 fuzzy sets (T2 FSs), originally introduced by Zadeh [3], provide additional design degrees of freedom in
Mamdani and TSK fuzzy logic systems (FLSs), which can be very useful when such systems are used in situations where lots of
uncertainties are present [4]. The implementation of this type-2 FLS involves the operations of fuzzification, inference,and output
processing. We focus on output processing, which consists of type reduction defuzzification. Type-reduction methods are extended
versions of type-1 defuzzification methods. In this paper we represent a comparision of dirrerent techniques of fuzzy logic systems.

Key Wordsfuzzy logic systems, interval sets, uncertainties,membership function, defuzzification.

I. INTRODUCTION
IN this paper, we introduce a new class of fuzzy logic systemstype-2 fuzzy logic systemsin which the antecedent or consequent
membership functions are type-2 fuzzy sets. The concept of a type-2 fuzzy set was introduced by Zadeh [1] as an extension of the
concept of an ordinary fuzzy set (henceforth called a type-1 fuzzy set). Such sets are fuzzy sets whose membership grades themselves
are type-1 fuzzy sets;they are very useful in circumstances where it is difficult to determine an exact membership function for a fuzzy
set; hence,they are useful for incorporating [Link] often, the knowledge used to construct rules in a fuzzy logic system
(FLS) is uncertain. This uncertainty leads to rules having uncertain antecedents and/or consequents, which in turn translates into
uncertain antecedent and/or consequent membership functions. For example:
1) A fuzzy logic modulation classifier described in [2] centers type-1 Gaussian membership functions at constellation points on the inphase/quadrature plane. In practice, the constellation points drift. This is analogous to the situation of a Gaussian membership function
(MF) with an uncertain mean. A type-2 formulation can capture this drift.
2) Previous applications of FL to forecasting do not account for noise in training data. In forecasting, since antecedents and
consequents are the same variable, the uncertainty during training exists on both the antecedents and consequents. If we have
information about the level of uncertainty, it can be used when we model antecedents and consequents as type-2 sets.
3) When rules are collected by surveying experts, if we first query the experts about the locations and spreads of the fuzzy sets
associated with antecedent and consequent terms, it is very likely that we will get different answers from each expert[4] This leads to
statistical uncertainties about locations and spreads of antecedent and consequent fuzzy sets. Such uncertainties can be incorporated
into the descriptions of these sets using type-2 membership functions. In addition, experts often give different answers to the same
rule-question; this results in rules that have the same antecedents but different [Link] such a case, it is also possible to
represent the output of the FLS built from these rules as a fuzzy set rather than a crisp number. This can also be achieved within the
type-2 framework[6].

2 TYPE 1 FLS
In a type-1 FLS, the inference engine combines rules and gives a mapping from input type-1 fuzzy sets to output type-1 fuzzy
sets. Multiple antecedents in rules are connected by the -norm (corresponding to intersection of sets). The membership
grades in the input sets are combined with those in the output sets using the sup-star composition. Multiple rules may be combined
using the -conorm operation (corresponding to union of sets) or during defuzzification by weighted summa- tion. In the type-2 case,
the inference process is very similar. The inference engine combines rules and gives a mapping from input type-2 fuzzy sets to output
type-2 fuzzy sets. To do this one needs to find unions and intersections of type-2 sets, as well as compositions of type-2 relations.
In a type-1 FLS, the defuzzifier produces a crisp output from the fuzzy set that is the output of the inference engine, i.e., a
type-0 (crisp) output is obtained from a type-1 set. In the type-2 case, the output of the inference engine is a type-2 set; so we use
extended versions (using Zadehs extension principle [5], [7], of type-1 defuzzification methods. This extended defuzzification
gives a type-1 fuzzy set. Since this operation takes us from the type-2 output sets of the FLS to a type-1 set, we call this operation
type reduction and the type-reduced set so obtained a type-reduced set.

195

[Link]

International Journal of Engineering Research and General Science Volume 2, Issue 4, June-July, 2014
ISSN 2091-2730

To obtain a crisp output from a type-2 FLS, we can defuzzify the type-reduced set. The most natural way of doing this seems
to be by finding the centroid of the type-reduced set, however, there exist other possibilities like choosing the highest
membership point in the type-reduced set.

From our discussions so far, we see that in order to develop a type-2 FLS, one needs to be able to: 1) perform the set theoretic
operations of union, intersection, and complement on type-2 sets [8]; 2) know properties (e.g., commutativity, associativity,
identity laws) of membership grades of type-2 sets [8]; 3) deal with type-2 fuzzy relations and their compo- sitions [8]; and 4)
perform type reduction and defuzzification to obtain a set-valued or crisp output from the FLS [8], [7].

Fig. 1. Type-1 FLS.

3 TYPE 2 FLS
Type-2 fuzzy sets allow us to handle linguistic uncertainties,as typified by the adage words can mean different things to different
people [20]. A fuzzy relation of higher type (e.g.,type-2) has been regarded as one way to increase the fuzziness of a relation and,
according to Hisdal [11], increased fuzziness in a description means increased ability to handle inexact information in a logically
correct manner. According to John [12], Type-2 fuzzy sets allow for linguistic grades of membership, s ssisting in knowledge
representation and they also offer improvement on inferencing with type-1 sets.

Fig. 2. The structure of a type-2 FLS.

196

[Link]

International Journal of Engineering Research and General Science Volume 2, Issue 4, June-July, 2014
ISSN 2091-2730

Fig. 2 shows the structure of a type-2 FLS. It is very similar to the structure of a type-1 FLS [26]. For a type-1 FLS, the
output processing block only contains the defuzzifier. We assume that the reader is familiar with type-1 FLSs, so that here we focus
only on the similarities and differences between the two FLSs.
The fuzzifier maps the crisp input into a fuzzy set. This fuzzy set can, in general, be a type-2 set, however, in this paper,
we consider only singleton fuzzification, for which the input fuzzy set has only a single point of nonzero membership
Fig.3 shows an example of product and minimum inference for an arbitrary single-input single-output type-2 FLS using
Gaussian type-2 sets. Uncertainty in the primary membership grades of a type-2 MF consists of a dark region that we call the
footprint of uncertainty of a type-2 MF. The footprint of uncertainty represents the union of all primary memberships. Darker
areas indicate higher secondary memberships. The principal membership function, i.e., the set of primary memberships having
secondary membership equal to one, is indicated with a thick line

(a)

(c)

(b)

(d)

Fig. 3. Illustrations of product and minimum inference in the type-2 case. (a) Gaussian type-2 antecedent set for a single
nput system. The membership of a certain input x = 4 in the principal membership function is also shown, equal to . (b)
Consequent set corresponding to the antecedent set shown in (a). (c) Scaled consequent set for x = 4 using product
inference. Observe that the secondary membership functions of the consequent set also change depending upon the
standard deviation of the membership grade of x. (d) Clipped consequent set for x = 4 using minimum inference.

REERENCES:
[1] L. A. Zadeh, The concept of a linguistic variable and its application to approximate reasoning1, Inform. Sci., vol. 8, pp. 199
249. 1975.
[2] W. Wei and J. M. Mendel, A fuzzy logic method for modulation classification in nonideal environments, IEEE Trans. Fuzzy
Syst., vol.7, pp. 333344, June 1999.
[3] L. A. Zadeh, The concept of a linguistic variable and its applicationto approximate reasoning-1, Inform. Sci., vol. 8, pp. 199
249, 1975
[4] , Type-2 fuzzy sets: Some questions and answers, IEEE Connections,vol. 1, pp. 1013, Aug. 2003.
[5] O. Castillo, P. Melin, Intelligent systems with interval type-2 fuzzy logic, International Journal of Innovative Computing,
Information and Control 4 (4) (2008) 771783.
[6] S. Coupland, R. John, A fast geometric method for defuzzication of type-2 fuzzy sets, IEEE Transactions on Fuzzy Systems
16 (4) (2008) 929941. [3] S. Coupland, R. John, New geometric inference techniques for type-2 fuzzy sets, International Journal
of Approximate Reasoning 49 (1) (2008) 198211. [7] D. Dubois, H. Prade, Fuzzy sets and systems: theory and applications,
Academic Press, Inc., New York, 1980.
[8] H. Hagras, Type-2 fuzzy logic controllers: a way forward for fuzzy systems in real world environments, Lecture Notes in
Computer Science (5050) (2008) 181200.
[9] R. O. Duda and P. E. Hart, Pattern Classification and Scene [Link] York: Wiley, 1973.
197
[Link]

International Journal of Engineering Research and General Science Volume 2, Issue 4, June-July, 2014
ISSN 2091-2730

[10] S. Ghosh, Q. Razouqi, H. J. Schumacher, and A. Celmins, A survey of recent advances in fuzzy logic in telecommunications
networks and new challenges, IEEE Trans. Fuzzy Syst., vol. 6, pp. 443447, Aug. 1998.
[11] E. Hisdal, The IF THEN ELSE statement and interval-valued fuzzy sets of higher type, Int. J. Man-Machine Studies, vol. 15,
pp. 385455,1981.
[12] R. I. John, Type 2 fuzzy sets: An appraisal of theory and applications,Int. J. Uncertainty, Fuzziness, Knowledge-Based Syst.,
vol. 6, no. 6, pp.563576, Dec. 1998.
[13] R. I. John, P. R. Innocent, and M. R. Barnes, Type 2 fuzzy sets and neuro-fuzzy clustering of radiographic tibia images, IEEE
Int. [Link] Syst., Anchorage, AK, May 1998, pp. 13731376.
[14] N. N. Karnik and J. M. Mendel, Introduction to Type-2 Fuzzy Logic Systems, presented at IEEE FUZZ Conf., Anchorage, AK,
May 1998.
[15] , Type-2 Fuzzy Logic Systems: Type-Reduction, presented at IEEE Syst., Man, Cybern. Conf., San Diego, CA, Oct. 1998.
[16] , An introduction to type-2 fuzzy logic systems, Univ. Southern California, Rep., Oct. 1998.
[17] , Applications of type-2 fuzzy logic systems: Handling the uncertainty associated with surveys, presented at FUZZ-IEEE Conf.,
Seoul, Korea, Aug. 1999.
[18] , Applications of type-2 fuzzy logic systems to forecasting of time-series, Inform. Sci., to be published.
[19] , Operations on type-2 fuzzy sets, Fuzzy Sets Syst., to be published.
[20] , Centroid of a type-2 fuzzy set, Inform. Sci., to be published.
[21] G. J. Klir and B. Yuan, Fuzzy Sets and Fuzzy Logic: Theory and Applications. Englewood Cliffs, NJ: Prentice-Hall, 1995

198

[Link]