CIRCUIT FOR DIVIDING DIVIDEND BY 2n THROUGH THE USE OF ROUND-UP OR ROUND-DOWN APPROXIMATION

    公开(公告)号:JP2000112761A

    公开(公告)日:2000-04-21

    申请号:JP32747799

    申请日:1999-11-17

    Abstract: PROBLEM TO BE SOLVED: To provide a circuit for calculating the attribute degree of the antecedent part of a fuzzy rule determining the intersection of a membership function and an input by a simple and economical system. SOLUTION: A circuit for dividing a dividend by 2n through the use of round-up or round-down approximation includes a register 32 which includes a binary number to be devided and an adder 48 which is connected to the register 32, receives the (n+1) bit of the binary number to be divided to a highest- order bit by a first input A and the n-numbered bit of the binary number to be divided by a second input B, adds the first input A to a second input B and outputting a total value A+B. Then, the binary number divided by the total value A+B outputted by the adder 48 is the result of round-up and round- down approximation of the binary number divided by 2n.

    2.
    发明专利
    未知

    公开(公告)号:DE69432349D1

    公开(公告)日:2003-04-30

    申请号:DE69432349

    申请日:1994-05-23

    Abstract: Method of parallel processing of multiple inference rules (R) organised in fuzzy sets or logical functions of multiple fuzzy sets comprising membership functions (I') defined in a so-called universe of discourse (U) and said inference rules (R) being configured essentially as IF-THEN rules with at least one antecedent preposition and at least one consequent implication and each preposition comprising at least one term (T) of comparison between membership functions (I') and a plurality of input data (I) and each term (T) being separated by logical operators (OL). The method associates with the logical operators (OL) maximum and minimum operations among two or more elements and calculates exhaustively the overall degree of truth ( OMEGA ) of a rule (R) with a maximum or minimum of N partial truth levels (w).

    3.
    发明专利
    未知

    公开(公告)号:DE69426229D1

    公开(公告)日:2000-12-07

    申请号:DE69426229

    申请日:1994-05-23

    Abstract: Method of parallel processing of multiple inference rules (R) organized in fuzzy sets or logical functions of multiple fuzzy sets comprising membership functions (I') defined in a so-called universe of discourse (U) and said inference rules (R) being configured essentially as IF-THEN rules with at least one antecedent preposition and at least one consequent implication and each preposition comprising at least one term (T) of comparison between membership functions (I') and a plurality of input data (I) and each term (T) being separated by logical operators (OL). The method comprises at least one phase of calculation of the weight ( OMEGA ) of each term (T) of the antecedent part of each fuzzy logic inference rule as the greatest value of the intersection between the set of input data (I) and the corresponding membership functions (I').

    4.
    发明专利
    未知

    公开(公告)号:DE69426229T2

    公开(公告)日:2001-05-23

    申请号:DE69426229

    申请日:1994-05-23

    Abstract: Method of parallel processing of multiple inference rules (R) organized in fuzzy sets or logical functions of multiple fuzzy sets comprising membership functions (I') defined in a so-called universe of discourse (U) and said inference rules (R) being configured essentially as IF-THEN rules with at least one antecedent preposition and at least one consequent implication and each preposition comprising at least one term (T) of comparison between membership functions (I') and a plurality of input data (I) and each term (T) being separated by logical operators (OL). The method comprises at least one phase of calculation of the weight ( OMEGA ) of each term (T) of the antecedent part of each fuzzy logic inference rule as the greatest value of the intersection between the set of input data (I) and the corresponding membership functions (I').

    5.
    发明专利
    未知

    公开(公告)号:DE69534914D1

    公开(公告)日:2006-05-18

    申请号:DE69534914

    申请日:1995-01-31

    Abstract: The present level shift circuit has a first (I1) and a second (I2) input respectively for input of a first and a second voltage signal and an output (OT) and comprises: a) a first transistor (Q1) having a control terminal (G1), a first (S1) and a second (D1) main conduction terminal identifying a main conduction path, and b) a second transistor (Q2) of the same type as said first transistor (Q1) and having a control terminal (G2), a first (S2) and a second (D2) main conduction terminal identifying a main conduction path. The first signal is applied essentially between said control terminal (G1) and said first terminal (S1) of said first transistor (Q1) and said second input (I2) is coupled with the control terminal (G2) of said second transistor (Q2). The currents flowing in the conduction paths of the first (Q1) and the second (Q2) transistors are mutually proportional and one made from the other. The output (OT) is coupled with the first terminal (S2) of the second transistor (Q2). The control terminal (G1) of said first transistor (Q1) is connected to a potential reference (GND). The first signal is applied essentially to said first terminal (S1) of said first transistor (Q1).

    6.
    发明专利
    未知

    公开(公告)号:DE69521027T2

    公开(公告)日:2001-09-06

    申请号:DE69521027

    申请日:1995-03-28

    Abstract: A method for memorizing membership functions (MF), comprising the following steps: memorizing the position of the vertex (CV) of the triangle that defines the function in the universe of discourse (UdD); memorizing a first distance (LVD) between the position of the vertex in the universe of discourse and the point of intersection between the left side of the triangle that defines the membership function and the axis of the universe of discourse; memorizing a second distance (RVD) between the position of the vertex in the universe of discourse and the point of intersection between the right side of the triangle that defines the membership function and the axis of the universe of discourse. The present invention furthermore relates to a circuit for calculating the grade of membership of the antecedent of a fuzzy rule, the particularity whereof resides in that it comprises means adapted to fuzzify an input variable by adopting the geometric proportions that occur between homologous sides of similar triangles defined by the position of the input value in the universe of discourse.

    7.
    发明专利
    未知

    公开(公告)号:DE69521027D1

    公开(公告)日:2001-06-28

    申请号:DE69521027

    申请日:1995-03-28

    Abstract: A method for memorizing membership functions (MF), comprising the following steps: memorizing the position of the vertex (CV) of the triangle that defines the function in the universe of discourse (UdD); memorizing a first distance (LVD) between the position of the vertex in the universe of discourse and the point of intersection between the left side of the triangle that defines the membership function and the axis of the universe of discourse; memorizing a second distance (RVD) between the position of the vertex in the universe of discourse and the point of intersection between the right side of the triangle that defines the membership function and the axis of the universe of discourse. The present invention furthermore relates to a circuit for calculating the grade of membership of the antecedent of a fuzzy rule, the particularity whereof resides in that it comprises means adapted to fuzzify an input variable by adopting the geometric proportions that occur between homologous sides of similar triangles defined by the position of the input value in the universe of discourse.

Patent Agency Ranking