In this form Qi is ∀ or ∃ where i=0,1,2...n and F is the quantifier-free formula. Q1x1 Q2x2...Qnxn is called the prefix, whereas F is called the matrix.
A formula with no quantifiers is called a trivial case of the prenex normal form.
Steps to convert into PNF
We'll follow the steps below to convert any expression into PNF:
- We eliminate all the occurrences of → and ↔ from the formula.
- We move all the negations inwards to appear only as a part of the literal.
- Standardize the variables apart if it is necessary.
- PNF is obtained by moving the quantifiers to the front of the formula.
Step 1
To remove the conditional → (if A then B) and bi-conditional ↔ (A if and only if B), we use the following logical equivalences:
- A→B≡¬A∨B
- A↔B≡(¬A∨B)∧(A∨¬B)
- A↔B≡(A∧B)∨(¬A∧¬B)
Step 2
We'll now try to move all the negations close to the literals instead of the negations occurring as a whole. We convert the ∀ symbol to the ∃ symbol and vice versa in this step when we shift the ¬ symbol. To accomplish step 2, we use the following logical equivalences:
- ¬¬A≡A
- ¬∀xA(x)≡∃x¬A(x)
- ¬∃xA(x)≡∀x¬A(x)
- De Morgan's law
Step 3
Renaming of the variables is called the standardizing of the variables apart. To achieve step 3, we use the following theorem to rename the variables to make them distinct.
Suppose we get A′ from A by making some replacements in A of the occurrences of Q(y)F(y) by Q(x)F(x). Q can either be an existential or universal quantifier.
Step 4
In this step, we'll shift all the ∀ and ∃ to the beginning of the expression. To achieve step 4, we use the following basic logical equivalences:
- A∨∀xF(x)≡∀x(A∨F(x)) where x does not occur in A.
- A∨∃xF(x)≡∃x(A∨F(x)) where x does not occur in A.
- A∧∀xF(x)≡∀x(A∧F(x)) where x does not occur in A.
- A∧∃xF(x)≡∃x(A∧F(x)) where x does not occur in A.
Example
Let's consider the following expression: