Equivalence of PDA, CFG

Conversion of CFG to PDA

1
 Overview
 When we talked about closure
properties of regular languages, it was
useful to be able to jump between RE
and DFA representations.
 Similarly, CFG’s and PDA’s are both
useful to deal with properties of the
CFL’s.

2
 Overview – (2)
 Also, PDA’s, being “algorithmic,” are
often easier to use when arguing that a
language is a CFL.
 Example: It is easy to see how a PDA
can recognize balanced parentheses;
not so easy as a grammar.
 But all depends on knowing that CFG’s
and PDA’s both define the CFL’s.
3
 Converting a CFG to a PDA
 Let L = L(G).
 Construct PDA P such that N(P) = L.
 P has:
 One state q.
 Input symbols = terminals of G.
 Stack symbols = all symbols of G.
 Start symbol = start symbol of G.

4
 Intuition About P
 Given input w, P will step through a
leftmost derivation of w from the start
symbol S.
 Since P can’t know what this derivation
is, or even what the end of w is, it uses
nondeterminism to “guess” the
production to use at each step.

5
 Intuition – (2)
 At each step, P represents some left-
sentential form (step of a leftmost
derivation).
 If the stack of P is , and P has so far
consumed x from its input, then P
represents left-sentential form x.
 At empty stack, the input consumed is a
string in L(G).
6
 Transition Function of P
1. δ(q, a, a) = (q, ε). (Type 1 rules)
 This step does not change the LSF
represented, but “moves” responsibility for a
from the stack to the consumed input.
2. If A ->  is a production of G, then
δ(q, ε, A) contains (q, ). (Type 2 rules)
 Guess a production for A, and represent the
next LSF in the derivation.
7
 Proof That L(P) = L(G)
 We need to show that (q, wx, S) ⊦* (q,
x, ) for any x if and only if S
=>*lm w.
 Part 1: “only if” is an induction on the
number of steps made by P.
 Basis: 0 steps.
 Then  = S, w = ε, and S =>*lm S is surely
true.
8
 Induction for Part 1
 Consider n moves of P: (q, wx, S) ⊦*
(q, x, ) and assume the IH for
sequences of n-1 moves.
 There are two cases, depending on
whether the last move uses a Type 1 or
Type 2 rule.

9
 Use of a Type 1 Rule
 The move sequence must be of the form
(q, yax, S) ⊦* (q, ax, a) ⊦ (q, x, ),
where ya = w.
 By the IH applied to the first n-1 steps,
S =>*lm ya.
 But ya = w, so S =>*lm w.

10
 Use of a Type 2 Rule
 The move sequence must be of the form
(q, wx, S) ⊦* (q, x, A) ⊦ (q, x, ),
where A ->  is a production and  = .
 By the IH applied to the first n-1 steps,
S =>*lm wA.
 Thus, S =>*lm w = w.

11
 Proof of Part 2 (“if”)
 We also must prove that if S =>*lm w,
then (q, wx, S) ⊦* (q, x, ) for any x.
 Induction on number of steps in the
leftmost derivation.
 Ideas are similar; read in text.

12
 Proof – Completion
 We now have (q, wx, S) ⊦* (q, x, ) for
any x if and only if S =>*lm w.
 In particular, let x =  = ε.
 Then (q, w, S) ⊦* (q, ε, ε) if and only if
S =>*lm w.
 That is, w is in N(P) if and only if w is in
L(G).
13
 From a PDA to a CFG
 Now, assume L = N(P).
 We’ll construct a CFG G such that L = L(G).
 Intuition: G will have variables generating
exactly the inputs that cause P to have the
net effect of popping a stack symbol X
while going from state p to state q.
 P never gets below this X while doing so.

14
 Variables of G
 G’s variables are of the form [pXq].
 This variable generates all and only the
strings w such that
(p, w, X) ⊦*(q, ε, ε).
 Also a start symbol S we’ll talk about
later.

15
 Productions of G
 Each production for [pXq] comes from a
move of P in state p with stack symbol X.
 Simplest case: δ(p, a, X) contains (q, ε).
 Then the production is [pXq] -> a.
 Note a can be an input symbol or ε.
 Here, [pXq] generates a, because reading
a is one way to pop X and go from p to q.
16
 Productions of G – (2)
 Next simplest case: δ(p, a, X) contains
(r, Y) for some state r and symbol Y.
 G has production [pXq] -> a[rYq].
 We can erase X and go from p to q by
reading a (entering state r and replacing
the X by Y) and then reading some w that
gets P from r to q while erasing the Y.
 Note: [pXq] =>* aw whenever
[rYq] =>* w.
17
 Productions of G – (3)
 Third simplest case: δ(p, a, X) contains
(r, YZ) for some state r and symbols Y
and Z.
 Now, P has replaced X by YZ.
 To have the net effect of erasing X, P
must erase Y, going from state r to
some state s, and then erase Z, going
from s to q.
18
 Picture of Action of P

a w x w x x

p r s q

X Y Z
Z

19
Third-Simplest Case – Concluded
 Since we do not know state s, we must
generate a family of productions:
[pXq] -> a[rYs][sZq]
for all states s.
 [pXq] =>* awx whenever [rYs] =>* w
and [sZq] =>* x.

20
Productions of G: General Case
 Suppose δ(p, a, X) contains (r, Y1,…Yk)
for some state r and k > 3.
 Generate family of productions
[pXq] ->
a[rY1s1][s1Y2s2]…[sk-2Yk-1sk-1][sk-1Ykq]

21
Completion of the Construction
 We can prove that (q0, w, Z0)⊦*(p, ε, ε)
if and only if [q0Z0p] =>* w.
 Proof is in text; it is two easy inductions.
 But state p can be anything.
 Thus, add to G another variable S, the
start symbol, and add productions
S -> [q0Z0p] for each state p.

22