Homework HW0 – Problem 2 CS62003 – Siva Dharman Naidu

Theorem: Prove that Post Correspondence Problem (PCP) is undecidable

Post Correspondence Problem

This problem is introduced by Emil Leon Post in 1946 [1]. Problem asks to find a match from the given
collection of two faced dominos where each face of dominos contains strings. A two faced domino will
𝑎
look like: [𝑎𝑏]

So the collection of two faced dominos can be represented in a set as:

𝑎 𝑏 𝑐𝑎 𝑎 𝑎𝑏𝑐
{[ ],[ ],[ ],[ ], [ ]}
𝑎𝑏 𝑐𝑎 𝑎 𝑎𝑏 𝑐

If we choose a list of dominos from the dominos (repetition is allowed while choosing) set in such a way
that concatenated of strings in one face must be same as concatenated string of other face. This list is
called match. This problem asks us to find a match from any given set of two faced dominos.

In the above collection of two faced dominos, there exists a match because the string “abcaaabc” can be
obtained by concatenating strings of each face of the dominos separately.

Problem Definition

An instance of this problem can be expressed in mathematically as:

Let D be the collection of n dominos.

u1 u2 u3 un
D = {[ ] , [ ] , [ ] , … , [ ]}
𝑣1 𝑣2 𝑣3 𝑣𝑛

And there exists match sequence k1, k1, k2, k3… kl where

u1u2u3 … ul = v1v2v3 ... vl (Concatenation of strings of one face is u and other face v) problem is to find
the existence of match in D. Let

PCP = {<D>| D is the instance of the Post Correspondence Problem with a match}

Proof:

Consider a Turing machine M to simulate this PCP’s input string w can be represented as <M, w>.

M = (Q,Σ,Γ,δ,q0,qaccept,qreject)

1|Page
Homework HW0 – Problem 2 CS62003 – Siva Dharman Naidu

If there is a match in the input string w, then Turing machine M halts in an accepting state. This halting
state of Turing machine is an acceptance problem ATM. We know that the acceptance problem ATM is
undecidable from proof mentioned in Sipser’s book [1]. Therefore PCP problem is also undecidable.

If a Turing Machine M accepts input w, then the computation history of M gives us the accepting
configuration, from which we can determine the instance D of the PCP problem has a match. How can
construct the instance D so that there exists a match and Turing Machine M halts in accept state on input
w? By choosing the dominos in such a way a concatenation of strings of each faces are same at each step
of the selection. We can force the simulation of M to accept the w.

To force the simulation of M, we make 2 modifications to Turing Machine M and a one change to our
PCP problem.

1. M on input w can never attempt to move tape head beyond the left end of the input tape.
2. If input is empty string € we use ˽
𝑢1
3. PCP problem starts the match with first domino [ ] This is called Modified PCP problem
𝑣1

MPCP = {<D>| D is an instance of PCP starts with first domino}

Construction Steps:

#
1. Put [#𝑞0𝑤1𝑤2..𝑤𝑛#] into D| as the first domino, where is instance of D| is MPCP. A partial match

is obtained in first domino is # at one face is same #symbol in other face.

2. Transition functions for Turing Machine M can have moves Left L, Right R
For every x,y z in Γ tape alphabets and q,r in Q where q is not equal to qreject.
𝑞𝑥 𝑧𝑞𝑥
If δ(q,x) = (r,y,R) put the domino [𝑏𝑦] into D| and δ(q,x) = (r,y,L) put the domino [𝑟𝑧𝑦] into D|
𝑥
3. For every tape alphabets x in Γ put [𝑥] into D|
# #
To mark the separation of each configurations put [#] and [˽#] into D|
𝑥𝑞𝑎 𝑞𝑎𝑥
4. To read input alphabets x in Γ even after Turing Machine is accepting state put [ 𝑞𝑎 ] and [ 𝑞𝑎 ]
𝑞𝑎#
and [ #
] into D|

These steps concludes the construction of D| Since this a instance of MPCP, we need to convert this to
PCP. So to convert to D, we consider the below dominos and strings matchings.

2|Page
Homework HW0 – Problem 2 CS62003 – Siva Dharman Naidu

Converting MPCP To PCP

Let u = u1,u2,…,un be any string of input length n and modify these strings as

$u = *u1*u2*u3* …*un

u$ = u1* u2* u3* … un*

$u$ = * u1* u2* u3* … un*

Let D be the set of two faced Dominos

u1 u2 u3 un $u1 $u2 ∗˽
D| = {[𝑣1] , [𝑣2] , [𝑣3] , … , [𝑣𝑛]} and {[$𝑣1$] , [𝑣2$] , , [ ˽ ]}

$u1
From above dominos collection, we could see only domino has partial match starts with [$𝑣1$] and to
∗˽
place marker the end of inputs[ ] There by we can avoid stating explicit requirement of domino should
˽

start with first domino.

If number of configurations of the Turing machine does not lie within the value of qngn [Lemma specified
in Sipsers Book], then Turing machine is looping state. It does not halt.

References

[1] Michael Sipser (2005). "A Simple Undecidable Problem". Introduction to the Theory of Computation
(2nd ed.). Thomson Course Technology. pp. 199–205. ISBN 0-534-95097-3.

[2] E. L. Post (1946). "A variant of a recursively unsolvable problem" (PDF). Bull. Amer. Math. Soc 52.

3|Page