.
Article
SCIENCE CHINA
Physics, Mechanics & Astronomy
May 2013 Vol. 56 No. 5: 941–946
doi: 10.1007/s11433-013-5048-y
c
Science China Press and Springer-Verlag Berlin Heidelberg 2013
phys.scichina.com www.springerlink.com
On a class of quantum Turing machine halting deterministically
LIANG Min & YANG Li
*
State Key Laboratory of Information Security, Institute of Information Engineering, Chinese Academy of Sciences, Beijing 100093, China
Received May 10, 2012; accepted October 23, 2012; published online April 15, 2013
We consider a subclass of quantum Turing machines (QTM), named stationary rotational quantum Turing machine (SR-QTM),
which halts deterministically and has deterministic tape head position. A quantum state transition diagram (QSTD) is proposed to
describe SR-QTM. With QSTD, we construct a SR-QTM which is universal for all near-trivial transformations. This indicates there
exists a QTM which is universal for the above subclass. Finally we show that SR-QTM is computational equivalent with ordinary
QTM in the bounded error setting. It can be seen that SR-QTMs have deterministic tape head position and halt deterministically,
and thus the halting scheme problem will not exist for this class of QTMs.
quantum Turing machine, quantum circuit, halting scheme, quantum computational complexity
PACS number(s): 03.67.Lx, 03.65.-w, 02.70.-c
Citation: Liang M, Yang L. On a class of quantum Turing machine halting deterministically. Sci China-Phys Mech Astron, 2013, 56: 941–946, doi:
10.1007/s11433-013-5048-y
Deutsch [1] formulated the model of quantum computer
which is referred to as quantum Turing machine (QTM).
Then it was pointed out that there exists a halting scheme
problem when a QTM takes different time steps for different
branches of computation [2]. In this case, a measurement of
the halt qubit can spoil the computation.
Ozawa [3] showed that the halting scheme is equivalent to
the quantum nondemolition (QND) monitoring of the output
observable. Linden and Popescu [4] argued that the halting
scheme proposed by Ozawa [3] is not consistent with the uni-
tarity of the evolution operator and the halting scheme prob-
lem is still unsolved. Ozawa [5] proposed QTM with well-
behaved halt flags, and arbitrary QTM can be simulated by
this kind of QTM. Moreover, this kind of QTMs obeys a halt-
ing protocol, which requires one measurement of the halt flag
after one time step. He showed that the halting protocol does
not affect the computational result of QTM. This indicates
the probability distribution of the output is not affected by
monitoring the halt flag. Thus the halt scheme problem is
*Corresponding author (email: yangli@iie.ac.cn)
solved though the QTM may halt probabilistically. There are
also some discussions about halting scheme problem [6–8].
This halting scheme problem was avoided by Bernstein
and Vazirani [9] who considered a simpler QTM that has the
same time step for all different branches. This kind of QTMs
can halt deterministically and the universal QTM was con-
structed in this sense. In quantum computing, this kind of
QTMs is usually the central model of the research in recent
times. Does there exist more simpler QTM? In this paper, we
propose a simpler class of QTMs (called SR-QTM), which
halt deterministically and have deterministic tape head posi-
tion.
Miyadera and Ohya [10] showed it is not possible to ef-
ficiently distinguish between those QTMs which halt deter-
ministically and those which halts probabilistically. They
suggested it is natural for QTM to halt probabilistically and it
is meaningless to define non-probabilistically halting QTM.
However, their proof necessarily indicates there does not ex-
ist a universal algorithm to decide whether arbitrary QTM of
the whole QTMs has the same time step of different branches,
thus it does not exclude the possibility that there exists a uni-