Relating the quantum mechanics of discrete systems to standard canonical quantum mechanics (Q474841)

The article is motivated by the philosophical question whether one can derive the fundamental principles of physics in terms of models based on integers only. The author has proposed a formulation of quantum mechanics using operators having purely discrete, ineger-valued, spectra. He then explains why the `real' world appears to be based on real-valued position-momenta. The basic idea is rather simple but elegant. It comes from the fact that to every real number \(x\), we can uniquely associate a pair \(([x], x - [x])\) consisting of the integral part and the fractional part of \(x\). Thus, real-valued functions can be in some sense `decomposed' into functions on integers and on the unit interval. But identifying the unit interval with the circle (after some care about the boundary values), one can use Fourier transform to get functions on integers. This is how we can associate a pair of integers to a real number in an (almost) bijective fashion. The formulation begins with associating to the standard quantum mechanical position-momentum pair \((q, p)\) a similar operator-pair \((Q, P )\) with purely integer spectra, satisfying, for example, \[ q = Q + \eta P , \] where \(Q = [q]\) and \(P\) is an operator with integer-spectrum obtained from the fractional part of q via the Fourier transform. This naive prescription needs some further adjustments and modifications to take care of issues related to boundary values (in terms of the `edge-states' in the language of the paper) and to achive certain symmetry between position and momentum. The author discusses the example of harmonic oscillator in this framework.