The so-called "replica method" of statistical physics is employed for the large system analysis of vector precoding for the Gaussian multiple-input multiple-output (MIMO) broadcast channel. The transmitter is assumed to comprise a linear front-end combined with nonlinear precoding, that minimizes the front-end imposed transmit energy penalty. Focusing on discrete complex input alphabets, the energy penalty is minimized by relaxing the input alphabet to a larger alphabet set prior to precoding. For the common discrete lattice-based relaxation, the problem is found to violate the assumption of replica symmetry and a replica symmetry breaking ansatz is taken. The limiting empirical distribution of the precoder's output, as well as the limiting energy penalty, are derived for one-step replica symmetry breaking. For convex relaxations, replica symmetry is found to hold and corresponding results are obtained for comparison. Particularizing to a "zeroforcing" (ZF) linear front-end, and non-cooperative users, a decoupling result is derived according to which the channel observed by each of the individual receivers can be effectively characterized by the Markov chain u-x-y, where u, x, and y are the channel input, the equivalent precoder output, and the channel output, respectively. For discrete lattice-based alphabet relaxation, the impact of replica symmetry breaking is demonstrated for the energy penalty at the transmitter.An analysis of spectral efficiency is provided to compare discrete lattice-based relaxations against convex relaxations, as well as linear ZF and Tomlinson-Harashima precoding (THP). Focusing on quaternary phase shift-keying (QPSK), significant performance gains of both lattice and convex relaxations are revealed compared to linear ZF precoding, for medium to high signal-to-noise ratios (SNRs). THP is shown to be outperformed as well. In addition, comparing certain lattice-based relaxations for QPSK against a convex counterpart, the latter is found to be superior for low and high SNRs but slightly inferior for medium SNRs in terms of spectral efficiency.

When a finite system is at equilibrium with a heat bath, the equilibrium temperature is dictated by the heat bath and not by the intrinsic thermostatistics of the finite system. If not sufficiently large, it may be necessary for the finite system to change its thermostatistics in order to be at equilibrium with the heat bath. We account for this process by invoking Landsberg's notion of temperature-dependent energy levels. We establish that the mismatch between the intrinsic temperature of the excited finite system and that of the heat bath drives a spectrum perturbation which enables thermal equilibrium. We show that the temperature-induced spectrum perturbation is equivalent to Hill's purely thermodynamic subdivision potential. The difference between intrinsic and equilibrium temperature provides us with a measure for how large a system can be before it no longer needs to be regarded as small. The theoretical framework proposed in this paper identifies the role of temperature in a bottom-up thermostatistical description of finite systems.

We show that, by properly adopting the notion of temperature-dependent energy levels, the standard tools of differential thermodynamics can be used for a consistent thermostatistical description irrespective of system size. In this framework the paradigmatic (yet not always descriptive) large-system limit is no longer a necessary assumption for differential thermodynamics. We present a generalized relation between temperature and internal energy which extends thermodynamics all the way to isolated quantum systems.

No abstract

Understanding how small systems exchange energy with a heat bath is important to describe how their unique properties can be affected by the environment. In this contribution, we apply Landsberg's theory of temperature-dependent energy levels to describe the progressive thermalization of small systems as their spectrum is perturbed by a heat bath. We propose a mechanism whereby the small system undergoes a discrete series of excitations and isentropic spectrum adjustments leading to a final state of thermal equilibrium. This produces standard thermodynamic results without invoking system size. The thermal relaxation of a single harmonic oscillator is analyzed as a model example of a system with a quantized spectrum than can be embedded in a thermal environment. A description of how the thermal environment affects the spectrum of a small system can be the first step in using environmental factors, such as temperature, as parameters in the design and operation of nanosystem properties.2

Negative thermophoresis (a particle moving up the temperature gradient) is a somewhat counterintuitive phenomenon which has thus far eluded a simple thermostatistical description. The purpose of this letter is to show that a thermodynamic framework based on the formulation of a Hamiltonian of mean force has the descriptive ability to capture this interesting and elusive phenomenon in an unusually elegant and straightforward fashion. We propose a mechanism that describes the advent of a thermophoretic force acting from cold to hot on systems that are strongly coupled to a non-isothermal heat bath. When a system is strongly coupled to the heat bath, the system's eigenenergies Ej become effectively temperature-dependent. This adjustment of the energy levels allows the system to take heat from the environment, +d Ej , and return it as work, −d T dEj/dT. This effect can make the temperature-dependence of the effective energy profile non-monotonic. As a result, particles may experience a force in either direction depending on the temperature.

No abstract

We propose a Hamiltonian-based approach to the nonextensive thermodynamics of small systems, where small is a relative term comparing the size of the system to the size of the effective interaction region around it. We show that the effective Hamiltonian approach gives easy accessibility to the thermodynamic properties of systems strongly coupled to their surroundings. The theory does not rely on the classical concept of dividing surface to characterize the system’s interaction with the environment. Instead, it defines an effective interaction region over which a system exchanges extensive quantities with its surroundings, easily producing laws recently shown to be valid at the nanoscale.