### 2009

Koch, Tobias; Lapidoth, Amos; Sotiriadis, Paul P

Channels That Heat Up Journal Article

In: IEEE Transactions on Information Theory, 55 (8), pp. 3594–3612, 2009, ISSN: 0018-9448.

Abstract | Links | BibTeX | Tags: additive noise channel, Capacity per unit cost, channel capacity, channels with memory, cooling, electronic circuits, heat dissipation, heat sinks, high signal-to-noise ratio, high signal-to-noise ratio (SNR), intrinsic thermal noise, low transmit power, network analysis, noise variance, on-chip communication, thermal noise

@article{Koch2009,

title = {Channels That Heat Up},

author = {Tobias Koch and Amos Lapidoth and Paul P Sotiriadis},

url = {http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=5165190},

issn = {0018-9448},

year = {2009},

date = {2009-01-01},

journal = {IEEE Transactions on Information Theory},

volume = {55},

number = {8},

pages = {3594--3612},

abstract = {This paper considers an additive noise channel where the time-A; noise variance is a weighted sum of the squared magnitudes of the previous channel inputs plus a constant. This channel model accounts for the dependence of the intrinsic thermal noise on the data due to the heat dissipation associated with the transmission of data in electronic circuits: the data determine the transmitted signal, which in turn heats up the circuit and thus influences the power of the thermal noise. The capacity of this channel (both with and without feedback) is studied at low transmit powers and at high transmit powers. At low transmit powers, the slope of the capacity-versus-power curve at zero is computed and it is shown that the heating-up effect is beneficial. At high transmit powers, conditions are determined under which the capacity is bounded, i.e., under which the capacity does not grow to infinity as the allowed average power tends to infinity.},

keywords = {additive noise channel, Capacity per unit cost, channel capacity, channels with memory, cooling, electronic circuits, heat dissipation, heat sinks, high signal-to-noise ratio, high signal-to-noise ratio (SNR), intrinsic thermal noise, low transmit power, network analysis, noise variance, on-chip communication, thermal noise},

pubstate = {published},

tppubtype = {article}

}

This paper considers an additive noise channel where the time-A; noise variance is a weighted sum of the squared magnitudes of the previous channel inputs plus a constant. This channel model accounts for the dependence of the intrinsic thermal noise on the data due to the heat dissipation associated with the transmission of data in electronic circuits: the data determine the transmitted signal, which in turn heats up the circuit and thus influences the power of the thermal noise. The capacity of this channel (both with and without feedback) is studied at low transmit powers and at high transmit powers. At low transmit powers, the slope of the capacity-versus-power curve at zero is computed and it is shown that the heating-up effect is beneficial. At high transmit powers, conditions are determined under which the capacity is bounded, i.e., under which the capacity does not grow to infinity as the allowed average power tends to infinity.