Asymptotic analysis of asymmetric MIMO links: EVM limits for joint decoding of PSK and QAM

Mikko Vehkaperä, Taneli Riihonen, Maksym Girnyk, Emil Bjornson, Merouane Debbah, Lars K. Rasmussen, Risto Wichman

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review


Hardware non-idealities in wireless transmitter electronics cause distortion that is not captured by conventional linear channel models; in fact, error-vector magnitude (EVM) measurements in conformance testing conceptually reduce their collective effect to an additive noise component at each subcarrier. Motivated by the EVM, the present paper considers a 'binoisy' multiple-input multiple-output (MIMO) channel model where the additional non-idealities manifest themselves as an additive distortion noise term at the transmit side. Through this extended MIMO relation, the effects of hardware impairments on the achievable rates of different digital modulation schemes are studied via large system analysis. The numerical results illustrate how tolerable EVM levels depend non-trivially on various factors, including the signal-to-noise ratio, modulation order and the level of asymmetry in antenna array configurations.

Original languageEnglish
Title of host publication2015 IEEE International Conference on Communications (ICC)
PublisherIEEE Institute of Electrical and Electronic Engineers
ISBN (Electronic)978-1-4673-6432-4
ISBN (Print)978-1-4673-6431-7
Publication statusPublished - 9 Sept 2015
MoE publication typeA4 Article in a conference publication
EventIEEE International Conference on Communications, ICC 2015: Smart City & Smart World - London, United Kingdom
Duration: 8 Jun 201512 Jun 2015

Publication series

SeriesIEEE International Conference on Communications


ConferenceIEEE International Conference on Communications, ICC 2015
Abbreviated titleICC
Country/TerritoryUnited Kingdom


Dive into the research topics of 'Asymptotic analysis of asymmetric MIMO links: EVM limits for joint decoding of PSK and QAM'. Together they form a unique fingerprint.

Cite this