EVM Prediction for Massive MIMO

Wael A G Boukley Hasan; Angela Doufexi; George Oikonomou; Mark A Beach

doi:10.1109/PIMRC.2019.8904437

EVM Prediction for Massive MIMO

Wael A G Boukley Hasan, Angela Doufexi, George Oikonomou, Mark A Beach

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

74 Downloads (Pure)

Abstract

Signal to interference plus noise ratio (SINR) is a widely common performance metric used in the majority of massive multiple-input, multiple-output (Ma-MIMO) research. This metric requires prior knowledge of the user channel vectors and the interference caused by inaccurate channel state information (CSI). However, the interference caused by inaccurate CSI can’t be calculated for realworld scenarios. On the other hand, a comprehensive performance indicator can be achieved by the Error Vector Magnitude (EVM) metric in real-world scenarios. This considers all impairments upon the transmitted symbol as seen at the receiver. However, measuring the EVM values for a subset of users requires each user to retransmit data symbols. This paper presents an estimation method with high accuracy by associating EVM to SINR values for MaMIMO with zero-forcing (ZF) and Minimum Mean Square Error (MMSE). Also introduced is a novel EVM prediction
method for subset of users taken from the original set of simultaneous users in a single cell Ma-MIMO. This method jointly relies on the channel correlation between users and the EVM performance to predict the EVM values for a subset of the available users without the need to retransmit data symbols. This method considers the user channel vector and the interference caused by inaccurate CSI, which makes it suitable for Ma-MIMO algorithms, such as user grouping and power control. Real-world experimental data-sets with real-time results are carried out to validate the EVM prediction method using software-defined radio Ma-MIMO testbed.

Original language	English
Title of host publication	2019 IEEE 30th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC)
Number of pages	7
ISBN (Electronic)	978-1-5386-8110-7
DOIs	https://doi.org/10.1109/PIMRC.2019.8904437
Publication status	Published - 11 Sep 2019
Event	IEEE International Symposium on Personal, Indoor and Mobile Radio Communications - Istanbul, Turkey Duration: 8 Sep 2019 → 11 Sep 2019 Conference number: 30 http://pimrc2019.ieee-pimrc.org/

Conference

Conference	IEEE International Symposium on Personal, Indoor and Mobile Radio Communications
Abbreviated title	IEEE PIMRC 2019
Country	Turkey
City	Istanbul
Period	8/09/19 → 11/09/19
Internet address	http://pimrc2019.ieee-pimrc.org/

Keywords

Massive MIMO
5G
EVM
SINR estimation

Access to Document

10.1109/PIMRC.2019.8904437

Full-text PDF (accepted author manuscript)Accepted author manuscript, 6.02 MB

Fingerprint Dive into the research topics of 'EVM Prediction for Massive MIMO'. Together they form a unique fingerprint.

Cite this

Boukley Hasan, W. A. G., Doufexi, A., Oikonomou, G., & Beach, M. A. (2019). EVM Prediction for Massive MIMO. In 2019 IEEE 30th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC) https://doi.org/10.1109/PIMRC.2019.8904437

@inproceedings{8448efa41ba54a2690ca45232ff442ba,

title = "EVM Prediction for Massive MIMO",

abstract = "Signal to interference plus noise ratio (SINR) is a widely common performance metric used in the majority of massive multiple-input, multiple-output (Ma-MIMO) research. This metric requires prior knowledge of the user channel vectors and the interference caused by inaccurate channel state information (CSI). However, the interference caused by inaccurate CSI can{\textquoteright}t be calculated for realworld scenarios. On the other hand, a comprehensive performance indicator can be achieved by the Error Vector Magnitude (EVM) metric in real-world scenarios. This considers all impairments upon the transmitted symbol as seen at the receiver. However, measuring the EVM values for a subset of users requires each user to retransmit data symbols. This paper presents an estimation method with high accuracy by associating EVM to SINR values for MaMIMO with zero-forcing (ZF) and Minimum Mean Square Error (MMSE). Also introduced is a novel EVM predictionmethod for subset of users taken from the original set of simultaneous users in a single cell Ma-MIMO. This method jointly relies on the channel correlation between users and the EVM performance to predict the EVM values for a subset of the available users without the need to retransmit data symbols. This method considers the user channel vector and the interference caused by inaccurate CSI, which makes it suitable for Ma-MIMO algorithms, such as user grouping and power control. Real-world experimental data-sets with real-time results are carried out to validate the EVM prediction method using software-defined radio Ma-MIMO testbed.",

keywords = "Massive MIMO, 5G, EVM, SINR estimation",

author = "{Boukley Hasan}, {Wael A G} and Angela Doufexi and George Oikonomou and Beach, {Mark A}",

year = "2019",

month = sep,

day = "11",

doi = "10.1109/PIMRC.2019.8904437",

language = "English",

isbn = "978-1-5386-8111-4",

booktitle = "2019 IEEE 30th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC)",

note = "IEEE International Symposium on Personal, Indoor and Mobile Radio Communications , IEEE PIMRC 2019 ; Conference date: 08-09-2019 Through 11-09-2019",

url = "http://pimrc2019.ieee-pimrc.org/",

}

TY - GEN

T1 - EVM Prediction for Massive MIMO

AU - Boukley Hasan, Wael A G

AU - Doufexi, Angela

AU - Oikonomou, George

AU - Beach, Mark A

N1 - Conference code: 30

PY - 2019/9/11

Y1 - 2019/9/11

N2 - Signal to interference plus noise ratio (SINR) is a widely common performance metric used in the majority of massive multiple-input, multiple-output (Ma-MIMO) research. This metric requires prior knowledge of the user channel vectors and the interference caused by inaccurate channel state information (CSI). However, the interference caused by inaccurate CSI can’t be calculated for realworld scenarios. On the other hand, a comprehensive performance indicator can be achieved by the Error Vector Magnitude (EVM) metric in real-world scenarios. This considers all impairments upon the transmitted symbol as seen at the receiver. However, measuring the EVM values for a subset of users requires each user to retransmit data symbols. This paper presents an estimation method with high accuracy by associating EVM to SINR values for MaMIMO with zero-forcing (ZF) and Minimum Mean Square Error (MMSE). Also introduced is a novel EVM predictionmethod for subset of users taken from the original set of simultaneous users in a single cell Ma-MIMO. This method jointly relies on the channel correlation between users and the EVM performance to predict the EVM values for a subset of the available users without the need to retransmit data symbols. This method considers the user channel vector and the interference caused by inaccurate CSI, which makes it suitable for Ma-MIMO algorithms, such as user grouping and power control. Real-world experimental data-sets with real-time results are carried out to validate the EVM prediction method using software-defined radio Ma-MIMO testbed.

AB - Signal to interference plus noise ratio (SINR) is a widely common performance metric used in the majority of massive multiple-input, multiple-output (Ma-MIMO) research. This metric requires prior knowledge of the user channel vectors and the interference caused by inaccurate channel state information (CSI). However, the interference caused by inaccurate CSI can’t be calculated for realworld scenarios. On the other hand, a comprehensive performance indicator can be achieved by the Error Vector Magnitude (EVM) metric in real-world scenarios. This considers all impairments upon the transmitted symbol as seen at the receiver. However, measuring the EVM values for a subset of users requires each user to retransmit data symbols. This paper presents an estimation method with high accuracy by associating EVM to SINR values for MaMIMO with zero-forcing (ZF) and Minimum Mean Square Error (MMSE). Also introduced is a novel EVM predictionmethod for subset of users taken from the original set of simultaneous users in a single cell Ma-MIMO. This method jointly relies on the channel correlation between users and the EVM performance to predict the EVM values for a subset of the available users without the need to retransmit data symbols. This method considers the user channel vector and the interference caused by inaccurate CSI, which makes it suitable for Ma-MIMO algorithms, such as user grouping and power control. Real-world experimental data-sets with real-time results are carried out to validate the EVM prediction method using software-defined radio Ma-MIMO testbed.

KW - Massive MIMO

KW - 5G

KW - EVM

KW - SINR estimation

U2 - 10.1109/PIMRC.2019.8904437

DO - 10.1109/PIMRC.2019.8904437

M3 - Conference Contribution (Conference Proceeding)

SN - 978-1-5386-8111-4

BT - 2019 IEEE 30th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC)

T2 - IEEE International Symposium on Personal, Indoor and Mobile Radio Communications

Y2 - 8 September 2019 through 11 September 2019

ER -