Adaptive Resource Scheduling for Energy Efficient QRD Processor with DVFS

Yangxurui Liu; Hemanth Prabhu; Liang Liu; Viktor Öwall

Adaptive Resource Scheduling for Energy Efficient QRD Processor with DVFS

Yangxurui Liu, Hemanth Prabhu, Liang Liu, Viktor Öwall

Forskningsoutput: Konferensbidrag › Konferenspaper, ej i proceeding/ej förlagsutgivet › Peer review

205 Nedladdningar (Pure)

Sammanfattning

This paper presents an energy efficient adaptive QR decomposition scheme for Long Term Evolution Advance (LTE-A) downlink system. The proposed scheme provides a performance robustness to fluctuating wireless channels while
maintaining lower workload on a reconfigurable hardware. A statistic based algorithm-switching strategy is employed in the scheme to achieve workload reduction and stable computing resource requirement for QR decomposition. With run time resource allocation, computing resources are assigned to highest performance gain segments to reduce performance loss. By utilizing the dynamic voltage and frequency scaling (DVFS) technique, we further exploit the potential of power saving in various workload situation while maintaining fixed throughput. The proposed technique brings power reduction upto 57.8% in EVA-5 scenario and 24.4% with a maximum SNR loss of 1 dB in
EVA-70 scenario, when mapped on a coarse grain reconfigurable vector-based platform.

Originalspråk	engelska
Status	Published - 2015
Evenemang	2015 IEEE Workshop on Signal Processing Systems - Hangzhou, Kina Varaktighet: 2015 okt. 14 → 2015 okt. 16

Konferens

Konferens	2015 IEEE Workshop on Signal Processing Systems
Land/Territorium	Kina
Ort	Hangzhou
Period	2015/10/14 → 2015/10/16

Ämnesklassifikation (UKÄ)

Elektroteknik och elektronik

Tillgång till dokument

SiPS_2015_submission_74

http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7345022

2 Doktorsavhandling (sammanläggning)

Efficient Processing and Storage for Massive MIMO Digital Baseband
Liu, Y., 2018, Department of Electrical and Information Technology, Lund University. 147 s.
Forskningsoutput: Avhandling › Doktorsavhandling (sammanläggning)

Öppen tillgång
Fil
Hardware Implementation of Baseband Processing for Massive MIMO
Prabhu, H., 2017 mars 7, The Department of Electrical and Information Technology. 168 s.
Forskningsoutput: Avhandling › Doktorsavhandling (sammanläggning)

Öppen tillgång
Fil

Citera det här

@conference{201dfd4d895b4d1585ddcf9c133e6700,

title = "Adaptive Resource Scheduling for Energy Efficient QRD Processor with DVFS",

abstract = "This paper presents an energy efficient adaptive QR decomposition scheme for Long Term Evolution Advance (LTE-A) downlink system. The proposed scheme provides a performance robustness to fluctuating wireless channels while maintaining lower workload on a reconfigurable hardware. A statistic based algorithm-switching strategy is employed in the scheme to achieve workload reduction and stable computing resource requirement for QR decomposition. With run time resource allocation, computing resources are assigned to highest performance gain segments to reduce performance loss. By utilizing the dynamic voltage and frequency scaling (DVFS) technique, we further exploit the potential of power saving in various workload situation while maintaining fixed throughput. The proposed technique brings power reduction upto 57.8% in EVA-5 scenario and 24.4% with a maximum SNR loss of 1 dB in EVA-70 scenario, when mapped on a coarse grain reconfigurable vector-based platform.",

keywords = "—QR decomposition, multiple-input multiple-output (MIMO), dynamic voltage and frequency scaling (DVFS)",

author = "Yangxurui Liu and Hemanth Prabhu and Liang Liu and Viktor {\"O}wall",

year = "2015",

language = "English",

note = "2015 IEEE Workshop on Signal Processing Systems ; Conference date: 14-10-2015 Through 16-10-2015",

}

TY - CONF

T1 - Adaptive Resource Scheduling for Energy Efficient QRD Processor with DVFS

AU - Liu, Yangxurui

AU - Prabhu, Hemanth

AU - Liu, Liang

AU - Öwall, Viktor

PY - 2015

Y1 - 2015

N2 - This paper presents an energy efficient adaptive QR decomposition scheme for Long Term Evolution Advance (LTE-A) downlink system. The proposed scheme provides a performance robustness to fluctuating wireless channels while maintaining lower workload on a reconfigurable hardware. A statistic based algorithm-switching strategy is employed in the scheme to achieve workload reduction and stable computing resource requirement for QR decomposition. With run time resource allocation, computing resources are assigned to highest performance gain segments to reduce performance loss. By utilizing the dynamic voltage and frequency scaling (DVFS) technique, we further exploit the potential of power saving in various workload situation while maintaining fixed throughput. The proposed technique brings power reduction upto 57.8% in EVA-5 scenario and 24.4% with a maximum SNR loss of 1 dB in EVA-70 scenario, when mapped on a coarse grain reconfigurable vector-based platform.

AB - This paper presents an energy efficient adaptive QR decomposition scheme for Long Term Evolution Advance (LTE-A) downlink system. The proposed scheme provides a performance robustness to fluctuating wireless channels while maintaining lower workload on a reconfigurable hardware. A statistic based algorithm-switching strategy is employed in the scheme to achieve workload reduction and stable computing resource requirement for QR decomposition. With run time resource allocation, computing resources are assigned to highest performance gain segments to reduce performance loss. By utilizing the dynamic voltage and frequency scaling (DVFS) technique, we further exploit the potential of power saving in various workload situation while maintaining fixed throughput. The proposed technique brings power reduction upto 57.8% in EVA-5 scenario and 24.4% with a maximum SNR loss of 1 dB in EVA-70 scenario, when mapped on a coarse grain reconfigurable vector-based platform.

KW - —QR decomposition

KW - multiple-input multiple-output (MIMO)

KW - dynamic voltage and frequency scaling (DVFS)

M3 - Paper, not in proceeding

T2 - 2015 IEEE Workshop on Signal Processing Systems

Y2 - 14 October 2015 through 16 October 2015

ER -

Adaptive Resource Scheduling for Energy Efficient QRD Processor with DVFS

Sammanfattning

Konferens

Ämnesklassifikation (UKÄ)

Tillgång till dokument

Fingeravtryck

Forskningsoutput

Efficient Processing and Storage for Massive MIMO Digital Baseband

Hardware Implementation of Baseband Processing for Massive MIMO

Citera det här