A General Framework for Temporal Fair User Scheduling in NOMA Systems

Shahram Shahsavari, Farhad Shirani, Elza Erkip

Research output: Contribution to journalArticle

Abstract

Non-orthogonal multiple access (NOMA) is one of the promising radio access techniques for next generation wireless networks. Opportunistic multi-user scheduling is necessary to fully exploit multiplexing gains in NOMA systems, but compared with traditional scheduling, inter-relations between users' throughputs induced by multi-user interference poses new challenges in the design of NOMA schedulers. A successful NOMA scheduler has to carefully balance the following three objectives: maximizing average system utility, satisfying desired fairness constraints among the users and enabling real-time, and low computational cost implementations. In this paper, scheduling for NOMA systems under temporal fairness constraints is considered. Temporal fair scheduling leads to communication systems with predictable latency as opposed to utilitarian fair schedulers for which latency can be highly variable. It is shown that under temporal fairness constraints, optimal system utility is achieved using a class of opportunistic scheduling schemes called threshold based strategies (TBS). One of the challenges in heterogeneous NOMA scenarios - where only specific users may be activated simultaneously - is to determine the set of feasible temporal shares. A variable elimination algorithm is proposed to accomplish this task. Furthermore, an (online) iterative algorithm based on the Robbins-Monro method is proposed to construct a TBS by finding the optimal thresholds for a given system utility metric. The algorithm does not require knowledge of the users' channel statistics. Rather, at each time-slot, it has access to the channel realizations in the previous time-slots. Various numerical simulations of practical scenarios are provided to illustrate the effectiveness of the proposed NOMA scheduling in static and mobile scenarios

Original languageEnglish (US)
JournalIEEE Journal on Selected Topics in Signal Processing
DOIs
StateAccepted/In press - Jan 1 2019

Fingerprint

Scheduling
Optimal systems
Multiplexing
Wireless networks
Communication systems
Throughput
Statistics
Computer simulation
Costs

Keywords

  • multi-user scheduling
  • Non-orthogonal multiple access
  • Robbins-Monro algorithm
  • temporal fairness
  • threshold-based strategies

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

A General Framework for Temporal Fair User Scheduling in NOMA Systems. / Shahsavari, Shahram; Shirani, Farhad; Erkip, Elza.

In: IEEE Journal on Selected Topics in Signal Processing, 01.01.2019.

Research output: Contribution to journalArticle

@article{da9dd9450c724c7ab0a8910b58304e50,
title = "A General Framework for Temporal Fair User Scheduling in NOMA Systems",
abstract = "Non-orthogonal multiple access (NOMA) is one of the promising radio access techniques for next generation wireless networks. Opportunistic multi-user scheduling is necessary to fully exploit multiplexing gains in NOMA systems, but compared with traditional scheduling, inter-relations between users' throughputs induced by multi-user interference poses new challenges in the design of NOMA schedulers. A successful NOMA scheduler has to carefully balance the following three objectives: maximizing average system utility, satisfying desired fairness constraints among the users and enabling real-time, and low computational cost implementations. In this paper, scheduling for NOMA systems under temporal fairness constraints is considered. Temporal fair scheduling leads to communication systems with predictable latency as opposed to utilitarian fair schedulers for which latency can be highly variable. It is shown that under temporal fairness constraints, optimal system utility is achieved using a class of opportunistic scheduling schemes called threshold based strategies (TBS). One of the challenges in heterogeneous NOMA scenarios - where only specific users may be activated simultaneously - is to determine the set of feasible temporal shares. A variable elimination algorithm is proposed to accomplish this task. Furthermore, an (online) iterative algorithm based on the Robbins-Monro method is proposed to construct a TBS by finding the optimal thresholds for a given system utility metric. The algorithm does not require knowledge of the users' channel statistics. Rather, at each time-slot, it has access to the channel realizations in the previous time-slots. Various numerical simulations of practical scenarios are provided to illustrate the effectiveness of the proposed NOMA scheduling in static and mobile scenarios",
keywords = "multi-user scheduling, Non-orthogonal multiple access, Robbins-Monro algorithm, temporal fairness, threshold-based strategies",
author = "Shahram Shahsavari and Farhad Shirani and Elza Erkip",
year = "2019",
month = "1",
day = "1",
doi = "10.1109/JSTSP.2019.2903745",
language = "English (US)",
journal = "IEEE Journal on Selected Topics in Signal Processing",
issn = "1932-4553",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - A General Framework for Temporal Fair User Scheduling in NOMA Systems

AU - Shahsavari, Shahram

AU - Shirani, Farhad

AU - Erkip, Elza

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Non-orthogonal multiple access (NOMA) is one of the promising radio access techniques for next generation wireless networks. Opportunistic multi-user scheduling is necessary to fully exploit multiplexing gains in NOMA systems, but compared with traditional scheduling, inter-relations between users' throughputs induced by multi-user interference poses new challenges in the design of NOMA schedulers. A successful NOMA scheduler has to carefully balance the following three objectives: maximizing average system utility, satisfying desired fairness constraints among the users and enabling real-time, and low computational cost implementations. In this paper, scheduling for NOMA systems under temporal fairness constraints is considered. Temporal fair scheduling leads to communication systems with predictable latency as opposed to utilitarian fair schedulers for which latency can be highly variable. It is shown that under temporal fairness constraints, optimal system utility is achieved using a class of opportunistic scheduling schemes called threshold based strategies (TBS). One of the challenges in heterogeneous NOMA scenarios - where only specific users may be activated simultaneously - is to determine the set of feasible temporal shares. A variable elimination algorithm is proposed to accomplish this task. Furthermore, an (online) iterative algorithm based on the Robbins-Monro method is proposed to construct a TBS by finding the optimal thresholds for a given system utility metric. The algorithm does not require knowledge of the users' channel statistics. Rather, at each time-slot, it has access to the channel realizations in the previous time-slots. Various numerical simulations of practical scenarios are provided to illustrate the effectiveness of the proposed NOMA scheduling in static and mobile scenarios

AB - Non-orthogonal multiple access (NOMA) is one of the promising radio access techniques for next generation wireless networks. Opportunistic multi-user scheduling is necessary to fully exploit multiplexing gains in NOMA systems, but compared with traditional scheduling, inter-relations between users' throughputs induced by multi-user interference poses new challenges in the design of NOMA schedulers. A successful NOMA scheduler has to carefully balance the following three objectives: maximizing average system utility, satisfying desired fairness constraints among the users and enabling real-time, and low computational cost implementations. In this paper, scheduling for NOMA systems under temporal fairness constraints is considered. Temporal fair scheduling leads to communication systems with predictable latency as opposed to utilitarian fair schedulers for which latency can be highly variable. It is shown that under temporal fairness constraints, optimal system utility is achieved using a class of opportunistic scheduling schemes called threshold based strategies (TBS). One of the challenges in heterogeneous NOMA scenarios - where only specific users may be activated simultaneously - is to determine the set of feasible temporal shares. A variable elimination algorithm is proposed to accomplish this task. Furthermore, an (online) iterative algorithm based on the Robbins-Monro method is proposed to construct a TBS by finding the optimal thresholds for a given system utility metric. The algorithm does not require knowledge of the users' channel statistics. Rather, at each time-slot, it has access to the channel realizations in the previous time-slots. Various numerical simulations of practical scenarios are provided to illustrate the effectiveness of the proposed NOMA scheduling in static and mobile scenarios

KW - multi-user scheduling

KW - Non-orthogonal multiple access

KW - Robbins-Monro algorithm

KW - temporal fairness

KW - threshold-based strategies

UR - http://www.scopus.com/inward/record.url?scp=85062675681&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062675681&partnerID=8YFLogxK

U2 - 10.1109/JSTSP.2019.2903745

DO - 10.1109/JSTSP.2019.2903745

M3 - Article

JO - IEEE Journal on Selected Topics in Signal Processing

JF - IEEE Journal on Selected Topics in Signal Processing

SN - 1932-4553

ER -