分布式机会调度协议优化D2D通信公平性

99 浏览量更新于2024-08-29 收藏 181KB PDF 举报

本文主要探讨了一种分布式机会性调度协议(Distributed Opportunistic Scheduling Protocol, DO-Fast)，针对基于正交频分复用(Orthogonal Frequency Division Multiplexing, OFDM)的设备到设备(Device-to-Device, D2D)通信系统。在当前的研究背景下，随着移动设备的普及和网络密度的提升，D2D通信作为一种无线直接通信方式，能够显著提高频谱效率和用户体验，尤其在大规模事件驱动的通信场景中。 D2D通信中的调度问题至关重要，因为它直接影响着多个用户间的公平性和系统性能。传统的集中式调度方法可能无法充分利用多路径传输带来的空间多样性，同时可能会导致某些链路资源被频繁占据，而其他链路资源利用率低下。为了克服这些问题，作者提出了DO-Fast协议，它是一种异步的、分布式的机会性调度策略，旨在实现公平性和资源的有效利用。 DO-Fast的关键在于引入了机会性调度机制，允许具有更好通道条件的链接优先获得访问权，从而提高了系统的整体效率。通过分布式的方式交换本地的信道状态信息(Channel State Information, CSI)，每个设备能够实时评估其与其他设备之间的连接质量，根据当前的无线环境动态调整优先级。此外，它结合了轮询策略，确保所有D2D链路都有平等的机会参与通信，避免了过度集中化的潜在问题。这种协议的优势在于它能够平衡多个D2D用户的竞争，减少了拥塞和冲突，同时通过利用多径效应来增强通信可靠性。DO-Fast的实施有助于优化频谱资源分配，提高链路利用率，并且由于其分布式特性，它具有较低的复杂性和对网络基础设施的依赖性，使得它在现实无线环境中更具适用性。然而，实现DO-Fast还需要考虑实际应用中的挑战，如网络同步问题、节点能量消耗以及数据传输的安全性。未来的研究可能着重于进一步优化协议算法以适应更复杂的网络环境，或者集成更多的无线技术，如多输入多输出(MIMO)和无线充电，以进一步提升D2D通信的性能。这篇文章对于理解分布式机会性调度在D2D通信中的作用，以及如何在公平性和效率之间找到平衡提供了有价值的研究成果，为无线通信领域的研究人员和工程师提供了新的设计思路和技术参考。

A Distributed Opportunistic Scheduling Protocol for

Device-to-Device Communications

Junyu Liu, Min Sheng, Member, IEEE, Yan Zhang, Member, IEEE,

Xijun Wang, Member, IEEE, Hongguang Sun and Yan Shi, Member, IEEE

State Key Laboratory of Integrated Service Networks

Institute of Information Science, Xidian University, Xi’an, Shaanxi, 710071, China

Email: jyliu@stu.xidian.edu.cn, msheng@mail.xidian.edu.cn, {yanzhang, xijunwang, hgsun, yshi}@xidian.edu.cn

Abstract—In this paper, we consider the distributed scheduling

problem for the OFDM based device-to-device (D2D) commu-

nications. In order to fully exploit the spatial diversity of the

channel variation as well as provide access fairness for all

D2D links, we propose a synchronous Distributed Opportunistic

scheduling protocol under Fairness constraints (DO-Fast). DO-

Fast incorporates the opportunistic scheduling with a round-

robin strategy. By exchanging local Channel State Information

(CSI) in a distributed way, the opportunistic scheduling strategy

enables the links with better channel conditions to take prece-

dence for higher access priorities. It leads to more concurrent

transmissions and higher system throughput than the random

scheduling strategy, where links are allocated with priorities in

a random manner regardless of channel conditions. Meanwhile,

we prompt a round-robin strategy so that the D2D links would

take high priorities alternately, which guarantees the short-term

fairness requirements of the links with poor channel conditions.

We show via simulations that DO-Fast achieves throughput

improvement over the existing scheduling protocol from the

network perspective with acceptable delay performance.

I. INTRODUCTION

Device-to-device (D2D) communications as an underlay to

cellular networks have been introduced as a promising compo-

nent to LTE-Advanced [1]. D2D pairs exchange information

directly rather than interact signals with the cellular base

station (eNB), achieving efﬁcient spatial reuse of the radio

resources.

D2D communications can be categorized into two modes

in terms of the operator control level [2]. The ﬁrst mode is

the fully controlled mode, where the eNB has full control

over the D2D users [3], [4], e.g., connection setup, resource

allocations, etc. The other one is the loosely controlled mode,

in which D2D communications are conducted autonomously

with little aid from the eNB. In this case, D2D links utilize

either unlicensed spectrum or dedicated carriers in licensed

spectrum so as to avoid interference to the cellular users

[5], [6]. In order to fully exploit the spectrum resources, the

scheduling for the D2D links is critical in both modes.

There has been much research on the scheduling for D2D

This paper is supported by National Natural Science Foundation of China

(61172079, 61231008, 61201141), China SAMSUNG Telecom R&D Center,

973 Program (2009CB320404), 111 Project (B08038), National S&T Major

Project (2012ZX03002009-003, 2012ZX03004002-003), Shaanxi Province

Science and Technology Research and Development Program (2011KJXX-

40).

communications. The SIR criterion based scheduling strategies

in [4] enable D2D users to utilize the spatial reuse in a more

efﬁcient way. In addition, priority scheduling facilitates inter-

ference and radio resource management in D2D networks by

assigning high priorities to desirable users and ﬂows [7], [8].

Moreover, the opportunistic subchannel scheduling scheme

proposed in [9] considers the time-varying channel condition

and QoS requirement of each user in order to maximize the

average sum rate of the system.

However, most work focuses on the centralized scheduling

schemes supported by the eNB. These scheduling strategies

may not be applicable where the eNB only provides access

authentication and synchronization for the D2D users in most

cases. Therefore, new challenges have been posed to the

distributed scheduling for D2D communications.

Recently, Qualcomm Inc. has designed and implemented an

OFDM-based synchronous Medium Access Protocol (MAC)

and Physical Layer (PHY) architecture for D2D communi-

cations in [6], referred to as FlashLinQ, which could be

deployed in the mixed license-unlicensed spectrum with loose

control from the eNB. With all devices globally synchronized,

FlashLinQ has introduced a distributed priority allocation

scheme to maintain fairness across links by randomizing their

access priorities over time. Moreover, an analog signal scheme

is incorporated to schedule the D2D links based on the SIR

criterion. Simulation results in [6] showed that FlashLinQ

outperforms the conventional 802.11g in both throughput and

delay performance. However, without considering the Channel

State Information (CSI), the random priority allocation scheme

adopted in FlashLinQ may possibly assign poor-quality links

high access priorities, thereby resulting in inferior throughput

performance.

Motivated by the prior work, we focus on the distributed

scheduling for D2D communications and propose a syn-

chronous Distributed Opportunistic scheduling protocol under

Fairness constraints (DO-Fast) compatible with the framework

in [6], aiming to improve the throughput performance from the

network point of view. The results show that DO-Fast achieves

throughput improvement over FlashLinQ as well as guarantees

the delay performance.

The rest of this paper is organized as follows. Section II

describes the system model. In Section III, the beneﬁt of

2013 IEEE 24th International Symposium on Personal, Indoor and Mobile Radio Communications: MAC and Cross-Layer Design

Track

下载后可阅读完整内容，剩余4页未读，立即下载

weixin_38545768

粉丝: 8
资源: 941

分布式机会调度协议优化D2D通信公平性

Task Scheduling for Multi-core and Parallel Architectures 2017

Resilient Distributed Datasets: A Fault-Tolerant Abstraction for In-Memory

Distributed model predictive control for plant-wide hot-rolled strip laminar cooling process

Application of Golay codes to distributed optical fiber sensor for long-distance oil pipeline leakage and external damage detection

Vulcan - Binary Transformation in a Distributed Environment (tr-2001-50)-计算机科学

RS-CPS: A Distributed Architecture of Robotic Surveillance Cyber-Physical System

Recovery schemes for different distributed connection management in generalized multi-protocol label switching networks

Dipper - Distributed PABX Reporting-开源

RabbitMQ-in-Action-Distributed-Messaging-for-Everyone.pdf

Distributed scheduling scheme for video streaming over multi-channel multi-radio multi-hop wireless networks

最新资源