Traffic estimation based on long short-term memory
neural network for mobile front-haul with XG-PON
Min Zhang (张 敏)
1
,BoXu(许 渤)
1,
*, Xiaoyun Li (栗晓云)
2
, Yi Cai (蔡 怡)
1
,
Baojian Wu (武保剑)
1
, and Kun Qiu (邱 昆)
1
1
Key Laboratory of Optical Fiber Sensing and Communications, Ministry of Education, University of Electronic
Science and Technology of China, Chengdu 611731, China
2
Business School, University of International Business and Economics, Beijing 100029, China
*Corresponding author: xubo@uestc.edu.cn
Received January 28, 2019; accepted April 12, 2019; posted online June 25, 2019
A novel predictive dynamic bandwidth allocation (DBA) method based on the long short-term memory (LSTM)
neural network is proposed for a 10-gigabit-capable passive optical network in mobile front-haul (MFH) links. By
predicting the number of packets that arrive at the optical network unit buffer based on LSTM, the round-trip
time delay in traditional DBAs can be eliminated to meet the strict latency requirement for MFH links. Our
study shows that the LSTM neural network has better performance than feed-forward neural networks. Based on
extensive simulations, the proposed scheme is found to be able to achieve the latency requirement for MFH and
outperforms the traditional DBAs in terms of delay, jitter, and packet loss ratio.
OCIS codes: 060.4250, 060.4510.
doi: 10.3788/COL201917.070603.
The cloud radio access network (C-RAN) is one of the key
technologies for fifth generation mobile communication
(5G)
[1–3]
. In a C-RAN, the digital baseband processing
units (BBUs) are moved from mobile base station sites
to a central location known as the BBU pool that serves
a group of distributed radio units known as remote radio
heads (RRHs)
[1]
. Mobile front-haul (MFH) is an optical
link that connects RRHs in multiple locations to the
BBU pool
[2]
. The number of MFH links is expected to
increase with the increasing traffic requirements of the
5G systems. To reduce the MFH link cost, a time division
multiplexed passive optical network (TDM-PON) is pro-
posed as it allows sharing of optical fibers and transmis-
sion equipment
[2,3]
. However, a TDM-PON suffers from
a large latency for forwarding uplink traffic because an
optical network unit (ONU) has a waiting time of several
milliseconds in a typical dynamic bandwidth allocation
(DBA) scheme
[3]
. This transmission waiting time in the
ONU is a critical problem since the latency requirement
for the MFH link is very strict, e.g., less than 250 μs
defined by the Third Generation Partnership Project
(3GPP)
[4]
.
Different methods have been proposed in the literature
to solve the latency issue of TDM-PON
[5–9]
. For example,
fixed bandwidth allocation (FBA) is used to meet the
latency requirement for MFH links
[3]
. However, the band-
width usage efficiency is low and the number of ONUs that
can be accommodated is limited by the FBA algorithm
[3,5]
.
Instead, a statistical DBA scheme has been proposed to
improve the bandwidth usage efficiency
[7]
. A disadvantage
of the statistical DBA is that it cannot deal with burst
of MFH traffic. Reference [
8] evaluated the performance
of group-assured GIANT (gGIANT) and round-robin
DBA (RR-DBA). Results show that neither RR-DBA
nor gGIANT satisfies the delay requirement for MFH,
and RR-DBA has a lower upstream delay than gGIANT.
Machine learning (ML), a branch of artificial intelli-
gence (AI), is regarded as one of the most promising
methodological approaches to performing different types
of network data analysis for automated network self-
configuration
[10]
. Recently, machine learning techniques
have been successfully applied in optical communication
and optical networks to improve the intelligence of such
systems
[10–12]
. With its powerful modeling capabilities,
artificial intelligence is also desirable to help solving the
latency issue of TDM-PON for MFH applications.
In this Letter, we propose a long short-term memory
(LSTM)-based predictive DBA method for a slow-latency
10-gigabit-capable passive optical network (XG-PON)
MFH for a C-RAN based on traffic estimation in Fig.
1.
First, the problem of predicting the number of packets
that arrive at the ONU is formulated as an ML function
approximation problem. Second, LSTM is investigated for
this problem and compared to a feed-forward neural net-
work (FNN). Results show that the LSTM neural network
has better prediction performance than FNN. Finally, the
Fig. 1. MFH architecture based on the XG-PON system.
COL 17 (7), 070603(2019) CHINESE OPTICS LETTERS July 2019
1671-7694/2019/070603(6) 070603-1 © 2019 Chinese Optics Letters