International Scientific Journal on Science Engineering & Technology

Vo1ume 17, No. 04, May 2014

189

Oscillator Characterization and Close Loop

Adaptation for IEEE 1588 Timing and
Synchronization
Karuppusamy.M, Dr. M. Suganthi and D. Balaji
Abstract--- The oscillators used in synchronization
modules of CDMA and 3G base stations are huge in size

I.

INTRODUCTION

and expensive. Applying cost effective and smaller, even

As a result of growth in the mobile telecommunication

more inaccurate, oscillators in timing modules are

and networking, lots of remote smaller one rack unit sized

progressing research theme. The IEEE 1588v2 PTP

equipments are being placed which either can act as a base

synchronization scheme along with an adaptive control

station equipment or mobile backhaul router. Because of the

algorithm is presented to enhance the oscillators to meet the

network growth, economy aspects and the statistical

requirements of base stations during holdover mode

multiplexing advantages of packet networking technology

especially situations like GPS is lost at the master and at the

the circuit switching networking is replaced in a rapid

time of network outage. An oscillator frequency stability

manner. This adds challenges to the next generation

model is developed for the adaptive control algorithm. This

technologies like 3G, 4G Wireless deployments. In order to

model takes into account the control loop which creates the

ensure smooth call hand off in the cell sites the timing and

correction signal when the timing module is in locked

synchronization plays a vital role.

mode. A recursive prediction error method is used to

Althoughthevarioustabletextstylesareprovided.Theformat

identify the system model parameters. Simulation results

terwillneedtocreatethesecomponents,incorporatingtheapplica

show that an oscillator enhanced by this scheme improves

blecriteriathatfollow.This

the synchronization performance significantly, compared

controlled crystal oscillator (OCXO) can be used in the

with uncorrected oscillators. NS2 tool is used for the

timing module along with an adaptive closeloop control

simulation of the PTP behavior. The results also show the

mechanism is used to predict and accommodate the

benefit of explicitly modeling and analysis of the control

oscillator variation. Using an OCXO in the timing module,

loop. Finally the results show that with IEEE 1588 slave in

the physical size and power consumption of the timing

operation the base station can provide an enhanced hold

module are reduced. The resulting timing module can be

over duration and even if master is lost for hours.

integrated onto the base station modem card and the cost is

Keywords--- Oscillator, IEEE1588, Adaptive Control,

paper

shows

that

an

oven

further reduced.

OCXO, PDV

II.

BACKGROUND

In order to achieve required accuracy of synchronization

GPS receiver would need to be placed in each of the base
station location (BTS, NodeB or eNodeB). From the
Karuppusamy.M, M.E in Communication Systems, Mahendra College
of Engineering, Salem, India. E-mail: mkaruppu@gmail.com
Dr.M. Suganthi, Professor and HOD, Electronics and Communication
Department, Mahendra College of Engineering, Salem
D. Balaji, M.E, Assistant Professor, Electronics and Communication
Department, Mahendra College of Engineering, Salem, India

operational perspective especially from external jammers

intruding to GPS operation, cost of equipment and ensuring
the satellite signal availability at all the locations, it creates

ISSN: 6814-4794 @ ISJSET

International Scientific Journal on Science Engineering & Technology

Vo1ume 17, No. 04, May 2014

190

additional operational challenges. So, there is a suitable

deployment adaptation. There is also network latency

mechanism accurate mechanism required to transport the

created by the hubs, routers, switches, cables and other

timing and synchronization to the base station devices over

hardware existing between clocks.

the packet networking and at the same time having a cost

effective (low stratum quality oscillator at the BTS) would
be well suitable from the network CAPEX and OPEX
perspective. The objective of the paper is to analyze different
methods that can be used to synchronize the timing in a
network,

to

understand

the

problems

causing

bad

The more recently developed protocol is the IEEE 1588

Precision Time Protocol (PTP). The IEEE 1588 PTP
overcomes the latency and jitter issues through hardware
time stamping at the physical layer of the network resulting
in sub micro second accuracy (10s of nano seconds). The
IEEE 1588 PTP is cost effective as it uses existing Ethernet

synchronization in a network and propose a solution to

LANs. The key factor in achieving the higher accuracy in

improve the synchronization accuracy. This also involves

IEEE 1588 is the use of boundary or transparent clock,

understanding of why the clocks drift, what are the factors

affecting the accuracy and the stability of the clock
oscillators and what can be done to counter those factors.

intelligence of the algorithm implementation at the timing

recovery end along with hardware assisted time stamping.
The hardware time stamping mitigates the system latency
while the deployment of boundary or transparent clocks

III.

SYNCHRONIZATION ISSUES

mitigates the network latency.

A clock accuracy and stability is directly related to the

frequency accuracy and stability of the underlying crystal

V.

oscillator. The frequency accuracy of an oscillator is a

measure of offset from a specified target. The factors
establishing the frequency accuracy of an oscillator include
the temperature, ageing and retrace. The frequency stability
of a crystal oscillator is determined over a period of time by
measuring the oscillator frequency variations from its
operational frequency. The factors affecting the frequency
stability of an oscillator include the reference signal noise (if
the oscillator is locked to a reference signal such as GPS),
oscillator tuning port noise, supply rail noise and the crystal
vibration

EXISTING SYSTEM
The closed loop predictive correction mechanism is

based an adaptive correction algorithm is to enhance the

oscillator performance in holdover mode. The oscillator used
in the timing module on a base station is usually phase
locked by a one pulse per second (1-pps) GPS signal. When
the GPS signal is lost, the timing module enters holdover
mode and the timing module accuracy becomes dependent
on the local oscillator. The adaptive correction model helps
to sustain the synchronization characteristics based on the
learnt pattern. The cumulative time error (CTE) should be
kept below 10microseconds for 3G and LTE deployments
which in fact needs a quite expensive oscillator system so

IV.

NETWORK SYNCHRONIZATION VIA

PACKETS

this adaptive correction mechanism becomes necessary.

Network synchronization means that the frequency,

The 10 s CTE amounts ~0.35 ppb of frequency drift

phase and time is synchronized with the synchronous timing

and therefore require an expensive oscillator such as a

supply unit. This is generally accomplished through line

double oven controlled crystal oscillator (DOCXO) to meet

timing, GPS central timing or synchronizing via protocols

the

like

and

temperature range of the oscillator. It is also suggested

synchronization attracts the deployments very much because

employing an adaptive model f the timing module during the

of lower in CAPEX and OPEX cost and also in terms of

locked period and then using the resulting model to correct

IEEE1588.

The

packet

based

timing

stability

requirement

over

extended

operational

the oscillator drifts in holdover mode. The demonstrated

ISSN: 6814-4794 @ ISJSET

International Scientific Journal on Science Engineering & Technology

Vo1ume 17, No. 04, May 2014

191

algorithm comprises of two parts one running on the Master

show the prediction mechanism follows the adaptive filter

side and another on the slave side. Figure 1 shows the

mechanism along with the PDV variation estimation using

modules present in the timing master end and having a

Gamma Distribution function.

correction predictor to adapt the Oscillator variation and

GPS outage adaptation. Adaptive filters like Kalman filter
mechanism allows the GPS outage and Oscillator variation
prediction. The Kalman filter model assumes that the state of
a system at a time t evolved from the prior state at time t-1
according to the equation
xt= Ftxt-1 + Btut+ wt
wherextis the state vector containing the parameter of

Fig. 2: Adaptive Correction on Slave

interest , utis the vector containing any control, Ft is the

state transition matrix which applies the effect of each
system state parameter at time t-1 on the system state at time
t , Btis the control input matrix
GPS receiver receives GPS signal and provides 1pps
signal and use it as a reference to generate PTP messages
from the PTP master
OCXO
Fig. 3: Gamma Distribution over Kalman prediction

Feedback control (correction predictor)

PTP synthesizer (on the slave)

VII.

DPLL

RESULTS AND CONCLUSION

Perl and Tcl scripts are written simulate the network

conditions and simulate the overall synchronization scheme
for the existing system and proposed system.

Fig. 1: Adaptive Correction on Master

VI.

PROPOSED SYSTEM

The proposed system considers estimating, predicting the

network packet delay variation (PDV) and adapts to slave
PTP recovery. The below figure 2 shows the modules
present in the slave including adaptive control to the PTP
clock synthesizer.The master end correction handles the
GPS outage compensation and

on the

slave

side

compensates the network outage compensation.Figure 3

Figure 5: Holdover Frequency Accuracy Measured without

having the 1588 Adaptation at Slave
ISSN: 6814-4794 @ ISJSET

International Scientific Journal on Science Engineering & Technology

Vo1ume 17, No. 04, May 2014

192

Figure 5 and 6 show the existing system results. The

existing system hold-over the synchronization requirements
for few hours at the master side of the timing equipment and
primarily does take care of the GPS receiver or signal
outage. Figure 7 and 8 show the proposed system results and
hold-over accuracies are approximately 3 times improved
than the existing system. The proposed mechanism can be
either implemented in the intermediate nodes or in the BTS
or NodeB elements.

Figure 8: Holdover Phase Accuracy Measured at Slave using

1588 Adaptation at Slave

REFERENCES
[1]
[2]
[3]

[4]
Figure 6: Holdover Phase Accuracy measured at slave
without having the 1588 Adaptation

[5]

[6]

[7]

[8]

[9]

Figure 7: Holdover Frequency Accuracy Measured at Slave

[10]

using 1588 Adaptation at Slave

[11]

Symmetricom white paper, The Importance of

Network Time Synchronization, 2009.
Symmetricom white paper, The Five Dangers of
Poor Network Timekeeping, 2009.
J. Vig, Quartz Crystal Resonators and Oscillators
for Frequency Control and Timing Applications - A
Tutorial, 2004 IEEE International Frequency
Control Symposium Tutorials, May 2004.
W. Zhou, Time, Frequency Measurement and
Control Technology, Xidian University Press,
2006. ISBN-10: 7560616720
T. Pialis and K. Phang, Analysis of Timing Jitter
in Ring Oscillators Due to Power Supply Noise,
Circuits and Systems, ISCAS03, Proceedings of
the 2003 International Symposium, pp. I-685 I688, 2003.
J. Vig, Introduction to Quartz Frequency
Standards Technical Report SLCET-TR-92-1,
Army Research Laboratory, Electronics and Power
Sources Directorate, 1992.
Symmetricom white paper, IEEE 1588 Precise
Time Protocol: The new standard in time
synchronization, 2009.
Symmetricom white paper, Improving Real World
Synchronization Accuracy with IEEE-1588
Transparent Clocks, 2009.
IEEE Standard for a Precision Clock
Synchronization
Protocol
for
Networked
Measurement and Control Systems, 2008.
H. Weibel, High Precision Clock Synchronization
according to IEEE
1588
Implementation
and Performance Issues, in Proc. Embedded
World 2005.
N. M. Freris, S. R. Graham, P. R. Kumar,
"Fundamental Limits on Synchronization
of

ISSN: 6814-4794 @ ISJSET

International Scientific Journal on Science Engineering & Technology

Vo1ume 17, No. 04, May 2014

[12]

[13]

[14]

[15]

[16]

[17]
[18]

[19]

[20]
[21]
[22]

[23]

[24]

[25]

Affine Clocks in Networks", Automatic Control,

IEEE Transactions on, On page(s): 1352 - 1364,
Volume: 56 Issue: 6,
June 2011.
D. Khler, A Practical Implementation of an IEEE
1588 supporting Ethernet Switch, Proc. IEEE Int.
Symp. Precision Clock
Synchronization Meas.,
Control Commun., pp. 134137, Oct. 2007.
J. Han, H. Chi and D-K.Jeong, A Clock
Synchronization System with IEEE 1588-2008
Adapters over Existing Gigabit Ethernet
Equipment, Circuits and Systems (ISCAS),
Proceedings of 2010
IEEE
International
Symposium, pp. 193 196, Aug. 2010.
H. Weibel and D. Bchaz, IEEE 1588
Implementation and
Performance of Time
Stamping Techniques, Proceedings of the 2004
Conference on IEEE 1588, Sep. 2004.
J. Han and D.K. Jeong, Practical Considerations in
the Design and Implementation
of
Time
Synchronization Systems Using IEEE 1588,
Communications Magazine, IEEE , vol.47, no.11,
pp. 164-170, Nov
2009.
N. Shenoy, J. Fischer, P. Myers and Z. Qui, A
Simulation Study
Performance of IEEE
1588 over Ethernet and IEEE 802.11b WLANs,
IEEE Conference, 2005.
CrystalOscillatorhttp://en.wikipedia.org/wiki/Cryst
al_oscillator
H. Zhou, C. Nicholls, T. Kunz and H. Schwartz,
"Frequency Accuracy
&
Stability
Dependencies of Crystal Oscillators", Carleton
University,
Systems
and
Computer
Engineering, Technical Report SCE-08-12,
November 2008.
H. Zhou, T. Kunz and H. Schwartz, "Adaptive
Correction Method for an OCXO and Investigation
of Analytical Cumulative Time Error
Upper
Bound", IEEE Transactions on Ultrasonic,
Ferroelectrics, and Frequency Control, vol. 58, no.
1, January 2011.
ITU-T G.8261/Y.1361., April 2008, Timing and
synchronization aspects in packet networks
NS-2 Simulator http://isi.edu/nsnam/ns/
T. Issariyakul and E. Hossain, Introduction to
Network Simulator
NS2, Springer Science,
Business Media, LLC, 2009.
C. H. Rentel and T. Kunz, "A Mutual Network
Synchronization Method for Wireless Ad Hoc and
Sensor Networks", IEEE transaction
on
mobile computing, vol. 7, no. 5, May 2008.
L. Xueqiao, C. Yuan, L. Shuang and D. Yaxing,
"Implementation and
Research of Hardware
Time Stamping Techniques Based on IEEE1588",
communication Software and Networks (ICCSN),
IEEE 3rd International Conference, pages 6 - 9,
May 2011.
Realize a Frequency Stability Measurement System
with PTP TeKwei Wang, Fan-Ren Chang,

193

Member, IEEE, and Shinn-Yan Lin, Member,

IEEE, May 2011.

ISSN: 6814-4794 @ ISJSET