Performance bounds for phase offset estimation in IEEE 1588 synchronization

There has been recent interest in the use of packet-based synchronization techniques based on the IEEE 1588 Precision Time Protocol, in order to meet challenges arising in mobile telecommunication networks. An important problem in this area is to design estimators that determine the phase offset of slave clocks, while being resilient to the degrading effects of random network traversal times. While a number of simple non-parametric estimation techniques for this problem have been described in literature, little is known about the best theoretically achievable accuracy. In this paper, we address this issue by describing two novel lower bounds on the error variance of estimators for such problems. These bounds are obtained by adapting two classical Bayesian estimation bounds, namely the Weiss-Weinstien bound and the Ziv-Zakai bound, to a non-Bayesian estimation scenario. The results provide new insights into scenarios where existing phase offset estimation techniques perform well, and where significant scope for performance improvements exists.

[1]  Philippe Forster,et al.  A Fresh Look at the Bayesian Bounds of the Weiss-Weinstein Family , 2008, IEEE Transactions on Signal Processing.

[2]  Ehud Weinstein,et al.  A general class of lower bounds in parameter estimation , 1988, IEEE Trans. Inf. Theory.

[3]  N. Draper,et al.  Applied Regression Analysis , 1966 .

[4]  Cheng-Xiang Wang,et al.  Dealing with Packet Delay Variation in IEEE 1588 Synchronization Using a Sample-Mode Filter , 2013, IEEE Intelligent Transportation Systems Magazine.

[5]  Ravi Subrahmanyan Timing Recovery for IEEE 1588 Applications in Telecommunications , 2009, IEEE Transactions on Instrumentation and Measurement.

[6]  Desmond C. McLernon,et al.  On the Clock Offset Estimation in an Improved IEEE 1588 Synchronization Scheme , 2013, EW.

[7]  Hans Weibel,et al.  Synchronizing IEEE 1588 clocks under the presence of significant stochastic network delays , 2005 .

[8]  Jacob Ziv,et al.  Some lower bounds on signal parameter estimation , 1969, IEEE Trans. Inf. Theory.

[9]  Ilija Hadzic,et al.  Improving IEEE 1588v2 clock performance through controlled packet departures , 2010, IEEE Communications Letters.

[10]  Ilija Hadzic,et al.  Adaptive packet selection for clock recovery , 2010, 2010 IEEE International Symposium on Precision Clock Synchronization for Measurement, Control and Communication.

[11]  Dennis R. Morgan,et al.  A Synchronization Algorithm for Packet MANs , 2011, IEEE Transactions on Communications.

[12]  Cheng-Xiang Wang,et al.  A sample-mode packet delay variation filter for IEEE 1588 synchronization , 2012, 2012 12th International Conference on ITS Telecommunications.

[13]  Yossef Steinberg,et al.  Extended Ziv-Zakai lower bound for vector parameter estimation , 1997, IEEE Trans. Inf. Theory.

[14]  Philipp Reinecke,et al.  Model-Based Evaluation and Improvement of PTP Syntonisation Accuracy in Packet-Switched Backhaul Networks for Mobile Applications , 2011, EPEW.

[15]  C. R. Rao,et al.  Information and the Accuracy Attainable in the Estimation of Statistical Parameters , 1992 .

[16]  S. Johannessen Time synchronization in a local area network , 2004, IEEE Control Systems.

[17]  Yukio Horiuchi,et al.  A packet filtering mechanism with a packet delay distribution estimation function for IEEE 1588 time synchronization in a congested network , 2011, 2011 IEEE International Symposium on Precision Clock Synchronization for Measurement, Control and Communication.

[18]  Aggelos Bletsas,et al.  Evaluation of Kalman filtering for network time keeping , 2003, Proceedings of the First IEEE International Conference on Pervasive Computing and Communications, 2003. (PerCom 2003)..

[19]  E. Barankin Locally Best Unbiased Estimates , 1949 .