Performance Analysis of Kalman-Filter-Based Clock Synchronization in IEEE 1588 Networks

Giorgi, Giada; Narduzzi, C.

doi:10.1109/tim.2011.2113120

Cited by 152 publications

(99 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This section describes a discrete-time dynamic model to characterize the dynamics of the phase variations in h (n,m) (τ ). Based on the two-state models in [14], [15], we define the discrete-time state of the n th transmit node's carrier as x…”

Section: A Oscillator Dynamicsmentioning

confidence: 99%

Asymptotic oscillator tracking performance analysis for distributed massive MIMO systems

Brown

Wang

Dasgupta

2014

2014 48th Annual Conference on Information Sciences and Systems (CISS)

View full text Add to dashboard Cite

Abstract-This paper considers the problem of estimating and tracking channels in a distributed transmission system with Nt transmit nodes and Nr receive nodes. Since each node in the distributed transmission system has an independent local oscillator, the effective channel between each transmit node and each receive node has time-varying phase and frequency offsets which much be tracked and predicted to facilitate coherent transmission. This paper presents a unified state-space model in which a single node aggregates all of the observations and tracks all of the channel states in the system. To quantify the performance of this tracking system, the steady-state prediction covariance is analyzed. Since the dimension of the prediction covariance matrix grows proportionally to the product NtNr, the resulting discrete-time algebraic Riccati equation is difficult to solve for large systems, e.g., distributed massive MIMO systems. It is shown, however, that the structure of the dynamic model allows the steady-state prediction covariance to be efficiently calculated irrespective of the number of transmit and receive nodes. An asymptotic analysis is also presented for systems with a large number of transmit and receive nodes with closed-from results for all of the elements in the asymptotic prediction covariance as a function of the carrier frequency, oscillator parameters, and channel measurement period. Numeric results confirm the analysis and demonstrate the effect of the oscillator parameters on the ability of the distributed transmission system to achieve coherent transmission.

show abstract

Section: A Oscillator Dynamicsmentioning

confidence: 99%

Asymptotic oscillator tracking performance analysis for distributed massive MIMO systems

Brown

Wang

Dasgupta

2014

2014 48th Annual Conference on Information Sciences and Systems (CISS)

View full text Add to dashboard Cite

show abstract

“…In our simulator, the slave clock is modeled by a "simple skew model" (SKM) introduced in [10]. In this model, the variations of offset and skew are considered as random processes and the dynamics of offset and skew are expressed as:…”

Section: B Slave Node's Inaccurate Clockmentioning

confidence: 99%

An Open Source Simulator for IEEE1588 Time Synchronization over 802.11 Networks

Huang

Yang

et al. 2013

2013 European Modelling Symposium

View full text Add to dashboard Cite

As a key technique in distributed systems, time synchronization plays an important role in time sensitive modern industrial wireless local area networks (WLANs). Based on time packet exchange technology, the IEEE 1588 precise time protocol (PTP) achieves high precise and low power consumption time synchronization in wired Ethernet. In WLANs, however, it is difficult for the wireless sensor nodes to acquire accurate timestamps due to their limited resources. Moreover, compared to wired Ethernet, WLANs suffer more from wireless channel sharing, fading and packet collisions, which leads to overwhelming transmission delay jitters. Therefore, analytic solution to PTP's performance in WLANs is very difficult and it is of importance to evaluate PTP's performance by realistic simulation. In this paper, based on the open source OMNeT++ simulation engine, we present the simulator we developed for PTP time synchronization in 802.11 WLANs. The behavior of the PTP time synchronization, the simulation results and the factors that affect the PTP performance are presented and evaluated.

show abstract

“…And the Kalman filter is a kind of hidden Markov model [11,12]. A Kalman filter may not always be realizable due to possible limitations of actual PTP devices, but it is equally of interest because it is known to provide a solution that, under suitable assumptions, can be considered optimal or near optimal [13]. In addition, the Kalman filter has many advantages: (i) lower mean error in tracking a Gaussian-distributed signal; (ii) lower computational overhead when the distribution is stationary; (iii) improve robustness in the presence of lost signals [7].…”

Section: Skew Estimating Model Based On Kalman Filtermentioning

confidence: 99%

A servo design for slave clock in IEEE 1588 synchronization networks

Fan

Liu

et al. 2013

Int J Communication

View full text Add to dashboard Cite

SUMMARYClock synchronization is critical for a variety of applications in communication networks. In this paper, we design a clock servo named VI_PID-K, which consists of a variable integral PID controller and a Kalman filter. It is a pre-correction mechanism for the skew based on feedback principle to compensate for the poor stability of local clocks in IEEE 1588 networks. Our contribution is establishing discrete differential equation models, bringing in a variable integral PID controller to adjust the skew error and a Kalman filter to act as a skew estimator to predict the skew during one synchronization interval. Simulation results demonstrate that VI_PID-K provides a more stable disciplined clock, improves the speed of skew convergence, and reduces Allan variance in a large time scale when compared to PI method or Fixed_Integral method.VI_PID-K can improve the stability of slave clock and reduce the frequency that master clock sends Sync messages, so the network overhead for clock synchronization can be reduced.

show abstract

Performance Analysis of Kalman-Filter-Based Clock Synchronization in IEEE 1588 Networks

Cited by 152 publications

References 11 publications

Asymptotic oscillator tracking performance analysis for distributed massive MIMO systems

Asymptotic oscillator tracking performance analysis for distributed massive MIMO systems

An Open Source Simulator for IEEE1588 Time Synchronization over 802.11 Networks

A servo design for slave clock in IEEE 1588 synchronization networks

Contact Info

Product

Resources

About