On a Perturbation Approach for the Analysis of Stochastic Tracking Algorithms
adaptive tracking algorithmsexpansion techniqueslinear regressionperformance analysistime-varying parameter
Signal detection and filtering (aspects of stochastic processes) (60G35) Linear systems in control theory (93C05) Estimation and detection in stochastic control theory (93E10) Identification in stochastic control theory (93E12) Least squares and related methods for stochastic control systems (93E24)
This paper considers the estimation problem of the time-varying vector parameter in a scalar linear regression model. This model encompasses many different applications, including channel equalization, time delay estimation and echo cancellation. To track the variations of the parameter, it is customary to use a recursive algorithm for updating an estimator of the unknown parameter. These algorithms may take many different forms depending on what one is willing to assume on the observation noise, the parameter variations, and the amount of computation that is acceptable; e.g., standard stochastic approximation with fixed step-size, recursive least-squares with forgetting factor or adaptations of Kalman-Bucy filters. In this paper, a perturbation expansion technique is introduced to decompose the tracking error of a general adaptive tracking algorithm. This method results in a tracking error bound and tight approximate expressions for the moments of the tracking error. These expressions allow the evaluation, both qualitatively and quantitatively, of the impact of several factors on the tracking error performance.
- Tracking randomly varying parameters: Analysis of a standard algorithm
- Perturbation expansions in polynomial root tracking
- Analyzing noise-induced tracking errors in control systems with saturation: a stochastic linearization approach
- Bridging the gap between constant step size stochastic gradient descent and Markov chains
- A recursive online algorithm for the estimation of time-varying ARCH parameters
- On the solution of stochastic optimization and variational problems in imperfect information regimes
- Tracking properties of adaptive signal processing algorithms
- Performance estimates in tracking under white-noise perturbations
- Frequency domain tracking characteristics of adaptive algorithms
- A result on the mean square error obtained using general tracking algorithms
- A robust algorithm for random parameter tracking
- A remark on the stability of the l.m.s. tracking algorithm
- Tracking in clutter with strongest neighbor measurements. I. Theoretical analysis
- Stochastic algorithms for visual tracking. Probabilistic modelling and stochastic algorithms for visual localsisation and tracking
- Performance analysis of general tracking algorithms
- On recursive estimation for time varying autoregressive processes
This page was built for publication: On a Perturbation Approach for the Analysis of Stochastic Tracking Algorithms
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4507483)