Topic
Recursive least squares filter
About: Recursive least squares filter is a research topic. Over the lifetime, 8907 publications have been published within this topic receiving 191933 citations.
Papers published on a yearly basis
Papers
More filters
TL;DR: This paper proposes an optimal recursive least square (RLS) parameter tracking algorithm, which significantly accelerates the convergence speed and eliminates the gradient noise.
Abstract: Digital signal processing (DSP) techniques have been proposed in recent years to adaptively track the control parameters of a power amplifier (PA) feedforward linearizer. In most of the propositions, gradient-based searching algorithms are applied to the parameter tracking. In this paper, we propose an optimal recursive least square (RLS) parameter tracking algorithm, which significantly accelerates the convergence speed and eliminates the gradient noise. There exist two problems for the RLS algorithm. First, the least square solution is not the optimal solution because of the nonlinearity of the PA. Second, the vector modulator (VM) which introduces the control parameters into the linearizer circuit may not be accurate enough to provide a precise power gain and phase shift calculated by the DSP. We solve both problems, respectively; by rearranging the circuit components and by constraining the VM characteristics. We also present simulation results to verify the performance improvement of the proposed algorithm.
55 citations
TL;DR: In this article, a fuzzy system based method for modeling both rate-independent and rate-dependent hysteresis in the piezoelectric actuator is proposed, where the antecedent structure of the fuzzy system is identified through uniform partition of its input variable.
55 citations
TL;DR: The auxiliary model-based recursive least-squares algorithm is used to estimate the parameters of one-step state-delay systems and the convergence of the proposed algorithm is studied by using the stochastic process theory.
Abstract: Based on the input–output representation of one-step state-delay systems, we use the auxiliary model-based recursive least-squares algorithm to estimate the parameters of the systems and study the convergence of the proposed algorithm by using the stochastic process theory. A simulation example is provided.
55 citations
TL;DR: In this article, a new adaptive recursive least squares (RLS) controller for HVAC systems is proposed, which can be described as a first order plus dead time model.
55 citations
TL;DR: Two modified RLS algorithms are derived by requiring robustness in its prediction performance to input perturbations to tackle the problem of diminishing weight decay effect as training progresses.
Abstract: Recursive least squares (RLS)-based algorithms are a class of fast online training algorithms for feedforward multilayered neural networks (FMNNs). Though the standard RLS algorithm has an implicit weight decay term in its energy function, the weight decay effect decreases linearly as the number of learning epochs increases, thus rendering a diminishing weight decay effect as training progresses. In this paper, we derive two modified RLS algorithms to tackle this problem. In the first algorithm, namely, the true weight decay RLS (TWDRLS) algorithm, we consider a modified energy function whereby the weight decay effect remains constant, irrespective of the number of learning epochs. The second version, the input perturbation RLS (IPRLS) algorithm, is derived by requiring robustness in its prediction performance to input perturbations. Simulation results show that both algorithms improve the generalization capability of the trained network.
55 citations