Subspace estimation plays an important role in a variety of modern signal processing applications. We present new approach for tracking the subspace recursively. It is based on novel interpretation as solution projection like unconstrained minimization problem. show that recursive least squares techniques can be applied to solve this problem by making appropriate approximation. The resulting algorithms have computational complexity O(nr) where n input vector dimension and r number desired...

This work presents a unified derivation of four rotation-based recursive least squares (RLS) algorithms. They solve the adaptive problems linear combiner, combiner without desired signal, single channel, and multichannel prediction transversal filtering. Compared to other approaches, authors' is simpler unified, may be useful readers for better understanding algorithms their relationships. Moreover, it enables improvements some in literature both computational numerical issues. All derived...

The author presents a new algorithm for tracking the signal subspace recursively. It is based on novel interpretation of as solution projection like unconstrained minimization task. shown that recursive least squares technique can be applied to solve this problem by approximation projections appropriately. resulting has computational complexity O(nr) where n dimension and r number desired eigencomponents, respectively. Simulation results show frequency capability virtually identical in some...

In this letter, we propose a Global navigation satellite system (GNSS) spoofing suppression method based on correlation peak cancellation (SCPC). The approach estimates the signal from baseband sampling sequence, and generates reverse sequence to suppress GNSS attack. Based this, receiver scheme with SCPC function is proposed by adding module general model. experimental verification conducted TEXBAT dataset. By comparing peaks, carrier-to-noise ratio trajectories of spoofed, mitigated...

In order to determine an appropriate attitude of three-axis stabilized communication satellites, this paper describes a novel determination method using direction arrival (DOA) estimation ground signal source. It differs from optical measurement, magnetic field inertial and global positioning system (GPS) determination. The proposed is characterized by taking the source as reference acquiring information DOA estimation. Firstly, measurement equation with derived in detail. Then, error...

Abstract Online novelty detection is of great importance in the series production motors. This study developed an online scheme for motors based on a one-class hyperdisk (OCHD) model. In OCHD approach, decision boundary estimated using (HD), which derived from training sample set. The HD model addresses underestimation issue commonly associated with convex-hull-based methods by providing more accurate estimation class region. Furthermore, optimal separating hyperplane constructed at nearest...

To solve the persistently multiple object tracking in cluttered environments, this paper presents a novel association approach based on shortest path faster algorithm. First, is formulated as an integer programming problem of flow network. Then we relax to standard linear problem. Therefore, global optimum can be quickly obtained using The proposed method avoids difficulties programming, and it has lower worst-case complexity than competing methods but better robustness accuracy complex...

Abstract This article proposes a direct position determination (DPD) algorithm for non‐circular sources observed by moving array using the self‐calibration technique in presence of mutual coupling. The method first utilises symmetric Toeplitz property uniform linear matrices with coupling and cyclic circular matrix, realising decoupled estimations target parameters sensor error parameters. Then multiple are directly determined based on subspace data fusion criterion manner, where subspaces...

We develop a rank tracking method for spherical subspace trackers. The adaptively sets threshold based on the averaged noise eigenvalue. signal eigenvalues are then compared with to reach decision rank. itself is chosen balance error probabilities due underfitting and overfitting. Simulation results show that our performs as well (and, very low SNRs, even better than) information theoretic criteria.

We prove the asymptotic convergence of subspace tracking (stochastic approximation) algorithms PAST and PASTd. also present new results about their rate. First we review algorithms. Then derive corresponding ordinary differential equation (ODE). The behaviour established by studying asymptotically stable equilibrium states ODE, computer simulation are shown.

A new, efficient two plane rotations (TPR) method for computing two-sided involved in singular value decomposition (SVD) is presented. By exploiting the commutative properties of some special types 2x2 matrices, we show that a rotation can be computed by only and few additions. Moreover, if use coordinate digital computer (CORDIC) processors to implement processing elements (PEs) SVD array given Brent Luk, computational overhead diagonal PEs due angle calculations avoided. The resulting has...

Vehicle pose estimation plays an significant role in vehicle tracking which is important part autonomous driving. In this paper, a new method based on Edge Distance proposed. The defined to describe the distribution of point clouds with respect their bounding rectangles. Based Distance, rectangle utilized fit outline. For case where are sparse or only one edge visible, modified according preset size. obtained rectangle. Experiments KITTI data sets show that proposed can obtain better...

A new algorithm for signal subspace tracking is presented. It based on an approximated singular value decomposition using interlaced QR-updating and Jacobi plane rotations. By forcing the noise to be spherical, computational complexity of brought down O(nr), where n problem dimension r desired number components. The lends itself a very efficient systolic array implementation, resulting in throughput O(n/sup 0/). Simulations show that frequency capabilities method are at least as good those...

The article comments on the error propagation analysis of recursive least squares algorithms presented by Ljung and (1985), Verhaegen (1989), Haykin (1991). In particular, it corrects some incorrect statements about given in literature. It considers problem spatial adaptive filtering using RLS minimization.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

The authors present a general adaptive array processing algorithm based on the MVDR (minimum variance distortionless response) beamformer for multiple steering vectors and frequencies. They show how they use sequence of Givens rotations to update output, implement by means two linear systolic arrays. pipelined CORDIC processor is proposed as element achieve high data throughput straightforward implementation. problem complex arithmetic frequency-domain discussed.< <ETX...

Traditional multi-target tracking algorithms assume that each target can generate at most one detection per scan. However, a may produce multiple detections (MDs) in many practical applications, e.g. over-the-horizon radar (OTHR), for extended and with sensors. In this study, the authors propose new algorithm targets MD observation. The proposed technique is based on labelled random finite set (RFS), which estimates number of trajectories their states. Furthermore, they two methods,...

Multi-target tracking is an extremely challenging task when targets move in the formation of groups and interact with each other. Group target has to deal this problem contrast independently moving as assumed most multi-target algorithms. A feasible approach for group estimate structure modify motion model prediction step tracker according structure. In paper, we propose ad hoc labeled multi-Bernoulli (LMB) filter interaction, which use stochastic differential equation joint by using graph...

Rank and subspace estimation is important in a variety of modern signal processing applications. In this paper we present new approach for tracking both the rank recursively. At arrival each sample, update eigenvectors spanning plus one or fixed number auxiliary eigenvectors, their corresponding eigenvalues, an averaged noise eigenvalue. Then apply information theoretic criteria to estimate signals. The resulting adaptive algorithm has computational complexity which linearly proportional...