Shrinkage estimation-based source localization with minimum mean squared error criterion and minimum bias criterion

Robust Shrinkage Range Estimation Algorithms Based on Hampel and Skipped Filters

“…ML-Based Shrinkage Range Estimation Algorithm. In the LOS situations, the shrinkage estimator is obtained by multiplying the ML estimator (̂M L = 10 ( − )/10 = ⋅10 ) and shrinkage factor (c) [4,23], where…”

Robust TOA source localisation algorithm using prior location

“…Localisation of point targets is of considerable interest in various research fields including telecommunication, radar, sonar, and mobile communications. Position estimation problems under line‐of‐sight (LOS) environments have been intensively studied in the previous works [1–6]. However, some open problems exist and thus a crucial task among location estimation problems is to determine the location of the source in LOS/non‐LOS (NLOS) mixed situations [7, 8].…”

Sequential source localisation and range estimation based on shrinkage algorithm

“…In addition, a closed-form two-step weighted least squares (WLS) positioning algorithm that achieves the Cramér-Rao lower bound (CRLB) in sufficiently small noise conditions was devised in [8][9][10]. Recently, a closed-form shrinkage-based localisation method using the minimum mean square error (MSE) criterion was investigated in [11] in which the localisation performance of the two-step WLS method ( [8][9][10]) is improved in large noise variance conditions by modifying the blind shrinkage estimator [12] in such a way to be used in the TOAbased source localisation context. This paper proposes shrinkage algorithms using the CRLB-based, Stein's unbiased risk estimator (SURE) [13] and Ledoit-Wolf methods [14] extending the existing shrinkage method.…”