Dynamic orientation of receiver arrays using particle swarm optimisation


ÇAKIR O., KAYA İ., YAZGAN A., ÇAKIR O.

ELECTRONICS LETTERS, cilt.49, sa.21, ss.1313-1315, 2013 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 49 Sayı: 21
  • Basım Tarihi: 2013
  • Doi Numarası: 10.1049/el.2013.2165
  • Dergi Adı: ELECTRONICS LETTERS
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.1313-1315
  • Anahtar Kelimeler: array signal processing, particle swarm optimisation, time-of-arrival estimation, dynamic orientation, particle swarm optimisation, time difference of arrival localisation system, transmitteraEuro"receiver geometry, time differences estimation error, TDOA estimation, emitter location findings, Cramer-Rao lower bound, optimum receiver arrays, Fisher information matrix, PSO
  • Karadeniz Teknik Üniversitesi Adresli: Evet

Özet

Considering a time difference of arrival (TDOA) localisation system, it is well known that the accuracy depends on the transmitter-receiver geometry, time differences estimation error and the selection of the localisation algorithm. Although many studies related to TDOA estimation and emitter location finding can be found in the literature, the effect of transmitter-receiver geometry on positioning accuracy has not been investigated sufficiently. Most of the studies focused on figuring out the optimum receiver arrays by minimising the Cramer-Rao lower bound (CRLB) or maximising the Fisher information matrix. In this reported work, depending on the target position, the optimum receiver arrays are dynamically oriented. By using particle swarm optimisation (PSO) the target is localised and the offset angle which makes the CRLB minimum is determined. After rotating the receiver array by the offset angle, the target is localised again. At the end of the process, a significant decrement in the positioning error is obtained. To show the effectiveness of the proposed method, a uniform angular array and a cross array are selected as optimum two-dimensional geometries.