Cooperative angle-of-arrival position localization Jun Xu, Maode Ma ⇑ , Choi Look Law School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore article info Article history: Received 31 October 2013 Received in revised form 8 September 2014 Accepted 11 September 2014 Available online 22 September 2014 Keywords: AOA UWB Cooperative localization abstract Cooperation among unknown-location nodes can improve network positioning coverage as well as localization accuracy particularly in the poor electronic conditions. This paper investigates the cooperative angle-of-arrival (AOA) positioning based on the Ultra Wide- band (UWB) technique. An UWB measuring system is built to estimate the AOAs, and then the weighted cooperative positioning algorithm is implemented to determine the locations of the unknown nodes. Theoretical analysis is conducted to evaluate the accuracy of the cooperative AOA localization, and the methods to select neighbors for the distributed local- ization are proposed as well to improve the localization performance. The results of exten- sive experiments have shown that the proposed cooperative AOA positioning approach is more accurate. Ó 2014 Elsevier Ltd. All rights reserved. 1. Introduction Accurate location information is critical for many appli- cations and UWB is a promising candidate technique for the accurate and robust indoor positioning because of its fine temporal resolution and ability to resolve the multi- path components [1–6]. The time-of-arrival (TOA), the time-difference-of-arrival (TDOA) and the AOA are the three commonly used ranging techniques. For TOA and TDOA ranging, time synchronization in the order of picoseconds is required between the transmitter and the receiver or among the different receivers. To avoid the requirement of the synchronization between the transmit- ter and the receiver, AOA technique can be employed to measure the angles between the blind node and the refer- ence node. Many AOA estimation techniques based on the UWB technology have been proposed. Most of the techniques work based on the maximum likelihood (ML) estimation approach or a scheme in the family of subspace algorithms such as the multiple signal classification (MUSIC) scheme and the estimation of signal parameters via rotational invariance technique (ESPRIT). An UWB propagation study has been conducted in [7], where AOA has been estimated by using a modified algorithm. The study has demonstrated that the signals received in the UWB wireless system have a large angle spread and the multipath components are delivered in clusters. A technique of joint TOA and AOA estimation on the UWB signals has been proposed in [8]. It uses the signals trans- formed between the time and the frequency domains and is able to resolve the TOA and AOA precisely. Zhang et al. [9] has investigated to estimate the indoor multipath prop- agation of AOA using UWB signals. The unitary ESPRIT algorithm has been employed to estimate the AOA includ- ing both the azimuth and elevation angles. The frequency dependence of AOA has also been investigated over the UWB frequency band. In this paper, a new AOA estimation system has been introduced. The key principle of the sys- tem design is to employ the advantages of TDOA and AOA techniques to avoid the introduction of extra system components or a structure that would degrade the fine time resolution of the UWB signals because it is critical to the accuracy of the position localization. In general, the TDOA ranges can be measured to determine the posi- tions of the blind nodes. The TDOA measurements can be http://dx.doi.org/10.1016/j.measurement.2014.09.023 0263-2241/Ó 2014 Elsevier Ltd. All rights reserved. ⇑ Corresponding author. Measurement 59 (2015) 302–313 Contents lists available at ScienceDirect Measurement journal homepage: www.elsevier.com/locate/measurement