In this paper, we propose a 3-tier ultra wide-band indoor localization system for autonomously powered sensor network applications. It consists of a large number of cost-effective tags, a number of cheap and low-power hubs and few synchronized base stations. Using the UWB characteristics and the hierarchical scheme, the localization system enables ultra-low-power, autonomous tags and precise positioning. We present how ambiguous solutions (coordinates and transmission time of an unknown node) can be eliminated with the help of the proper geometry of 4 reference nodes. A method for optimizing the number and placement of hubs is proposed. Different positioning algorithms are discussed and compared based on the position accuracy a.f.o range errors. We determine the optimal algorithm for different scenarios. Simulations are conducted to investigate the performance degradation due to timing errors.
"However, in a lot of applications RFID-tags need to be localizable. This can be achieved by measuring the time of arrival of UWB-pulses . To be able to receive the UWB-pulses, they must be transmitted at an accurate rate . "
[Show abstract][Hide abstract] ABSTRACT: In RFID-tags with pulse-based UWB communication, accurate supply-independent low-power oscillators are required. The 24 MHz oscillator presented was realized in a 130 nm CMOS technology. It has an ultra-low supply voltage dependency of 104 ppm/V over a voltage range of 1.4 V to 0.4 V. This was achieved by the use of two nested ultra-low-power voltage regulators and a novel circuit technique based on the attraction of two oscillator frequencies. The mean power consumption is 33 muW over the 1 V voltage span. No external biasing and no trimming or calibration was used.
"Thus multiple hub placements can achieve the minimum number of hubs.  intuitively presents 2 different hub placements that both achieve this minimum. In some of these, however, hubs are more concentrated in the central area. "
[Show abstract][Hide abstract] ABSTRACT: We present a 3-tier UWB-based indoor localization system. It consists of a large number of energy-scavenging-based cost-effective transmit-only tags, a small number of battery powered hubs as relay stations and a few base stations. This hierarchical scheme is driven by the energy available at each node. Localization is based on the arrival time of the UWB pulses at reference nodes.We describe how the coordinates and transmit time of a tag are determined and how the ambiguous solution is eliminated with the proper geometry of 4 reference nodes. We formulate where to place the hubs as an optimization problem. The localization performance of the system is investigated as a function of several parameters such as non-ideal hub placement, hub localization error and TOA error.
" we proposed a 3-tier UWB-based indoor localization system, which consists of three levels: a large number of tags, a small number of hubs and a much smaller number of base stations. The differentiation in three levels not only allows distributing complexity over separate blocks (shifting the burden from tags to hubs and base stations), but also provides the necessary freedom to distribute energy consumption according to the energy source available at each level. "
[Show abstract][Hide abstract] ABSTRACT: Ultra-wideband (UWB) enabled wireless sensor networks (WSNs) are capable of identification, localization and recording of sensor readings. For reasons of cost, size and above all energy consumption constraints, scavenging-based ultra-lower-power (ULP) sensor nodes (SNs) are designed with only UWB transmitters. Transmit-only SNs are unaware of each other and broadcast data packets as soon as enough energy is harvested from the environment. The trade-off is that the resulting uncoordinated multiuser access causes packet collisions so that many SNs become undetectable. In this paper, we propose several techniques to control the transmit time of SNs. These techniques help to alleviate packet collisions and improve SNs detectability. They are evaluated and compared as a function of the network density (the number of SNs per m<sup>3</sup>). We determine the effective techniques for different network density. Results obtained from simulations indicate that by adding randomness in the transmissions while introducing pause time between two consecutive transmissions, all SNs can be identified up to quite a large network density (500 SNs/m<sup>3</sup>) in this scavenging-based transmit-only scenario.
Communication Systems, 2008. ICCS 2008. 11th IEEE Singapore International Conference on; 12/2008
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.