Schemes for Indoor Network Synchronization in Uwb Positioning

Location Positioning is a major technology for ubiquitous computing. A research on Location Positioning using UWB is ongoing. In order to construct an indoor location network, synchronization of base stations is very important. NTP is popularly used as a clock synchronization protocol ranging from LAN to WAN. The Master-Slave scheme, however, is the simplest method for synchronizing an indoor network that uses UWB. In this paper, we compare and analyze the NTP and Master-Slave schemes according to the statistical channel model for an indoor multi-path propagation environment. Error ranges are calculated at various circumstances such as when the indoor network expands from one primary base station to several base stations. In particular, we compare the correctness of the NTP and Master-Slave synchronization methods. We have found that NTP is a more reasonable synchronization protocol in UWB positioning. Introduction The term Location Positioning is widely used today to indicate the functions and procedures for determining the location of mobile terminals. Moreover, Location Positioning is a major technology for ubiquitous computing. Location Positioning methods can be classified according to their transmission media such as infrared rays, ultrasonic waves, radio frequency signal and UWB (Ultra Wide Band). There is an ongoing active research on Location Positioning using UWB and its standardization is ongoing in IEEE 802.15.4a TG. UWB technology can be incorporated into low-cost, high-performance devices and has a bandwidth greater than 25 percent of center frequency. It reduces multi-path fading and in-band interference while increasing the system's capacity. Multi-path resolution techniques in narrowband systems have been well developed [1]. To construct an indoor location network, synchronization of base stations is very important. Especially when the UWB is used as a medium, clock synchronization among base stations should have a unit precision of a nanosecond. A one-nanosecond difference corresponds to a 30-centimeter error range. The Network Time Protocol (NTP) is a clock synchronization protocol widely used in the Internet. It synchronizes the timeserver clocks distributed in the global network. Its coverage is from LAN to WAN. NTP consists of three functional blocks: Intersection, Clustering and Combining. The Intersection block is used to measure time differences between a local system clock and a number of timeserver clocks in the network. The Clustering block is used to select the best subset to reference among the timeservers. The Combining functional block is used to combine the differences of timeserver clocks and produce an accurate clock offset [3]. NTP is said to provide nominal accuracies of low tens of milliseconds on WANs, submilliseconds on LANs, and submicroseconds using a precision time source such as a cesium oscillator or GPS receiver [5]. Key Engineering Materials Vols. 277-279 (2005) pp. 233-241 online at http://www.scientific.net © (2005) Trans Tech Publications, Switzerland Online available since 2005/01/15 All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of the publisher: Trans Tech Publications Ltd, Switzerland, www.ttp.net. (ID: 130.203.154.169-30/01/08,08:44:47) Title of Publication (to be inserted by the publisher) Master-Slave is the simplest method for synchronizing an indoor network that uses UWB. The Slave node can synchronize its local time by point-to-point operation with the high-performance system Master. In this paper, we estimate error ranges for indoor network synchronization in UWB Location Positioning and then compare and analyze the NTP and Master-Slave schemes according to the statistical channel models for indoor multi-path propagation. Then we propose the possibilities of adopting the NTP and Master-Slave schemes for UWB indoor network synchronization. “Synchronization Methods” section describes the NTP and Master-Slave schemes. “Channel Model of UWB” section details the channel model for UWB indoor positioning. In the “Simulation and Results” section, conditions and main results are presented. And lastly, we will give a conclusion of this paper. Synchronization Methods Network Time Protocol. Network Time Protocol (NTP) synchronizes clocks of hosts and routers in the Internet. Primary base stations (stratum 1) synchronize to national time standards via radio, satellite and modem. Secondary base stations (more than stratum 2) are synchronized to primary servers via hierarchical subnets. Base stations operate in master/slave, symmetric or multicast modes with or without cryptographic authentication. Redundant servers and diverse network paths ensure reliability. Engineered algorithms reduce jitter, mitigate multiple sources and avoid improperly operating servers. The System clock is disciplined in time and frequencies using an adaptive algorithm responsive to network time jitter and clock oscillator frequency wander. The components of NTP are shown in Fig. 1. The roles of each block are given in [4-11]. Fig. 1. Components of Network Time Protocol Timestamps are exchanged between the client and each of possibly several other subnet peers at intervals ranging from a few seconds to several hours. Fig. 2 shows how NTP timestamps are numbered and exchanged between peers A and B [3, 4]. Fig. 2. Measuring Delay and Offset We can get the time-offset, average time difference and network delay information using four timestamps. The time-offset of clocki relative to clockj is the time difference between them, that is, Xij ≅ Ti(t) Tj(t) [3]. The average time difference from A to B is noted as θ. The sum of network delay On the Convergence of Bio-, Information-, Enrivonmental-, Energy-, Spaceand Nano-Technolgi s 234 Title of Publication (to be inserted by the publisher) differences from A to B and from B to A is called differential delay and noted as δ. The Eq. 1 and Eq. 2 are as follows [3]: θ = (T2-T1+T3-T4)/2 (1) δ = T2 T1 + T4 – T3 (2) The maximum error that can be made under any condition is called the synchronization distance, λ. The error expected under nominal conditions is called the dispersion, ε [3]. Given ε, λ is calculated with Eq. 3.