WIRELESS SENSOR POSITIONING WITH ULTRAWIDEBAND FINGERPRINTING

Similar documents
A Cooperative MIMO Mobile Multihop Relay for Cellular Networks

Wireless Channels Path Loss and Shadowing

IEEE pc-00/44

Bending Geometry Factor For Profile Corrected Involute Gear Tooth With Trochoidal Fillet

Journal of Kerbala University, Vol. 7 No.2 Scientific. 2009

A Complexity Cost Function for the Signal Processing in a WCDMA. Basestation for dimensioning of a Software Defined Radio.

Compatibility Analysis for Wireless Systems in VHF/UHF Bands with Geographic Information

I. SUMMARY II. NETWORK LEVEL ANALYSIS

Performance Analysis of MIMO Satellite Communications Via Multiple Terrestrial Non-Regenerative Relay Nodes

Computer Graphic with Matrices Math 308A Project

A Study on Interference Analysis based on Rec. ITU-R P.1546 with Geographic Information

Development of On-Board Orbit Determination System for Low Earth Orbit (LEO) Satellite Using Global Navigation Satellite System (GNSS) Receiver

Low-Complexity Time-Domain SNR Estimation for OFDM Systems

ELG3175 Introduction to Communication Systems. VSB and Introduction to Angle Modulation

Design of compact joint transform correlator

Long PN Code Based Traceback in Wireless Networks

Long PN Code Based Traceback in Wireless Networks

White Rose Research Online URL for this paper: Version: Accepted Version

Interconnect Planning, Synthesis, and Layout for Performance, Signal Reliability and Cost Optimization. Project Overview

Segmentation of Fluorescence Microscopy Cell Images Using Unsupervised Mining

Parameters of spinning AM reticles

PERFORMANCE OF TOA ESTIMATION TECHNIQUES IN INDOOR MULTIPATH CHANNELS

100G and 200G single carrier transmission over 2880 and 320 km using an InP IQ modulator and Stokes vector receiver

Assessing the Effect of Calibration on Nonresponse Bias in the 2005 ARMS Phase III Sample Using 2002 Census of Agriculture Data

A Comparison of Modulation Schemes in Bandlimited AWGN Channels

Compensation Estimation Method for Fast Fading MIMO- OFDM Channels Based on Compressed Sensing

Antenna fundamentals: With answers to questions and problems (See also Chapter 9 in the textbook.)

OBSOLETE 427, 424, 425, 426, 432, 428, 429, 422, AD530, AD531. I t. t I. t t I r

VLSI Implementation of Low Complexity MIMO Detection Algorithms

Design of A Circularly Polarized E-shaped Patch Antenna with Enhanced Bandwidth for 2.4 GHz WLAN Applications

ECE 6560 Multirate Signal Processing Chapter 7

Performance Comparison of Multihop Wireless ADHOC Routing Protocols AODV, OLSR & TORA

ISSN: [Reddy & Rao* et al., 5(12): December, 2016] Impact Factor: 4.116

where and are polynomials with real coefficients and of degrees m and n, respectively. Assume that and have no zero on axis.

Comparison Between Known Propagation Models Using Least Squares Tuning Algorithm on 5.8 GHz in Amazon Region Cities

WIRELESS SENSORS EMBEDDED IN CONCRETE

DESIGN A FSS USING JERUSALEM CROSS STRUCTURE AS A BSF FOR SATELLITE APPLICATIONS AT 10GHz

Phase Locked Loop based Pulse Density Modulation Scheme for the Power Control of Induction Heating Applications

TELE4652 Mobile and Satellite Communications

ABSTRACTT FFT FFT-' Proc. of SPIE Vol U-1

N2-1. The Voltage Source. V = ε ri. The Current Source

Optimal Design of Smart Mobile Terminal Antennas for Wireless Communication and Computing Systems

Spectrum Sharing between Public Safety and Commercial Users in 4G-LTE

S11 PHY114 Problem Set 8

Proceedings of Meetings on Acoustics

PSO driven RBFNN for design of equilateral triangular microstrip patch antenna

DESIGN AND PARAMETRIC EVALUATION OF RECTANGULAR MICROSTRIP PATCH ANTENNA FOR GSM APPLICATION

ONE-WAY RADAR EQUATION / RF PROPAGATION

Chamber Influence Estimation for Radiated Emission Testing in the Frequency Range of 1 GHz to 18 GHz

Chapter 14: Bandpass Digital Transmission. A. Bruce Carlson Paul B. Crilly 2010 The McGraw-Hill Companies

Wireless Communication (Subject Code: 7EC3)

MRI. Chapter 1. Larmor equation. Net magnetization. Rotating frame of reference 1/31/2017. ω0 (rad/s) = γ (rad/s/tesla) x B0 (Tesla)

Noise Reduction/Mode Isolation with Adaptive Down Conversion (ADC)

Received September 9, 2012; revised October 15, 2012; accepted October 26, 2012

Graphs fundamental to many problems. Web graphs. Biology. Other.

Feasibility of a triple mode, low SAR material coated antenna for mobile handsets

School of Electrical and Computer Engineering, Cornell University. ECE 303: Electromagnetic Fields and Waves. Fall 2007

Realistic Simulation of a Wireless Signal Propagation in an Urban Environment

Near-field Computation and. Uncertainty Estimation using Basic. Cylindrical-Spherical Formulae

Performance Analysis of High-Rate Full-Diversity Space Time Frequency/Space Frequency Codes for Multiuser MIMO-OFDM

This article presents the

Channel Modelling ETIM10. Fading Statistical description of the wireless channel

Current Compensation Techniques for Lowvoltage High-performance Current Mirror Circuits

Low Profile MIMO Diversity Antenna with Multiple Feed

Experimental Investigation of Influence on Non-destructive Testing by Form of Eddy Current Sensor Probe

Figure Geometry for Computing the Antenna Parameters.

Cyclic Constellation Mapping Method for PAPR Reduction in OFDM system

Analysis of a Fractal Microstrip Patch Antenna

EE 330 Lecture 24. Amplification with Transistor Circuits Small Signal Modelling

Experimental Analysis of Parameter Limitations in High-Frequency Resonant Gate Driver

ScienceDirect. Simplified Generation of Electromagnetic Field within EMC Immunity Test Area

Study and Design of Dual Frequency High Gain and Conformal Microstrip Patch Antenna

A Transmission Scheme for Continuous ARQ Protocols over Underwater Acoustic Channels

ANALYSIS OF CIRCULAR MICROSTRIP ANTENNA ON THICK SUBSTRATE

When answering the following 25 questions, always remember that there is someone who has to grade them. So please use legible handwriting.

Published in: International Journal of Material Forming. Document Version: Peer reviewed version

1 Performance and Cost

Lecture 4. EITN Chapter 12, 13 Modulation and diversity. Antenna noise is usually given as a noise temperature!

Wall Compensation for Ultra Wideband Applications

Feasibility of Air Target Detection Using GPS as a Bistatic Radar

P. Bruschi: Project guidelines PSM Project guidelines.

Hexagonal Shaped Microstrip Patch Antenna for Satellite and Military Applications

The Periodic Ambiguity Function Its Validity and Value

Demosaicking using Adaptive Bilateral Filters

How to Shorten First Order Unit Testing Time. Piotr Mróz 1

Design of Microstrip Antenna for Wireless Local Area Network

Multiagent Reinforcement Learning Dynamic Spectrum Access in Cognitive Radios

Analysis of the optimized low-nonlinearity lateral effect sensing detector

DESIGN OF DUAL BAND 1X2 MICROSTRIP PATCH ANTENNA ARRAY FOR GPS RECEIVERS

Gas Tube Arresters. Certifications, Device Selection Purpose, Operation, Installation Part Number Construction, Part Marking. General Information

Memorandum on Impulse Winding Tester

A 10 MHz GaNFET Based Isolated High Step-Down DC-DC Converter: Design and Magnetics Investigation

OTA MEASUREMENTS OF SMALL RADIO TERMINALS BY A SPHEROIDAL COUPLER

Electrical characterization of thin film ferroelectric capacitors

A New Method of VHF Antenna Gain Measurement Based on the Two-ray Interference Loss

Localization Algorithm for a mobile robot using igs

Solution of ECE 342 Test 2 S12

Motion-blurred star image acquisition and restoration method based on the separable kernel Honglin Yuana, Fan Lib and Tao Yuc

MATLAB/SIMULINK TECHNOLOGY OF THE SYGNAL MODULATION

Transmit Beamforming with Reduced Feedback Information in OFDM Based Wireless Systems

Transcription:

WIRELESS SENSOR POSITIONING WITH ULTRAWIDEBAND FINGERPRINTING Wasim Q. Malik and Ben Allen Depamen of Engineeing Science, Univesiy of Oxfod, Paks Road, Oxfod OX 3PJ, Unied Kingdom Email: wasim.malik@eng.ox.ac.uk; ben.allen@eng.ox.ac.uk (Invied Pape) Keywods: Localizaion, posiioning, RF fingepining, ulawideband (UWB). Absac Ulawideband (UWB) sysems offe high spaioempoal esoluion and ae heefoe well suied o senso localizaion applicaions. In his pape, we conside he use of UWB signals fo posiioning and anging based on fingepining using a channel impulse esponse daabase. We use indoo measuemens o demonsae he educion in he locaion esimaion ambiguiy and false alam pobabiliy wih an incease in he channel bandwidh. Inoducion Accuae posiioning is one of he majo equiemens fo he nex geneaion of wieless communicaions sysems and senso newoks. Recenly enaced egulaions in he US and Euope now equie cellula sevice povides o deemine and epo he locaion of a mobile ansmie in ode o deal wih emegency siuaions [, 2]. Thee is also consideable inees in senso localizaion by he miliay, since i can be useful fo asse and pesonnel acking, and cove communicaions. Finally, locaion-based billing, mobile yellow pages and inelligen anspo sysems ae some of he majo commecial applicaions. Posiion esimaion wih he saellie-based Global Posiioning Sysem (GPS) is highly accuae in many siuaions, bu suffes sevee degadaion when he eceive is in an uban canyon o an indoo envionmen. Gound-based saions use he eceived signal sengh (RSS), ime-of-aival (TOA), ime-diffeence-of-aival (TDOA) o angle-of-aival (AOA) fo locaion esimaion ha usually involves iangulaion [3]. These appoaches ae highly sensiive o he availabiliy of line-of-sigh (LOS), wih posiively biased ange esimaes in non-los (NLOS) siuaions. Mulipah popagaion also causes goss posiioning eos, due o which sysems wih lage bandwidh, and consequenly fine mulipah esoluion capabiliy, yield bee pefomance [4, 5]. Fuhemoe, he geomeic diluion of pecision (GDOP) of many posiioning echniques is small only when he posiioning saions ae muually fa apa, inceasing significanly if hey ae spaially poximal. RF fingepining is an alenaive echnique fo locaion esimaion [3, 6]. I consiss of wo seps: () a pioi consucion of a map elaing vaious poins in space o a given popagaion paamee, and (2) un-ime esimaion of ha paamee fom he eceived signal and is coelaion wih he soed daabase. This signal paamee ha acs as a unique locaion signaue may, fo example, be he RSS o he angula delay pofile (ADP). The channel impulse esponse (CIR) may also be used fo his pupose [7, 8], leading o geae accuacy han ime-based appoaches in uban bu no sububan envionmens [9]. Compaed o GSM sysems, he accuacy of RF fingepining impoves wih UMTS due o is inceased bandwidh and ime esoluion [9]. Fuhe impovemens can be obained wih adapive appoaches such as neual newok aining [] o Kalman fileing [] o incease he obusness. Fig. illusaes he concep of senso posiion esimaion wih fingepining and daabase coelaion based on he spaio-empoal impulse esponse of he popagaion channel. CIR-based fingepining offes a poenially obus soluion o some of he poblems faced by he ohe posiioning echniques. LOS blockage and dense mulipah do no deeioae is accuacy, which acually inceases wih geae scaeing. On his basis, his echnique is especially suied o indoo envionmens wih dense, hee-dimensional mulipah popagaion. As i does no involve iangulaion o AOA esimaion, he equiemen fo disibued o collocaed anenna sysems is also obviaed. I pefoms bee in wideband channels han in naowband channels [9], and heefoe may be a good choice fo senso localizaion using ulawideband (UWB) echnology [2, 3]. On he flip side, howeve, he consucion of a high-esoluion map based on measuemens is a vey ime-consuming pocess and is feasible only in small egions wih easy access. Also, he map is only valid fo a adio popagaion envionmen whee he main feaues of he CIR, dependen on he dominan scaeing cenes, emain saionay, a condiion ha can be fulfilled by lage buildings in oudoo envionmens, o walls and lage objecs in indoo channels.

h (, Tansmie x() y (, Receive Coelao Esimaed posiion, ˆ Channel Daabase Fig.. Schemaic epesenaion of locaion fingepining based on daabase coelaion. In his pape, we invesigae he use of UWB signals fo senso locaion fingepining in an indoo envionmen using CIR measuemens. Of paicula inees is he impac of LOS pesence and channel bandwidh, which we conside in deail in ou analysis. The es of his pape is oganized as follows. Sec. 2 gives an oveview of he pinciple of wideband RF fingepining, Sec. 3. descibes he indoo channel esponse measuemen pocedue used in he analysis in his pape, Sec. 4 discusses he key esuls fom his analysis, and Sec. 5 summaizes he main conclusions. 2 RF Fingepining We now pesen an analyical famewok fo locaion esimaion using fingepining based on wideband adio signals. Le us conside a wideband wieless communicaions sysem opeaing in a fequency selecive channel. The eceived signal is given in he geneal fom by he expession y, = E h, x + z, () whee ( ) x ( ) () () is he eceive locaion, () E x z() h(, x is he ansmied signal, is he ansmi signal enegy, is he addiive whie Gaussian noise, is he spaio-empoal impulse esponse of he channel, and denoes convoluion. Unde he apped delay line channel model, we can wie L jφ, l h(, = α, le δ ( l) l= whee α,l, φ,l and,l denoe he magniude esponse, phase esponse and excess ime delay of he l h mulipah componen (MPC), and L is he numbe of esolved MPCs in he CIR coesponding o locaion. The pinciple of RF fingepining involves he measuemen and soage of h, fo R, whee R denoes he ( ) ensemble of locaions in he egion of inees. While fingepining can be used fo volumeic localizaion, we will esic ou analysis o wo spaial dimensions in his pape, so,, (2) ha ( x, y) = epesens a poin in he hoizonal plane. When a pe-ecoded CIR daabase is available, an esimae, ˆ, of he ue insananeous posiion,, of a mobile can be obained using he coesponding insananeous esimaed CIR, h ˆ,. Hee we do no assume pefec channel ( ) esimaion and allow fo impefecions due o noise, unesolved mulipah, inefeence, o synchonizaion eos. The posiion esimaion poblem can hen be fomulaed as he maximizaion of he CIR coss-coelaion coefficien, i.e., = R R ˆ ag max. Fo m = h(, and n h( ˆ, coefficien,, (3) =, he complex coelaion R E{ } E{ } E{ }, is he channel spaial coelaion defined as mn mn m n R = 2 2 2 2 ( E{ m } E{ m} )( E{ n } E{} n ) whee u is he complex conjugae of u and E {}. denoes he expecaion opeao. Fom (4), R = coesponds o ˆ =, leading o pecise localizaion. The pefec coelaion condiion is, howeve, impacical due o impefecions in he channel esimaion pocess a he eceive. A lowe coelaion heshold,, can heefoe be used insead, such as R h =.9,.7, o.5. The coesponding egion in space is epesened by R ˆ R such ha R ˆ if R R fo aound ˆ h, and he span of ˆR defines he coheence disance, dc, of he channel. Some sidelobes wih vaying ampliudes may be pesen in he spaial coelaion funcion, as is ypical fo naowband Rayleigh channels [4]. If he ampliude of any such sidelobe exceeds, i may esul in sevee locaion esimaion R h eos, and heefoe R h should be chosen caefully in accodance wih he channel popagaion condiions., (4) R h

Analogous o his ime-domain eamen, he fequencydomain channel ansfe funcions (CTFs) can also be used fo fingepining. If U = F u denoes he discee Fouie {} ansfom opeaion, hen we can ewie (4) as R = F { MN } (5) whee M = H, f = F h, ( ) ( ) { ( )} K jθ, k a, ke δ f k f k = = ( ) and N = H ˆ, f is defined similaly. In (6), α,k and θ,k denoe he ampliude and phase of he k h fequency componen. Fo a channel wih bandwidh W sampled a K f = W K. fequency poins, he fequency esoluion is ( ) 3 Analysis Mehodology We conduc indoo channel measuemens in a numbe of small office envionmens o evaluae he posiioning abiliy of UWB fingepining. The measuemens ae in he FCCallocaed UWB band beween fequencies f l = 3. GHz and f h =.6 GHz, wih K = W f + = 6 discee fequency samples ove he = ( f ) 2= h l (6) W f GHz bandwidh. A veco newok analyze is used o obain he CTF, H (, f ), fom he S 2 (f) measuemen. LOS and NLOS channel daases ae obained in muliple ooms, each compising of he CTFs measued in a local egion using a ecangula gid posiione spanning m m wih. m esoluion. Fo each daase, he ansmie is fixed a one locaion while he eceive is anslaed hoizonally using he posiione. Discone anennas ae used fo omnidiecional ansmission H f, we obain he and ecepion. Fom each measued (, ) coesponding educed-bandwidh CTFs, H (, f b ), wih bandwidh W b, using ideal bandpass files cened a f c, such H, f = H, f if f f W 2, and ohewise, ha ( ) ( ) b whee f ( f f ) 2 6.85 GHz c l h c = + = is mainained in each case. The numbe of fequency poins, K, is kep consan fo fai compaison by zeo-padding he disceized CTFs in he ineval Wb 2 < f fc W b, simila o [5]. Fuhe deails of he channel measuemen pocedue, he popagaion envionmen, and basic daa pocessing can be found in [5]. 4 Resuls The measuemen-based analysis of fingepining wih UWB signals demonsaes he effeciveness of his echnique. We find fom he complex spaial coelaion magniude, R, in Fig. 2 ha in he GHz wide UWB channel, hee is lile pobabiliy of false posiion esimaion if he coelaion heshold is chosen caefully. Fom ou analysis, a small numbe of false maches may occu a R h =.5 fo some CIRs due o he coelaion sidelobes, bu R h =.7 povides b.75.5. 5.75.5. 5 5 5 moe eliable esuls. Significan sidelobes, which decay wih disance, ae obseved only along he coss-ange diecion, indicaing ha he popagaion is no pefecly isoopic. The compaison of Figs. 2(a) and (b) eveals ha he posiioning accuacy is highe and he incidence of false a alam lowe in NLOS han in LOS, due o he moe ponounced sidelobes in he lae, esablishing ha he pefomance of UWB fingepining impoves unde inceased scaeing. The posiion ambiguiy egion has a adius of appoximaely 2 cm wih R h =.5. This accuacy is compaable o ha obained fom TOA-based UWB posiioning wih iangulaion in [5]. (a) Line-of-sigh envionmen 5 5 (b) Non-line-of-sigh envionmen Fig. 2. An esimae of fingepin ambiguiy in ems of he channel impulse esponse spaial coelaion fo a GHz wide UWB channel. The ue senso locaion is = (,). Fig. 3 depics he coelaion magniudes in he NLOS channel fo vaious values of he channel bandwidh, W b. I is found ha he localizaion accuacy depends significanly on W b. As W b inceases, he obusness of he coelaion measue also inceases, wih fewe false esimaes appeaing 5 5

above he appopiaely seleced coelaion heshold a vey lage bandwidh. We also find ha a low W b, he coelaion heshold mus be inceased o a value appoaching uniy in ode o educe he occuence of false posiion esimaes. A consequence of his obsevaion is ha he condiions on channel esimaion qualiy can be elaxed when he bandwidh is lage, leading o educed equiemens fo he associaed hadwae complexiy. The widh of he coelaion funcion mainlobe does no appea o vay subsanially wih W b, and he main conibuion of inceasing bandwidh is hus he impoved disciminaion fom disan locaions. We now invesigae he dependence of coelaion on channel bandwidh in fuhe deail by examining he spaial coelaion funcion a vaious values of W b. The mean coelaion behavio is analyzed along he ange and cossange diecions sepaaely. In his analysis, he coss-ange coelaion, R x, is obained as a funcion of he coss-ange offse, x, fo a given se of fixed-ange CIRs using (4) wih m = h( x, and n = h( ˆ x,, whee ˆ x denoes he esimae of he ue posiion. We hen aveage his coelaion funcion ove he ange, y, o obain R { x y Rx = E }. A simila pocedue is adoped o evaluae of he mean ange coelaion, R, and he pocess is epeaed fo vaious W b. y Fig. 4 shows ha boh he mean ange and coss-ange coelaion sidelobes decay shaply wih W b. The coelaion decay is lage in ange han in coss-ange. The coss-ange sidelobes obseved in Fig. 3 ae mos noiceable a low W b. The mean widh of he coelaion funcion mainlobe, which deemines he spaial esoluion of he senso posiion esimae, is also found o be smalle in he ange diecion han in he coss-ange diecion. When he coelaion heshold is aken as.5, his posiioning ambiguiy egion spans appoximaely d = 4 cm and d x = 6 cm along he ange and coss-ange diecions, especively, a he UWB bandwidh (W b 5 MHz), so ha he aea of he ambiguiy ellipse is π dd 4= 8.8 cm 2. x.9.8.7.6.5 5.9.8.7.6.5 5 5 5 (a) W b = 5 MHz 5 5 (b) W b = 2 GHz 5 5 5 Conclusion Senso posiion esimaion wih fingepining povides a viable alenaive o ohe posiioning echniques based on TOA, TDOA and AOA esimaion, poviding a suiably deailed signaue daabase is available. Is accuacy, achieved wih a single anenna, is compaable o he ohe candidae echniques. I is also obus o hash popagaion condiions such as NLOS channels wih dense mulipah. Ou analysis has shown ha he posiioning accuacy and eliabiliy obained wih CIR-based fingepining is impoved subsanially wih an incease in he signal bandwidh. As accuae posiion esimaion can hen be achieved even wih elaively low coelaion, he UWB channel esimaion equiemens can be elaxed and sysem complexiy educed. In conclusion, UWB fingepining is well suied o senso localizaion applicaions in saic indoo envionmens..9.8.7.6.5 5 5 5 (c) W b = GHz Fig. 3. Fingepin ambiguiy in ems of he channel impulse esponse spaial coelaion fo a UWB channel wih bandwidh W b in NLOS. The ue senso locaion is = (,). 5

Acknowledgemen This wok was suppoed by he Engineeing and Physical Sciences Reseach Council, UK, via Gan GR/T2769/. Refeences [] "Revision of he Commission's ules o insue compaibiliy wih enhanced 9 emegency calling sysems," Fedeal Communicaions Commission, Washingon, DC, USA FCC 94-2, July 996. [2] "Commission pushes fo apid deploymen of locaion enhanced 2 emegency sevices," Euopean Commission, Bussels, Belgium IP/3/22, July 23. [3] F. Gusafsson and F. Gunnasson, "Mobile posiioning using wieless newoks," IEEE Sig. Poc. Mag., vol. 22, no. 4, July. [4] S. Gezici, e al., "Localizaion via ula-wideband adios: a look a posiioning aspecs fo fuue senso newoks," IEEE Sig. Poc. Mag., vol. 22, no. 4, July. [5] Z. Taique, W. Q. Malik, and D. J. Edwads, "Bandwidh equiemens fo accuae deecion of diec pah in mulipah envionmen," IEE Elecon. Le., vol. 42, no. 2, Jan. 26. [6] S. Juuakko and W. Backman, "Daabase coelaion mehod wih eo coecion fo emegency locaion," Wieless Pes. Commun., vol. 3, no. 2, Sep. 24. [7] O. Hilsenah and M. Wax, "Radio ansmie locaion finding fo wieless communicaion newok sevices and managemen," in US Paen 6,26,34. USA: US Wieless Cop., 2. [8] M. Wax and O. Hilsenah, "Signaue maching fo locaion deeminaion in wieless communicaion sysems," in US Paen 6,2,95. USA: US Wieless Cop., 2. [9] S. Ahonen and H. Laiinen, "Daabase coelaion mehod fo UMTS locaion," in Poc. IEEE Veh. Technol. Conf. (VTC). Seoul, Souh Koea, Ap. 23. [] C. Neguizian, C. Despins, and S. Affes, "Geolocaion in mines wih an impulse esponse fingepining echnique and neual newoks," IEEE Tans. Wieless Commun., vol. 5, no. 3, Ma. 26. [] T. Nypan, K. Gade, and O. Hallingsad, "Vehicle posiioning by daabase compaison wusing he Box-Cox meic and Kalman fileing," in Poc. IEEE Veh. Technol. Conf. (VTC). Bimingham, AL, USA, May 22. [2] B. Allen, M. Dohle, E. E. Okon, W. Q. Malik, A. K. Bown, and D. J. Edwads (ed.), Ula-Wideband Anennas and Popagaion fo Communicaions, Rada and Imaging. London, UK: Wiley, 26. [3] B. Allen, "Ula wideband wieless senso newoks," in Poc. IEE Wokshop Ula Wideband. London, UK, July 24. [4] G. L. Sübe, Pinciples of Mobile Communicaions, 2nd ed. Nowell, MA, USA: Kluwe Academic Publishes, 2. [5] W. Q. Malik, B. Allen, and D. J. Edwads, "Impac of bandwidh on small-scale fade deph," in Poc. IEEE Global Commun. Conf. (Globecom). Washingon, DC, USA, Nov. 27..75.5..75.5..5.5 3 4.5 Bandwidh, GHz 3 4.5 Bandwidh, GHz 6 5 (a) Coss-ange diecion 6 5 (b) Range diecion Fig. 4. The vaiaion of fingepin ambiguiy wih channel bandwidh, in ems of he UWB channel impulse esponse spaial coelaion, along coss-ange and ange diecions in NLOS. The ue senso locaion is a offse. 5 5