A Preamble Pattern Identification based Synchronization System for UWB-based Wireless Networks

Transcription

1 66 JOURAL OF ETWORKS VOL. 7 O. 4 APRIL 212 A Preamble Pattern Identification based Synchronization System for UWB-based Wireless etwors Zhihong Qian College of Communication Engineering Jilin University dr.qzh@163.com Xue Wang Shuang Zu College of Communication Engineering Jilin University jluwangxue@yahoo.cn; zhushuang5969@163.com Abstract In this paper we address a pattern identification based synchronization system for UWB-based distributed wireless networs. A correlation based pattern identification method is introduced. A networ domain division and nodes tree setting approach according to the principle of Voronoi diagram is proposed and discussed. Therefore synchronization concluding not only synchronization between transmitter and receiver but also cloc drift synchronization for an entire distributed wireless networs is available. Simulation results demonstrate the relationship between patterns and correlation outcomes. Meanwhile a model nodes tree is simulated for UWB-based distributed wireless networs. Index Terms pattern correlation Voronoi diagram synchronization UWB distributed wireless networs I. ITRODUCTIO With the boom of wireless communication technology various wireless communication systems come forth one after another which introduces conflicts of the use of available frequency resources. However demands on applications of wireless communication system are driven tightly day by day. And systems with higher data transfer rate low cost and low power are strongly required. Moe Z. Win a distinguished research in wireless networs noticed novel features and advantages of impulse radio Ref. [1]. Impulse radio communicates with baseband pulses of very short duration typically on the order of a nanosecond thereby spreading the energy of the radio signal very thinly from near dc to a few gigahertz. When this pulse is applied to an appropriately designed antenna the pulse propagates with distortion. Impulse radios must contend with a variety of interfering signals and also must insure that they do not interfere with narrow-band radio systems operating in dedicated bands. These requirements necessitate the use of spread spectrum techniques. A simple means for spreading the spectrum of the ultra-wide bandwidth low duty cycle pulse trains is time hopping with data modulation accomplished by additional pulse position modulation at the rate of many pulses per data symbol which remains of the concept of Ultra Wide Band (UWB). owadays UWB technology has attracted so much attention to meet the great demand and become the focus of research and development in short range wireless communication field which is regarded as one of ey technologies of the next generation wireless communication. In accordance with terms of FCC Ref. [2] Ultra-Wide Band is not defined just to pulse transmission Ref. [3] but can be extended to a continuous transmission technology as long as absolute signal bandwidth is greater than 5MHz. Multi-band orthogonal frequency division multiplexing (MB-OFDM) Ref. [4] based Ultra-Wideband (UWB) systems divide the allocated 7.5 GHz spectrum into 14 bands each with a bandwidth of 528 MHz whereby information is transmitted using OFDM modulation on each band. It fulfills the definition of UWB according to FCC. The very high data rate (48 Mbps and beyond) capability of UWB technology would provide a compelling cable-replacement wireless technology. MB-OFDM based UWB system has been proposed for the IEEE a Ultra Wideband standard Ref. [5] the new Wireless-USB PHY layer standard the standard ECMA-368 Ref. [6] and ECMA-369. Synchronization is evidently a significant issue for both UWB systems and OFDM based systems. Critical multipath effect in wireless channel would lead to transmitting signal synchronization loss and subcarrier drifts Ref. [7]. Synchronization loss could cause inter carrier interferences (ICI) and inter symbol interferences (ISI) what s worse cause orthogonality loss of OFDM subcarriers as a result degrade system performance. There are a number of researches on synchronization of UWB-based wireless networs. Reference [8] proposes effective frequency offset estimation based on BLUE synchronization proposal for MB-OFDM based UWB systems. Reference [9] is a classical frequency offset estimation approach for OFDM based systems which is referred a lot. Reference [1] proposes an integer frequency offset estimator by frequency domain spreading for UWB multiband OFDM. References [11] [12] give non-coherent methods of UWB signal acquisition based on genetic algorithm. Since the structure for MB-OFDM based UWB systems has been defined in 212 ACADEMY PUBLISHER doi:1.434/jnw

2 JOURAL OF ETWORKS VOL. 7 O. 4 APRIL literature [6] already corresponding schemes are in need to improve system synchronization capability. Adaptive timing synchronization estimators are proposed in Ref. [13] [14] which are implemented by using energy ratio of received symbols. Reference [15] proposes a multi-band estimation and compensation scheme for ultra wideband communications. Reference [16] studies a synchronization design for UWB-based wireless multimedia systems. Reference [17] gives us a robust and accurate frequency and timing synchronization method using chirp signals. Reference [18] is an improved method for CFO synchronization in MB-OFDM UWB. All of the above researches on synchronization have solved a great number of issues for UWB-based systems but there are still some critical problems to be wored out. Classical methods e.g. Ref.[8] and [9] are of good adaptability to most based systems but could not be used to implement synchronization in MB-OFDM based UWB systems directly because of the special structure of in this system. Secondly concerning the wor of synchronization of MB-OFDM based UWB systems a majority of estimation algorithms proposed are of great performance for pattern 1 ( or equivalently 2) but the performance for pattern 3 (or equivalently 4) are ignored. Thirdly most algorithms are proposed on the assumption that modes are nown to receivers. However studies about how to now the exact mode used is lac of research. Meanwhile synchronization issues in UWB based distributed networs have not been wored out perfectly. Aiming to solve the problems we address a pattern identification scheme and a synchronization system for UWB-based distributed networs based on this. The rest of the paper is organized as follows: Section II presents UWB-based systems and synchronization description. The proposed identification synchronization and systems for UWB-based distributed networs are given in Section III. Section IV shows the simulation results and discussions. Conclusion and summary are provided in Section V. II. SYSTEMS AD SYCHROIZATIO DESCRIPTIO A. MB-OFDM Specifications In the MB-OFDM-based UWB system the carrier frequency is hopped with a pre-defined set of carrier frequencies according to a time-frequency code. ECMA standard specifies three types of time-frequency codes (TFCs): TFI TFI2 and FFI. Preamble patterns are associated with different time-frequency codes. Each pattern is constructed by 23 synchronization sequences and 6 channel estimation sequences (CE). Figure 1 shows the structure of pattern 1 and 2 according to TFI. The first 21 sequences of synchronization sequences are PS and the other three are FS. Preamble pattern 3 and 4 which are defined according to TFI2 are interleaved. Figure 1. Preamble structure B. UWB Channel Model The IEEE channel modeling sub-committee has adopted modified Saleh-Valenzuela (S-V) model that can distinguish clusters and rays arrival rates. The IEEE a UWB RF channel model described in Ref. [19] is given by L K h = X α δ ( t T τ ) (1) l= = whereα l is the channel coefficient for th ray of lth cluster; T l is the delay of lth cluster; τ l is the delay of th ray related to lth cluster arrival time; X is the log-normal shadowing on the amplitude. C. Signal Model Suppose frequency offset has been estimated and compensated perfectly. The transmitted sequence in band can be expressed as S l = { s () s (1) s ( n 1) s ( s ( n + 1) }; n (2) The received sequence considering channel response is addressed as R = { r () r (1) r ( n 1) r ( r ( n + 1) }; n (3) The nth sample of lth OFDM transmitted symbol in band is L b r l n = 1 ( ) s ( l n i) h ( i) + w ( l ;1 l L (4) i= where s ( l is the nth sample of lth symbol in band. w ( n l) is the corresponding AWG sample. The transmitted signals can be described using a complex baseband signal notation. The actual RF transmitted signal is related to the complex baseband signal as follows: = 1 rrf Re r ( t TSYM )exp( j2π f t) (5) = where Re( ) represents the real part of a complex variable r (t) is the complex baseband signal of the th OFDM l l 212 ACADEMY PUBLISHER

3 662 JOURAL OF ETWORKS VOL. 7 O. 4 APRIL 212 symbol and is nonzero over the interval from to T SYM is the number of OFDM symbols T SYM is the symbol interval and f is the center frequency for the th band. The exact structure of the th OFDM symbol depends on its location within the pacet: r r = r r header data < header < < data header (6) All of the OFDM symbols r (t) can be constructed using an inverse Fourier transform with a certain set of coefficient C n where the coefficients are defined as either data pilots or training symbols: ST / 2 r = Cn exp( j2πnδ f ) n= ST / 2 t [ T ] ( t T ) t [ T T + T ] t [ T FFT FFT + T T FFT + T + T ] GI (7) The parameters f and ST are defined as the subcarrier frequency spacing and the number of total subcarriers used respectively. The resulting waveform has duration of T FFT = 1/f. Shifting the time by T creates the circular prefix which is used in OFDM to mitigate the effects of multipath. The parameter T GI is the guard interval duration. D. Synchronization in UWB-based systems Synchronization problems in UWB-based wireless distributed system mainly conclude carrier frequency synchronization symbol timing synchronization sampling synchronization and cloc drift synchronization. Carrier frequency synchronization means the synchronization between the receiver and transmitter and that between the sub-carrier frequency synchronization which will mae a direct impact on the sub-carrier orthogonality resulting in inter-carrier interference (ICI). Symbol timing synchronization refers to how to find the correct symbol start position at receivers so as to do error-free demodulation to data. When timing errors estimated mae the FFT symbol window beyond the borders ISI and ICI will wor. Sampling synchronization is to estimate and compensate the sampling frequency asynchronies between the transmitters A/D and receivers A/D. sampling synchronization error will lead to ICI among sub-carriers sampled. Cloc offset and drift synchronization is to estimate cloc offset in a networ and compensate it. Although wide bandwidth decreases the sensitivity to frequency offset the demands of high transfer rate and fast frequency-hopping mae it more difficult to achieve carrier frequency synchronization. The capturing speed and accuracy requirement enhance difficulties to timing synchronization. Sampling synchronization maes much less impact to the system. Cloc drift synchronization is of significant value for UWB-based distributed networs which would lead to miss receiving for cloc time loss. Therefore synchronization systems for UWB-based distributed networs are seriously required which should consider not only synchronization in physical layer but also up layers. III. PREAMBLE PATTER IDETIFICATIO AD SYCHROIZATIO SYSTEMS FOR UWB-BASED WIRELESS ETWORKS A. Preamble patterns specification The standard PL which is shown in Figure consists of three distinct portions: pacet synchronization sequence frame synchronization sequence and the channel estimation sequence. The pacet synchronization sequence shall be constructed by successively appending 21 periods denoted as {PS PS 1 PS 2 } of a time-domain sequence. Each piconet will use a distinct time-domain sequence. Each period of the timing synchronization sequence shall be constructed by pre-appending 32 zero samples and by appending a guard interval of 5 zero samples to time domain sequences. This portion of the can be used for pacet detection and acquisition coarse carrier frequency estimation and coarse symbol timing. Similarly the frame synchronization sequence shall be constructed by successively appending 3 periods denoted as {FS FS 1 FS 2 } of an 18 degree rotated version of the time-domain sequence specified above. Again each period of the frame synchronization sequence shall be constructed by pre-appending 32 zero samples and by appending a guard interval of 5 zero samples to the sequences mentioned above. This portion of the can be used to synchronize the receiver algorithm within the. Finally the channel estimation sequence shall be constructed by successively appending 6 periods denoted as {CE CE 1 CE 5 } of the OFDM training symbol. This training symbol is generated by passing the frequency-domain sequence though the IFFT and pre-appending the output with 32 zero samples and appending and a guard interval consisting of 5 zero samples to the resulting time-domain output. This portion of the can be used to estimate the channel frequency response for fine carrier frequency estimation and fine symbol timing. TFCs are listed in table 1. C C 127 C C 1... C 127 Pacet Sync Sequence 21 OFDM symbols C C 127 C C 1... C 127 PS PS 1 PS 2 FS FS 1 FS 2 CE CE 1 CE 5 Frame Sync Sequence 3 OFDM symbols μs Channel Est Sequence 6 OFDM symbols Figure 2. Standard PL format for a Mode 1 device 212 ACADEMY PUBLISHER

4 JOURAL OF ETWORKS VOL. 7 O. 4 APRIL TABLE 1 TFCS CORRESPODIG WITH PREAMBLE PATTERS TFCs Preamble pattern number Sub-bands sequences 1 TFI TFI TFI TFI FFI FFI FFI B. Correlation based pattern identification The cross correlation between the lth received OFDM symbol and sequence is addressed as R ( l τ ) = 1 r n= ( l n + τ ) d ( ; τ 1 1 l L (8) SYM where {} d { d d 2 } = is the predefined 1 d sequence. Auto correlation between consequence received symbols in frequency band is where ( l + d R ( = ( l r r (9) ( l l+ d L) 3 d = 6 SYM SYM 1 & 2 3 & 4 ( TFI ) ( TFI 2) (1) L is the total number of symbols in one band. By substituting (4) into (9) we obtain R ( l j d SYMs / ( e x ( l x ( l + d + G W ) 2πε + (11) = where ( l is the channel output signal samples x corresponding to the th frequency band. 1 x ( l = 1 m= C ( m) H ( m) e j2πmn (12) where H (m) is the channel transfer function for mth frequency band. G + e = e j2πε ( n+ pre + l SYM ) / j2πε ( n+ pre + ( l+ d ) SYM ) / W x ( l w x ( l + d) w + ( l + d ( l (13) = w ( l w ( l d (14) The transmitter uses different patterns and then the pea of auto correlation output would have different locations. If pattern 1 is used then the output in cases that d= SYM d=3 SYM and d=6 SYM are totally different. The output in the case of d= SYM would be of a rather small value for there is no transmitted signal in the frequency sub-band for pattern 1. The pea would be the output of the current symbol and the symbol in a d=3 SYM distance. If pattern 3 is adopted the pea would be the output of the current symbol and the symbol in a d=6 SYM distance. This feature could be used for pattern identification. C. Synchronization proposals for UWB-based distributed networs For UWB-based distributed networs synchronization is not only about synchronization between a transmitter and a receiver. Synchronization of the whole networs is in order which refers to a large scale of synchronization concluding frequency offset estimation timing synchronization sampling synchronization as well as cloc offset and drift estimation and its compensation. Synchronization about frequency and timing sampling could be operated from one node to the other node. But cloc synchronization ought to tae all the nodes in a distributed networ e.g. wireless sensor networs (WS) into consideration. Therefore if we want to get a synchronization of the whole networs an effective approach to lin all nodes is of significant importance. To do advantages to system synchronization and reduce energy consumption as much as possible we intend to utilize Voronoi diagram to partition nodes domain so as to get a nodes classification to build lins among all the nodes. In mathematics a Voronoi diagram is a special ind of decomposition of a metric space determined by distances to a specified discrete set of objects in the space e.g. by a discrete set of points. Voronoi diagram has a property that the nodes in a sub-domain have a smaller distance to ey node in the domain than to other ey nodes. Therefore we consider building a nodes tree by domain division according to the principle of Voronoi diagram. odes relationship of distance in a Voronoi diagram is shown in Figure 3. According to this feature and principle of Voronoi diagram nodes in UWB-based distributed networs can be classified. The nodes that have a same distance to two adjacent ey nodes can be acted as a special line. In Figure 3 the arrow shows the special line. The nodes on the left of the special line have a smaller distance to A than to B. odes on the right side of the line have a smaller distance to B than to A. Utilizing this feature we can divide the whole networs into several sub-domains nodes inside a certain domain have the smallest distance to the ey node of the domain. Applying the feature to the 212 ACADEMY PUBLISHER

5 664 JOURAL OF ETWORKS VOL. 7 O. 4 APRIL 212 process of node trees building then an energy saving nodes tree could be achieved. d a = d b d a < d b B A d a > d b Figure 3. odes distance in a Voronoi diagram IV. SIMULATIO AD DISCUSSIO We simulate our proposal in IEEE a Ref. [5] channel model 1 and 2 Ref. [19].The channel is invariant for the duration of. We have illustrated in the previous sections that there are 24 synchronization sequences in one MB-OFDM based UWB frame 21 pacet synchronization sequences and 3 frame sequences. In the simulating process we set pf =6 to simulate and analyze the method for convenience which is constructed by PS and FS sequences as [PS 1 PS 2 PS 3 PS 4 FS 1 FS 2 ] for pattern 1 and 2 and [PS 1 PS 2 PS 3 PS 4 FS 1 PS 5 ] for pattern 3 and 4. We adopt TFC1 and TFC3 to simulate our proposals in 1 and 3 respectively. Parameters in our simulations are according to the specifications Ref. [6] which are shown in Table II. Correlation features of different patterns are simulated firstly. A node tree model is simulated to show the distributed wireless networs nodes domain division and nodes tree setting up process the structure of which saves energy consumption and can be used to get a synchronization system. A. Preamble pattern identification We simulate the correlation of TFC1 and TFC3 to see the correlation properties and polarities in correlation. Simulation results are shown in Figure 4 through Figure 7 from which we can get the conclusion that the polarity would precipitate in correlation. Especially in the curves of cross correlation the polarities are indicated more evidently. Therefore we can see that patterns and polarities could be achieved by the outcome of correlation no matter auto correlation or cross correlation. TABLE II. PARAMETERS I SIMULATIO P umber of pilot subcarriers 12 G umber of guard subcarriers 1 T Total number of subcarriers used 122=( D + P + G ) D f Subcarrier frequency spacing MHz(=f s / FFT ) T FFT IFFT and FFT period ns ZPS umber of samples in zero-padded suffix 37 T ZPS Zero-padded suffix duration in time 7.8 ns(= zps /f s ) T SYM Symbol interval 312.5(=T FFT +T ZPS ) F SYM Symbol rate 3.2 MHz(=T -1 SYM ) tot Total number of samples per symbol 165 pf umber of symbols in the pacet/frame synchronization 6 sequence T pf Duration of the pacet/frame synchronization sequence 1.25 ns Correlation value Correlation value Samples index Figure 4. Auto Correlation for TFC Samples index Parameter Description Value f s Sampling frequency 528 MHz FFT Total number of subcarriers(fft size) 128 D umber of data subcarriers 1 Figure 5. Auto Correlation for TFC3 212 ACADEMY PUBLISHER

6 JOURAL OF ETWORKS VOL. 7 O. 4 APRIL Then a node tree base on domain division can be wored out as Figure 9. Therefore nodes in the networ are connected with a ind of prority. And syncrhonizaiton no matter syncrhonizatoin in physical layer or networ layer MAC layer could be implemented with the ind of node tree. Cloc offset and drift synchronization can be carried out in both sub-domain and the entire networs which is up to the syncrhonizaiton demand. Energy consumption would decrease under this principle. When nodes within networs get increasing the node tree could divide new domain adaptively Figure 6. Cross correlation for TFC Figure 9. odes tree of UWB-based distributed networs Figure 7. Cross correlation for TFC3 B. Voronoi diagram in UWB-based distributed networs To express our proposal more clearly we simulate one condition that there are n nodes with 5 ey nodes in. Then we divide the domain into 5 sub-domains with the principle of Voronoi diagram each with one ey node the result of which is shown in Figure 8. 5 C. pattern identification based synchronization analysis Aiming to analyze the features of different patterns we simulate a timing synchronization algorithm based on pattern identification synchronization probabilities of which are shown in Figure 1. It indicates the importance of patterns identification for synchronization of different patterns are really different synchronization ratio (%) in CM1 and TFC1 in CM1 and TFC3 in CM2 and TFC1 in CM2 and TFC Figure 8. Domain division with Voronoi diagram principle SR(dB) Figure 1. Synchronization probability of pattern identification based timing synchronization scheme 212 ACADEMY PUBLISHER

7 666 JOURAL OF ETWORKS VOL. 7 O. 4 APRIL 212 V. COCLUSIO This paper addresses a pattern identification based synchronization systems for UWB-based distributed networs. Correlation based pattern identification and Voronoi diagram based networ domain division are expressed as well as nodes tree building up. Simulations demonstrate the features of correlation and the importance of pattern identification. Meanwhile the domain division and nodes tree setting based on Voronoi diagram provides a principle to synchronization all nodes in a distributed networ with a relatively small consumption. The pattern identification idea and networs nodes tree can be applied to any UWB-based distributed wireless networs. ACKOWLEDGMET This paper is supported by the ational atural Science Foundation of China (o.6941 o ) the Doctoral Fund of Ministry of Education of China (o ) and Graduate Innovation Fund of Jilin University (o ). REFERECES [1] M. Z. Win R. A. Scholtz Impulse radio: how it wors IEEE Commun. Lett. vol. 2 no. 2 pp [2] Federal Communications Commission Revision of part 15 of the commission s rules regarding Ultra-Wideband transmission systems: First report and order Technical Report FCC [3] M. Z. Win R. A. Scholtz Ultra-Wide bandwidth time-hoppingspread-spectrum impulse radio for wireless multiple-access communications IEEE Trans. Commun. vol. 48 no. 4 pp [4] Batra et al. Multi-band OFDM physical layer proposal for IEEE tas group 3a IEEE P /268r3 Orlando FL USA Mar. 24. [5] IEEE P82.15 Wireless Personal Area etwors (WPAs) Group 3a Multi-band OFDM physical layer proposal for IEEE tas group 3a Mar. 24. [6] Standard ECMA-368 High rate ultra wideband PHY and MAC standard. 1st Edition Dec. 25. [7] H.Steendam M.Moeneclaey Synchronization sensitivity of multi-carrier systems European Commun. ETT special issue on multi-carrier spread spectrum vol. 52 no. 5 pp [8] Yinghui Li Hlaing Minn Jacobs T Win M Frequency offset estimation for MB-OFDM-based UWB systems IEEE Trans. Commun. vol. 56 no. 6. pp [9] H. Minn P. Tarasa Improved maximum lielihood frequency offset estimation based on lielihood metric design IEEE Trans. Signal Processing vol. 54 no. 6 pp [1] Yang. H Jeong. KS Yi. JH et al. Integer frequency offset estimator by frequency domain spreading for UWB multiband-ofdm IEICE Transactions on Fundamentals of Electronics Communications and Computer Sciences vol. E93A no. 3 pp [11] Yang Zhihua Zhang Qinyu Zhang aitong. A non-coherent method of UWB siganl acquisition based on genetic algorithm Chinese Journal of Electronics vol.38 no.7 pp [12] Kreiser D. Olonbayar S. Efficient synchronization method for IR-UWB a non-coherent energy detection receiver 21 IEEE/ACM SCom 18-2 Dec. 21 pp [13] Sen. D Charabarti. S Kumar. RVR Some interesting results on compatible BER analysis issues related to multi-band timing and frequency synchronizers applicable for MB-OFDM based UWB communications Digital Signal Rrocessing vol. 21 no. 2 pp [14] Sen. D Charabarti. S Kumar. RVR An adaptive timing synchronization scheme for multi-band orthogonal frequency division multiplexing based Ultra-Wideband communication systems Wireless Personal Communication vol. 53 no. 2 pp [15] Sen D et al.a multi-band timing estimation and compensation scheme for Ultra-Wideband communications IEEE GLOBECOM 28 pp [16] Ye Z. Z Duan C. J et al. A synchronization design for UWB-based Wireless Multimedia systems IEEE Transactions on Broadcasting vol. 56 no.2 pp [17] Sandrine Boumard Aarne Mammela. Robust and accurate frequency and timing syncrhonization using chirp signals IEEE Transactions on Broadcasting vol.55 no.1 pp [18] Aymen M. Karim Masuri Othman Improved fine CFO syncrhonization for MB-OFDM UWB IEEE Communications Letters vol.14 no.4 pp [19] A. F. Molisch J. R. Foerster M. Pendergrass Channel models for ultrawideband personal area networs IEEE Wireless Commun. Mag. Vol.1 no.6 pp Zhihong Qian professor of Communication and Information System at the College of Communication Engineering Jilin University P.R. China. In 1982 he received the B.Sc. degree in Communication from the Xian College of Aeronautic Engineering China. He graduated with the M.Sc. degree in Communication and Electronics Systems at Jilin University of Technology (JUT) in 1991 and the Ph.D. in Communication and Information Systems at Jilin University China in 21. He wored for the Department of Electronic Engineering at Aeronautic Institute of technology from 1982 to 1996 as Teaching Assistant and Assistant Professor of Communication Engineering respectively. He joined the College of Communication Engineering at Jilin University in 1996 and wored as Associate Professor and currently Professor and Ph.D. Candidate Students Supervisor. He has wored as a visiting researcher at University of Massachusetts (21) University of Texas (25) and Virginia Tech (26) all in USA respectively. His research wor focuses on wireless communication and networs including ZigBee Radio Frequency Identification (RFID) Wireless Sensor etwors (WS)UWB and Internet of Things (IoT). In particular he is currently involved in a number of projects on the application of wireless networs. He is the author of 3 monographs has been granted 3 patents has completed 2 research projects with his co-operators as a principal investigator or main co-operator and authored and co-authored more than 8 research papers in national or international academic journals and conferences. Xue Wang was born in Jilin province China. She received the B.Sc. and M.Sc. from Jilin University in 27 and 29 Jilin China. And now she is a doctorial student in Jilin University. Her research interests lie in the areas of wireless communications and networing including synchronization scheme and ultra-wideband communications. 212 ACADEMY PUBLISHER