NOH, Min Seok (211 Second Floor, Miraegio OfficetelGuro 3-dong,Guro-gu, samSeoul 152-053, KR)
KWON, Yeon Hyeon (402 Smile Village, 103-6 Yuljeon-dongJangan-gu,Suwon-si, Gyeonggi-do 440-320, KR)
PARK, Hyun Hwa (1133, Shinan Metro Kahn1115, Bisan-dong,Dongan-gu,Anyang-si, Gyeonggi-do 431-050, KR)
LEE, Hyun Woo (204 Neo Village, 1103-3 Hogye-dongDongan-gu,Anyang-si, Gyeonggi-do 431-080, KR)
KIM, Dong Cheol (401 Daemyung Sweet Village, 763 Naeson-dong,Uiwang-si, Gyeonggi-do 437-080, KR)
HAN, Seung Hee (42-35 Yeokchon 1-dong, Eunpyeong-gu, Seoul 122-899, KR)
NOH, Min Seok (211 Second Floor, Miraegio OfficetelGuro 3-dong,Guro-gu, samSeoul 152-053, KR)
KWON, Yeon Hyeon (402 Smile Village, 103-6 Yuljeon-dongJangan-gu,Suwon-si, Gyeonggi-do 440-320, KR)
PARK, Hyun Hwa (1133, Shinan Metro Kahn1115, Bisan-dong,Dongan-gu,Anyang-si, Gyeonggi-do 431-050, KR)
LEE, Hyun Woo (204 Neo Village, 1103-3 Hogye-dongDongan-gu,Anyang-si, Gyeonggi-do 431-080, KR)
KIM, Dong Cheol (401 Daemyung Sweet Village, 763 Naeson-dong,Uiwang-si, Gyeonggi-do 437-080, KR)
CLAIMS
1. A method of generating a code sequence in a wireless communication
system, the method comprising:
5 recognizing a desired length of the code sequence;
generating a code sequence having a length different from the desired length;
and
modifying the length of the generated code sequence to equal the desired
length,
LO wherein the step of modifying includes discarding at least one element of the
generated code sequence or inserting at least one null element to the generated code
sequence.
2. The method of claim 1 , wherein the length of the generated code sequence is
L 5 a prime number.
3. The method of claim 1, wherein the length of the generated code sequence is
a smallest prime number larger than the desired length.
4. The method of the claim 1, wherein the length of the generated code
sequence is a largest prime number smaller than the desired length.
5. The method of claim 1, further comprising selecting the length of the
generated code sequence nearest to the desired length between a smallest prime number
larger than the desired length or a largest prime number smaller than the desired length.
6. The method of claim 1, further comprising performing circular shift to the
generated code sequence.
7. The method of claim 1, further comprising performing circular shift to the
generated code sequence having the modified length.
8. The method of claim 1, further comprising adjusting a power constraint of
the generated code sequence having modified length to a total power constraint of the
recognized the code sequence having the desired length.
9. The method of claim 1, wherein a number of at least one element to be
discarded corresponds to a difference in number between the desired length and the length
of the generated code sequence.
10. The method of claim 1, wherein a number of at least one null element to be
inserted corresponds to a difference in number between the length of the generated code
sequence and the desired length.
11. The method of claim 1, wherein the step of modifying further includes
inserting a cyclic prefix or a cyclic postfix to the generated code sequence.
12. A method of generating a code sequence in a wireless communication
10 system, the method comprising:
recognizing a desired length of a first code sequence;
generating a second code sequence having a length different from the desired
length of the first code sequence; and
modifying the length of the second code sequence to equal the desired length
L 5 of the first code sequence,
wherein the step of modifying includes discarding at least one element of the
modified code sequence if the length of the modified code sequence is longer than the
desired length of the first code sequence or inserting at least one null element to the
modified code sequence if the length of the modified second code sequence is shorter than
.0 the desired length of the first code sequence.
13. The method of claim 12, wherein the length of the second code sequence is a
prime number.
14. The method of claim 12, wherein the length of the second code sequence is a
smallest prime number larger than the desired length of the first code sequence.
15. The method of the claim 12, wherein the length of the second code sequence
is a largest prime number smaller than the desired length of the first code sequence.
10
16. The method of claim 12, further comprising selecting the length of the
second code sequence nearest to the desired length between a smallest prime number larger
than the desired length or a largest prime number smaller than the desired length.
L 5 17. The method of claim 12, further comprising performing circular shift to the
second code sequence.
18. The method of claim 12, further comprising performing circular shift to the
second code sequence having the modified length.
.0
19. The method of claim 12, further comprising adjusting a power constraint of
the second code sequence having modified length to a total power constraint of the
recognized the code sequence having the desired length.
20. The method of claim 12, wherein a number of at least one element to be
discarded corresponds to a difference in number between the desired length and the length
of the second code sequence.
21. The method of claim 12, wherein a number of at least one null element to be
.0 inserted corresponds to a difference in number between the length of the second code
sequence and the desired length.
22. The method of claim 12, wherein the step of modifying further includes
inserting a cyclic prefix or a cyclic postfix to the second code sequence.
23. An apparatus for generating a code sequence in a wireless communication
system, the apparatus comprising:
a sequence selection unit for recognizing a desired length of the code
sequence, generating a code sequence having a length different from the desired length, and
! 0 modifying the length of the generated code sequence to equal the desired length, wherein the sequence selection unit discards at least one element of the generated code sequence or
inserts at least one null element to the generated code sequence in modifying the length of
the generated code sequence; and
a transmitting unit for transmitting the modified generated code sequence via
at least one antenna. |
METHOD AND APPARATUS FOR GENERATING AND TRANSMITTING CODE
SEQUENCE IN A WIRELESS COMMUNICATION SYSTEM
TECHNICAL FIELD
The present invention relates to a method of generating and transmitting code
sequence, and more particularly, to a method and apparatus for generating and transmitting
code sequence in a wireless communication system.
BACKGROUND ART
Usually, a pilot signal or a preamble of a wireless communication system is referred
to as a reference signal used for initial synchronization, cell search, and channel estimation.
Further, the preamble is comprised of a code sequence, and the code sequence is further
comprised of orthogonal or quasi-orthogonal which represent good correlation properties.
For example, a Hadamard matrix of 128x128 is used in a portable internet (PI) to
insert the code sequence to the frequency domain. In so doing, 127 types of code sequences
are used.
Although the Hadmard code sequence and a poly-phase Constant Amplitude Zero
Auto-Correlation (CAZAC) code sequence are orthogonal codes, a number of codes used to
maintain orthogonality is limited. For example, a number of N orthogonal codes in a NxN
Hadamard matrix is N, and a number of N orthogonal codes that can be expressed by the
CAZAC codes is N and a prime number smaller than N (David C. Chu, "Polyphase Codes
with Good Periodic Correlation Properties," Information Theory IEEE Transaction on, vol.
18, issue 4, pp. 531-532, July 1972). With respect to CAZAC sequence types, GCL CAZAC
and Zadoff-Chu CAZAC are often used.
If the code sequence is generated using the Hadamard codes, N number of sequence
types corresponding to the entire length of the codes is generated. However, the if the code
sequence is generated using the CAZAC codes, only half or N/2 number of sequence types
are generated.
DISCLOSURE OF THE INVENTION
Accordingly, the present invention is directed to a method and apparatus for
generating and transmitting code sequence in a wireless communication system that
substantially obviates one or more problems due to limitations and disadvantages of the
related art.
An object of the present invention is to provide a method of generating a code
sequence in a wireless communication system.
Another object of the present invention is to provide an apparatus for generating a
code sequence in a wireless communication system.
Additional advantages, objects, and features of the invention will be set forth in part
in the description which follows and in part will become apparent to those having ordinary
skill in the art upon examination of the following or may be learned from practice of the
invention. The objectives and other advantages of the invention may be realized and
attained by the structure particularly pointed out in the written description and claims hereof
as well as the appended drawings.
To achieve these objects and other advantages and in accordance with the purpose of
the invention, as embodied and broadly described herein, a method of generating a code
5 sequence in a wireless communication system includes recognizing a desired length of the
code sequence, generating a code sequence having a length different from the desired length,
and modifying the length of the generated code sequence to equal the desired length. Here,
the step of modifying includes discarding at least one element of the generated code
sequence or inserting at least one null element to the generated code sequence.
LO In another aspect of the present invention, method of generating a code sequence in
a wireless communication system includes a recognizing a desired length of a first code
sequence, generating a second code sequence having a length different from the desired
length of the first code sequence, and modifying the length of the second code sequence to
equal the desired length of the first code sequence. Here, the step of modifying includes
L 5 discarding at least one element of the modified code sequence if the length of the modified
code sequence is longer than the desired length of the first code sequence or inserting at
least one null element to the modified code sequence if the length of the modified second
code sequence is shorter than the desired length of the first code sequence.
In a further aspect of the present invention, an apparatus for generating a code
_ 0 sequence in a wireless communication system includes a sequence selection unit for
recognizing a desired length of the code sequence, generating a code sequence having a
length different from the desired length, and modifying the length of the generated code
sequence to equal the desired length, wherein the sequence selection unit discards at least
one element of the generated code sequence or inserts at least one null element to the
generated code sequence in modifying the length of the generated code sequence, and a
transmitting unit for transmitting the modified generated code sequence via at least one
antenna.
It is to be understood that both the foregoing general description and the following
detailed description of the present invention are exemplary and explanatory and are
intended to provide further explanation of the invention as claimed.
BRIEF DESCRIPTION OF THE DRAWINGS
The accompanying drawings, which are included to provide a further understanding
of the invention and are incorporated in and constitute a part of this application, illustrate
embodiment(s) of the invention and together with the description serve to explain the
principle of the invention. In the drawings;
FIG. 1 illustrates a structure of an apparatus for transmitting data using Orthogonal
Frequency Division Multiplexing (OFDM) or OFDM Access (OFDMA) scheme;
FIG. 2 illustrates a structure of an apparatus for receiving data using
OFDM/OFDMA scheme;
FIG. 3 is a flow diagram illustrating adjusting a code sequence;
FIG. 4 illustrates cross-correlation properties of the generated code sequence;
FIG. 5 illustrates a generated CAZAC sequence a N^ ^ N using N (=1024);
FIG. 6 illustrates a cross-correlation properties cumulative distribution function
(CDF) of the code sequences that can be generated according to the code sequence a N MXN
and the CAZAC sequence α w NXN when N= 1024;
5 FIG. 7 illustrates the cross-correlation properties CDF of the code sequences that
can be generated based on the CAZAC sequence generated using the prime number of
N=I 031 and a code sequence set a N xN having length of 1024 (seven (7) elements
removed);
FIG. 8 illustrates a method of generating CAZAC sequence using a length required
0 by a communication system;
FIG. 9 illustrates a method of generating a CAZAC sequence using a padding
portion;
FIG. 10 illustrates an exemplary application of circular shift;
FIG. 11 is an exemplary diagram illustrating application of circular shift to the
L 5 generated code sequence after the elements of the code sequence are removed;
FIG. 12 is an exemplary diagram illustrating application of circular shift to the
generated code sequence prior to removing the elements of the code sequence;
FIG. 13 is an exemplary diagram illustrating application of circular shift to the
generated code sequence after a padding portion is attached;
!0 FIG. 14 is an exemplary diagram illustrating application of circular shift to the
generated code sequence prior to attaching a padding portion;
FIG. 15 is an exemplary diagram of a padding portion of the code sequence in which
the padding portion is used as a lower bandwidth guard interval;
FIG. 16 is a structural diagram for transmitting the code sequence. Depending on
whether the transmission of the code sequence is made in a downlink direction or an uplink
direction, the structure can be in different foπn;
FIG. 17 is a structural diagram illustrating a basic code sequence generation unit and
a code sequence length adjustment unit;
FIG. 18 illustrates cross-correlation characteristics of the code sequence;
FIG. 19 illustrates cross-correlation characteristics of the code sequence; and
FIG. 20 is an exemplary diagram illustrating boosting the power of the generated
code sequence.
BEST MODE FOR CARRYING OUT THE INVENTION
Reference will now be made in detail to the preferred embodiments of the present
invention, examples of which are illustrated in the accompanying drawings. Wherever
possible, the same reference numbers will be used throughout the drawings to refer to the
same or like parts.
Figure 1 illustrates a structure of an apparatus for transmitting data using Orthogonal
Frequency Division Multiplexing (OFDM) or OFDM Access (OFDMA) scheme. Figure 2
illustrates a structure of an apparatus for receiving data using OFDM/OFDMA scheme.
In Figure 1, traffic data and control data are multiplexed at a muxer 11. Here, the
traffic data is used to provide service from a transmitting end to a receiving end, and the
control data is used to facilitate transmission from the transmitting end to the receiving end.
The discussion relating to the present invention regarding the code sequence which relates
5 to a type of a code sequence of the control data. The code sequence can be used for initial
synchronization, cell search, or channel estimation.
Depending on the communication system, the code sequence can be used in various
forms. For example, the code sequence in an IEEE 802.16 wideband wireless access system
can be used in a preamble or a pilot signal format, and in a multi-input, multi-output
10 (MIMO) system, the code sequence can be used as a midamble format.
After being processed at the muxer 11, the multiplexed traffic and control data is
then channel coded by a channel coding module 12. Channel coding is used to allow the
receiving end to correct error that can occur during transmission by adding parity bits.
Examples of channel coding include convolution coding, turbo coding, and low density
L 5 parity check (LDPC) coding.
Thereafter, the channel coded data is modulated by a digital modulation module 13
in which data symbols are mapped using algorithms such as a quadrature phase shift keying
(QPSK) or a 16-quadrature amplitude modulation (16QAM). The mapped data symbols are
then processed by a subchannel modulation module 14 through which the data symbols are
.0 mapped to each subcarrier of the OFDM system or OFDMA system. Then, the data symbols
mapped to subcarriers are processed by an inverse fast Fourier transform (IFFT) module 15
which transform the data symbols into a signal in a time domain. The transformed data
symbols are then processed through a filter 16 and further processed through a digital-to-
analog conversion (DAC) module 17 where the filtered data symbols are converted to
analog signals. Lastly, the analog signals are converted into a radio frequency (RF) by a RF
module 18 which is then transmitted via an antenna 19 to the receiving end.
Based on the type of generated code (e.g., CAZAC code), the steps of channel
coding and/or symbol mapping can be omitted. Figure 2 illustrates a receiving end whose
processes are inverse to those of the transmitting end.
A code sequence is used for transmitting control information, which includes
identification (ID) and synchronization information, to classify types of sequences in a
communication system. In order for more effective reception of the control information
using code sequence, the code sequence can be adjusted or modified. Further, the code
sequence can be applied to all of the channels that use code sequence for control signaling
such as a random access channel (RACH), downlink/uplink reference symbol, channel
quality information (CQI), and Acknowledgement (ACK)/Negative Acknowledgement
(NACK).
Figure 3 is a flow diagram illustrating adjusting a code sequence. More specifically,
a length of the code sequence is defined as N, a number of codes in the code sequence is
defined as N seq _ N , and a code sequence set defined as a N∞ NXN . In operation, the code
sequence set a N NXN having N seq N number of codes can be extended to a code sequence
set a Nχ uxN having N seq _ M number of codes.
Equation a N N is a matrix of N seq N xN of
a ^_ » * = K « ,_ » * a k,_ » * ' '' a »£l* \ > and CI N ' ^_ NXN is a row vector of
<, . ^ = K ,. ^λr(0)4 Kϊ _ ff ^(l)-< ϊ _^(^-l)J • Furthermore, <„_„,» indicates
n(= 0,1,2, • • • , N - 1) element of k(= 0,1,2, • • • , N seq N - 1) code sequence.
Referring to Figure 3, a code sequence set a N xM , having N seq M number of code
sequence(s) where each code sequence has length M, can be generated based on the code
generation algorithm based on code type in which a value of length M is a natural number
greater than a value of length N (S301). Here, the code types include Hadamard code,
Pseudo Noise (PN) code, and a Constant Amplitude Zero Auto-Correlation (CAZAC) code,
among others to be used for initial synchronization, cell search, and channel estimation in
the wireless communication system. The code sequence set having length M per each code
type can be generated by various schemes as discussed. As for the CAZAC code, the value
of length M is a smallest prime number greater than the value of length N, preferably.
Subsequently, a code sequence set a N M%N , having N seq M number of code
sequences, can be generated where a resulting length of the code sequence is length N.
More specifically, the code sequence set a N xM , having N seq M number of code
sequences where each code sequence has length M (from step S301), can have elements of
the code sequence removed. That is, elements which comprise each code sequence can be
removed from the code sequence allowing the length of the code sequence to be adjusted or
shortened. Here, M- N number of elements can be removed from the code sequence whose
length corresponds to length M. By removing elements from the code sequence with length
M, a code sequence having length N can be generated. As discussed, N is smaller than M.
Consequently, a code sequence set a N uxN , having N se(J M number of code sequences in
which each code sequence has length N, can be generated (S302).
A code sequence is used for transmitting control information, which includes
identification (ID) and synchronization information, to classify types of sequences in a
communication system. Currently in 3 ld Generation Partnership Project (3 GPP) Long Term
Evolution (LTE), a CAZAC sequence is being considered.
The CAZAC sequence can be used by channels to output various IDs and
information. The channels include channels for downlink synchronization (e.g., primary
synchronization channel, secondary synchronization channel, and broadcast channel),
uplink synchronization (e.g., random access channel), and pilot channels (e.g., data pilot
and channel quality pilot). Further, the CAZAC sequence can be used in scrambling as well
as channels that use code sequence such as RACH.
Although there are various types of the CAZAC sequences, there are two types of
often used CAZAC sequences - GCL CAZAC and Zadoff-Chu CAZAC. The Zadoff-Chu
CAZAC sequence can be defined by the following equations.
[Equation 1]
c( / k 7 ;N λ γ ,M t s -). = exp f jπMk(k + ϊ)\ ( ,f~or od J d 1 N λ γ ).
{ N J
[Equation 2]
c(k;N,M) (for even N)
Here, k denotes sequence index, N denotes a length of CAZAC to be generated, and
M denotes sequence ID.
If the Zadoff-Chu CAZAC sequence and the GCL CAZAC sequence are expressed
by c(k;N,M) as shown in Equations 1 and 2, then the sequences have the following three
(3) characteristics as presented in following equations.
[Equation 3]
c(k; N,M)\ = 1 (for all k,N,M)
[Equation 4]
* ,^)={ [ l 0 l , ( ( / f w or a ≠ =0 0 ) )
[Equation 5]
R U^ I; N ( d ) = P (f or alϊ M \ > M 2 and N)
According to Equation 3, the CAZAC sequence always has a size of 1, and the
CAZAC sequence of Equation 4 has an auto-correlation function denoted by a delta
function. Here, the auto-correlation is based on circular correlation. Further, Equation 5 is a
cross-correlation which is constant if N is a prime number.
If the length to be applied in the wireless communication system for generating the
CAZAC sequence is denoted by L, a method for generating the CAZAC sequence sets N of
Equations 1 and 2 to equal L (N = L) - identified as step (1). Step (2) can be identified by a
method where a value of iVis set to be a prime number greater than a value of length L for
) generating the CAZAC sequence.
Referring to the characteristics of the generated CAZAC sequence having a
specified length of L, if L is not a prime number, the generated CAZAC sequence can
include M= 1, 2, ... L-X number of codes, some of which are repeated codes. In other words,
the number of different codes is less than L-X number of codes. On the contrary, if L is a
5 prime number, there is L-X number of different codes. The above descriptions may also be
applied to other types of code sequences and are not limited to Zadoff-Chu CAZAC
sequence.
Further, the following technique has been considered. More specifically, if the
length of code to be applied to the system is not a prime number, and there are a large
L 0 number of codes to be used, a smallest prime number greater than L was selected. Using the
selected prime number, the CAZAC sequence was generated, and discards or removes at
least one element of the generated CAZAC sequence for use. This technique is different
than the technique introduced with respect to step 1.
For example, assume that a number of codes of a CAZAC code sequence (N) is
.5 1024. The following equation can be used to express an algorithm for generating a Zadoff-
Chu CAZAC code.
[Equation 6]
a index(A)
In Equation 6, A and M are natural numbers, and index(A)(= 0,l,2,- - -,N seq _ M -l)is
an index of A in ascending order. In order to extend the CAZAC sequence where iV=1024, a
smallest prime number greater than 1024 is used. That is, the smallest prime number greater
than 1024 is 1031. As such, the code sequence set a N ^ where M=I 031 is applied to
Equation 6.
Since M (=1031) is a prime number, N M =1030. Furthermore, A can be referred
to as a seed value used to generate a code sequence maintaining CAZAC properties. If M is
a prime number, a total of Af - 1 number of code sequences can be generated. In other
words, for example, if M= 1024, a total of 512 (=1024/2 or N/2) number of code sequences
are generated. However, if M= 1031, a total of 1030 number of code sequences (M- I) are
generated. Moreover, the cross-correlation properties of the generated code sequence are
better if Mis a prime number than a composite number.
In order to adjust or modify the CAZAC code sequence set a N ^ u where M =
1031 to a code sequence set a N M whose length is N = 1024, M- N number of elements
can be removed from index n = N, • ■ ■ , M - 1 , generating a code sequence set a N N .
In determining the value of M, although the number of code sequences can increase
with increase in value of N, it is preferable to determine the value of M based on the code
sequence whose length is N that promotes maintenance of good correlation properties. In
case of the CAZAC code, optimum correlation properties can be attained if the value of
length M is the smallest prime number greater than the value of length N.
If the code sequence set a N^ ^ N generated using length N=I 024 is compared with
the code sequence set a N xff , a total number code sequences of the former can be
represented by N/2 or 512 (=1024/2) code sequences having an index
0,1,2, • • •, N 12 -1(N = 1024) , and a total number of code sequences of the latter can be
5 represented by M - 1 or 1030 having an index 0,1,2, • • • , M - 2 (Af = 1031) .
Figure 4 illustrates cross-correlation properties of the generated code sequence.
More specifically, the cross-correlation properties of a N k λN (k = l,2,- --,N seq M — X)
associated with the remaining N seq M (1029) code sequences for a N ° λN code sequence of
the code sequence set a N xλ , . The figure illustrates this with respect to amplitude, code
L 0 index, and time index.
Further, Figure 5 illustrates a generated CAZAC sequence a N λN using N (=1024).
More specifically, the figures illustrate cross-correlation properties of
a Nm<1 MxN ( k = ^ 2 ^"' N seq _ M ~ V) regarding the remaining N seq N (511) code sequences. The
figure illustrates this with respect to amplitude, code index, and time index. Between Figure
.5 4 and Figure 5, the cross-correlation properties of the generated code sequence of Figure 4
are better.
Figure 6 illustrates a cross-correlation properties cumulative distribution function
(CDF) of the code sequences that can be generated according to the code sequence a N λN
and the CAZAC sequence a N^ NXN when N= 1024.
' 0 Figure 7 illustrates the cross-correlation properties CDF of the code sequences that
can be generated based on the CAZAC sequence generated using the prime number of
N= 1031 and a code sequence set a N ^ ωχN having length of 1024 (seven (7) elements
removed). The performance lines of Figures 4 - 7 indicate that the code sequence set with
seven (7) elements removed has equivalent cross-correlation properties compared to the
original code sequence set.
As discussed, the codes in addition to the CAZAC code are available, such as the
PN code and the Hadamard code. The discussion with respect to the CAZAC code sequence
can also be applied to the PN code and the Hadamard code. With respect to the PN code, a
modular shift register generator is used to generate the code sequences. If a number of shift
registers generated is represented by N, a code sequence having a length of 2 N - 1 is
generated. Thereafter, a value "1" is added to the shift register, resulting in a length 2 N+1 - 1,
and then, adjust the length to equal 2 N .
With respect to the Hadamard codes, a number of code sequences, which equal the
length of the code sequence, make up a code sequence. However, for example, if M number
of code sequences having length N is required (M > N), then M number of code sequences
having length M are generated, followed by removing a specified number of elements to
make the length of the code sequence equal length N.
Figure 8 illustrates a method of generating CAZAC sequence using a length required
by a communication system. That is, the required (or desired) length of the CAZAC
sequence can be represented by length L. Further, the codes types can be extended.
However, since a generated code sequence can be truncated or have elements discarded to
correspond to the desired length L, the auto-correlation and cross-correlation properties of
the truncated code sequence can experience deterioration. Similarly, even if a code
sequence portion is added/attached to the generated code sequence (e.g., zero-padding or
cyclic prefix) to correspond to the desired length L, the auto-correlation and cross-
correlation properties can experience deterioration. Here, auto-correlation properties relate
to the auto-correlation value being 1 when the delay is 0. Otherwise, the auto-correlation
value is 0 when the delay is a value other than 0. Further, the cross-correlation properties
having a constant value is negatively affected.
Assuming that the code sequence having poor auto-correlation and cross-correlation
properties are removed, the remaining number of code sequences may be less than L — 1.
In order to attain a desired length and a maximum number of CAZAC sequence
types corresponding to the desired length, a smallest prime number, X, greater than the
desired length, L, (X > L) can be selected. Although the CAZAC sequence can be generated
using X, due to deterioration of the correlation properties, the correlations properties of
CAZAC sequence as shown in Equations 4 and 5 cannot be attained. Further, when
selecting a length of the generated code sequence, the length that is nearest to the desired
length L which is between a smallest prime number larger than the desired length or a
largest prime number smaller than the desired length can be selected.
Referring to Figure 8, the generated CAZAC sequence has length X. Thereafter, the
generated CAZAC sequence having length X has elements of the code sequence removed
(or truncated) so as to make the length of the generated CAZAC sequence correspond to the
desired length L.
Figure 9 illustrates a method of generating a CAZAC sequence using a padding
portion. As discussed, the generated CAZAC sequence is truncated. With respect to auto¬
correlation and cross correlation properties, delay of 0 indicates an auto-correlation value of
1, as shown in Equation 4, and a delay not equaling 0 indicates a value of 0. Moreover, the
5 properties where the cross-correlation value is always a prime number is not deteriorated
whereby effective correlation is maintained. Further, additional control information can be
transmitted by using the information inputted to the fading unit.
Referring to Figure 9, the generated CAZAC sequence has length X. Here, the value
of X is a largest prime number less than the value of L. In other words, X is a prime number
L 0 less than L. Thereafter, the generated CAZAC sequence having length Xhas elements added
or a padding portion added to the CAZAC sequence so as to make the length of the
generated CAZAC sequence correspond go the desired length L. Here, Cl represents the
length of the CAZAC sequence having length X, and C2 represents the padding portion. By
combining Cl and C2 (Cl + C2), the generated CAZAC sequence can have a length
.5 corresponding to the desired length L.
Figure 10 illustrates an exemplary application of circular shift. The circular shift is
typically applied to increase amount of control information transmitted to the
communication system. That is, for example, a back portion of the sequence is re-allocated
to a front portion of the sequence, and accordingly, the remaining sequence is shifted in the
0 direction of the back portion of the sequence in amount (or length) corresponding to the re¬
allocated back portion, as illustrated in Figure 2. Further, if specified control information is
applied the circular shift as described above, then the control information amount that can
be transmitted via corresponding sequence increases.
Discussions of above relate to the methods of generating the sequence using the
desired length L, and of increasing transmitted control information using the circular shift.
5 If these methods are applied in generating the sequence, the following processes take place.
First, select a smallest prime number greater than L or a largest prime number less than L,
which is referred to as X. Second, remove or add a sequence unit having a length
corresponding to X-L or L-X. Third, apply the circular shift to the resulting sequence.
Figure 11 is an exemplary diagram illustrating application of circular shift to the
0 generated code sequence after the elements of the code sequence are removed. Referring to
Figure 11, the code sequence 1102 is generated based on length X which is the smallest
prime number greater than length L. In other words, the generated code sequence 1102 has a
length equaling length X which is longer than the desired length L. From the generated code
sequence 1102, a portion having a length corresponding to length X - L is removed,
L 5 resulting in a code sequence having length L 1103. Thereafter, the result of the generated
code sequence 1103 having length L is applied circular shift thereto, resulting in the code
sequence 1104.
Figure 12 is an exemplary diagram illustrating application of circular shift to the
generated code sequence prior to removing the elements of the code sequence. In other
! 0 words, circular shift is performed to the generated CAZAC sequence having length X, and
after circular shift is performed, the elements of the code sequence are removed.
Referring to Figure 12, the code sequence 1202 is generated based on length X
which is the smallest prime number greater than length L. In other words, the generated
code sequence 1202 has a length equaling length X which is longer than the desired length L.
A circular shift is then performed to the generated code sequence 1203 having length X.
Thereafter, a portion of the generated code sequence having a length corresponding to
length X- Z, is removed, resulting in a code sequence 1204 having length L.
Figure 13 is an exemplary diagram illustrating application of circular shift to the
generated code sequence after a padding portion is attached. Referring to Figure 13, the
code sequence 1302 is generated based on length X which is the largest prime number
smaller than the value of length L. To the generated CAZAC sequence 1302, a padding
portion is added 1303. The length of the padding portion corresponds to a length L —X. As
discussed, the padding portion can be comprised of zeroes or cyclic prefix/postfix. With the
addition of the padding portion, the length of the CAZAC sequence equals the desired
length L. Thereafter, the result of the generated code sequence having length L 1303 is
applied circular shift thereto, resulting in the CAZAC sequence 1304.
Figure 14 is an exemplary diagram illustrating application of circular shift to the
generated code sequence prior to attaching a padding portion. In other words, circular shift
is performed to the generated CAZAC sequence having length X, and after circular shift is
performed, the padding portion is attached.
Referring to Figure 14, the code sequence 1402 is generated based on length X
which is the largest prime number smaller than the value of the desired length L. To the
generated CAZAC sequence 1402, circular shift is performed. The circularly-shifted
CAZAC sequence 1403 still has length X To the CAZAC sequence 1403, a padding portion
is added, resulting in the CAZAC sequence 1404. The length of the padding portion
corresponds to a length L — X. As discussed, the padding portion can be comprised of zeroes
or cyclic prefix/postfix. With the addition of the padding portion, the length of the CAZAC
sequence 1404 equals the desired length L.
Between Figures 11 and 12, the difference is that circular shift is performed either
before or after the elements of the CAZAC sequence are removed. By performing circular
shift before removing the elements (or adjusting the length to equal the desired length),
0 correlation deterioration can be reduced. To put differently, the CAZAC sequence does not
have discontinuous codes.
Between Figures 13 and 14, the difference is that circular shift is performed either
before or after the padding portion is added to the generated CAZAC sequence. By
attaching the padding portion after performing circular shift, better correlation properties
5 can be attained, especially since the padding portion is placed at the end of the code
sequence.
Further, according to the discussion above, the desired length L (or required length)
is first recognized. As illustrated with respect to Figures 11 - 14, the generated code
sequence is adjusted/modified based on the desired length L. Based on that, after the desired
D length L recognized, a determination can be made as to whether the generated length X
should be shortened or extended. In other words, the determination can be made whether to
remove or discard at least one element of the generated code sequence or to add or insert at
least one element to the generated code sequence. As discussed, the elements to be inserted
can be a null (0) element (e.g., zero padding) or cyclic prefix/postfix, for example. In order
to make the determination between discarding the element(s) or adding the element(s), the
5 system can choose to select the length closest to the desired length L.
For example, if the desired length L is 75, the value of the smallest prime number
greater than 75 is 79, and the value of the largest prime number smaller than the 75 is 73.
Here, the prime number 73 can be selected since 73 is closer to 75 than 79 is to 75.
Although the illustration above selects the prime number closest to the desired
L 0 length L, selection regarding removal or addition of the element(s) is not limited to the
example of above and other implementations may be applied.
Regarding padding, there are five (5) schemes by which padding can be
accomplished. As a first padding scheme, the padding portion can be comprised of a
constant number (e.g., Os). Although the padding portion is used to fill the portion of the
.5 code sequence so that the length of the code sequence coincides with the desired length, it is
possible for the padding portion to be less then completely full. In other words, it is possible
for that the length of the code sequence with padded portion is not equal to or is shorter than
the code sequence with the desired length. That is, when the code sequence is used for
functions deemed less important, such as for cell search or random access, it is not
! 0 necessary to use the entire length of the code sequence, and as such, the padding portion
does not need to be completely occupied to correspond to the desired length of the code
sequence.
As a second padding scheme, the padding portion can be comprised of a repeated
portion. In other words, the portion corresponding to L - X of the code sequence 1204 can
be duplicated and inserted/attached to the end of the code sequence 1204. This can be
referred to as cyclic postfix. Here, the code sequence uses the entire length L. When
determining the identification (ID) of the code sequence, the entire length L is used to
facilitate identifying of the code sequence ID. At the same time, the generated code
sequence does not experience distortion by using the entire length L. In the discussion
above, the cyclic postfix is used. Alternatively, cyclic prefix can also be used.
As a third padding scheme, the padding portion can be comprised of additional
information through which different messages can be delivered. More specifically, the
desired length L of the code sequence can be used to generate a supplemental code sequence
whose length equals the desired length L (N- L). The code sequence portion corresponding
to L - X is extracted from the supplemental code sequence and inserted/attached to the
generated code sequence as the padded portion.
As a fourth padding scheme, a portion corresponding to length L — X is extracted
from the code sequence and inserted as the padding portion. Here, the code sequence
inserted to the padding portion may be a different code sequence than the code sequence
1204. Put differently, the code sequence inserted to the padding portion may be a CAZAC
sequence having a length of M, for example, which is different from the code sequence
1204 having a length of L. Further, the code sequence inserted to the padding portion can be
a different code sequence other than the CAZAC sequence. By using different code
sequence, additional information can be delivered including information related to type of
code sequence adjustments.
As a fifth padding scheme, the padding portion can be used as lower bandwidth
guard interval. During transmission of control information using a prescribed sequence, the
following possible scenarios can occur such as transmitting data without establishing
synchronization with an access channel, transmitting data by a plurality of users within a
communication system, and distortion of frequency of the received data.
Furthermore, the padding portions can be placed at both ends of the code sequence
to use the padding portions as guard intervals of the lower bandwidth. Consequently, a more
reliable acquisition of control information from the received data can take place despite
distorted frequency signals. In the padding portions used as guard intervals, constant
numbers (e.g., Os) can be used or cyclic prefix or postfix of the generated code sequence can
be used.
If the padding portions are placed at both ends of the code sequence and used as
guard intervals of the lower bandwidth, the code sequences can be protected from frequency
signal distortions. Moreover, if Os are inserted between the guard intervals or put differently,
within the code sequence, interference to neighboring codes can be reduced. Alternatively,
if cyclic prefix/postfix is used as guard intervals, the code sequences can be protected from
frequency distortions and can be used to transmit the control information containing the
sequence ID if there is no frequency distortion.
Figure 15 is an exemplary diagram of a padding portion of the code sequence in
which the padding portion is used as a lower bandwidth guard interval. Referring to Figure
15, the code sequence 1501 can be divided into three (3) parts - a portion (Cl), which is
generated based on length X, and the other two portions (C2 and C3) are attached to both
ends of the code sequence 1501.
In the discussions above, five (5) padding schemes are introduced. However, the
padding schemes are not limited to the discussed schemes, and there can be other types of
padding schemes.
Besides the first padding scheme in which no information is inserted, the other four
padding schemes insert additional information in the padding portions to allow expansion of
the code sequence and/or transmission of message(s). Various information can be inserted
into the padding portion including, for example, initial access information, timing update
information, resource request information, user ID information, channel quality information
(CQI), user group ID information related to a random access channel (RACH). Furthermore,
the information can include cell ID information, multi-input multi-output (MIMO)
information, and synchronization channel information of a synchronization channel (SCH),
for example. In addition, the padding portion can be used for transmitting data for message
transmission as well as arbitrary information using a code sequence having a length of L - X.
Figure 16 is a structural diagram for transmitting the code sequence. Depending on
whether the transmission of the code sequence is made in a downlink direction or an uplink
direction, the structure can be in different form. With that, Figure 16 is described with
respect to a general transmitting end for transmitting the control signal.
5 Referring to Figure 16, the transmitting end 1601 comprises a sequence selection
unit 1602 and a transmitting unit 1603. The sequence selection unit 1602 is used to generate
the code sequence for transmitting the control information. More specifically, the sequence
selection unit 1602 performs an operation to select a code sequence having a desired length
of L. In other words, the sequence selection unit 1602 stores the value of the desired length
L 0 L, and then selects an appropriate code sequence for expressing the control information to
be transmitted where the code sequence has a length of L.
The code sequence that can be selected by the sequence selection unit 1602 has a
length of L as illustrated in Figures 12 and 14 (e.g., code sequence 1204 and code sequence
1404). Moreover, the code sequence is applied circular shift (e.g., code sequences 1203 and
.5 1403) to which a padded portion corresponding to lengths L - X or X- L is removed or
inserted/added. As a result, discontinuous parts are not formed within or in the code
sequence to promote superior correlation characteristics.
Although it is preferable to use length X which is a smallest prime number greater
than the length of L or a largest prime number smaller than the length of L, as long as the
0 value of length X is a prime number, different or other prime numbers can be used as the
value of length X
Figure 17 is a structural diagram illustrating a basic code sequence generation unit
and a code sequence length adjustment unit. In Figure 17, the basic code sequence
generation unit 1701 further comprises a code sequence generation unit 1701a and a circular
shift application unit 1701b. The code sequence generation unit 1701a is used to generate a
5 first code sequence (Cl). Here, Cl can be defined as a code sequence having a length of X
where the value of length X is a smaller prime number larger than the value of length L or a
code sequence having a length of X where the value of length X is a larger prime number
smaller than the value of length L. Cl is then applied circular shift by the circular shift
application unit 1701b. More specifically, the circular shift application unit 1701b receives
LO Cl having length of X, applies circular shift, and outputs a second code sequence (C2) to the
code sequence length adjustment unit 1702.
The code sequence length adjustment unit 1702 further comprises a control unit
1702a, a code sequence removing unit 1702b, and a padding unit 1702c. More specifically,
the control unit 1702a receives C2 as well as the value of length L. The control unit 1702a
L 5 determines whether to remove a portion/section of C2 or insert/add a portion/section to C2.
Based on the determination from the control unit 1702a, C2 is delivered to the sequence
removing unit 1702b in which a portion/section of C2 corresponding to a length of X— L is
removed. Alternatively, C2 can be delivered to the padding unit 1702c for inserting/adding
a portion/section of C2 whose length corresponds to the length of L -X.
!0 If C2 and the value of length L are provided to the control unit 1702a, the control
unit 1702a compares the value of length X which identifies the length of C2 with the value
of the length L. Here, if X is greater than L, then C2 is inputted into the sequence removing
unit 1702b. From C2, the portion length of C2 corresponding to length X - L is removed,
resulting in C3. However, if X is less than L, then C2 is inputted into the padding unit 1702c.
From C2, the padding portion length corresponding to length L -X is inserted/added to C2,
5 resulting in C4. Here, the padding portion can be inserted to either end or both ends of C2.
Figures 18 and 19 illustrate cross-correlation characteristics of the code sequence.
The illustrations of Figures 18 and 19 is based on the value of length X being the smallest
prime number greater than the value of the desired length L; however, the illustrations are
not limited to the smallest prime number greater than length L but can have a prime number
L 0 value of length X smaller than the value of length L.
Referring to Figures 18 and 19, the x-axis represents values of circular shift while
the y-axis represents un-normalized cross-correlation values. Furthermore, a thinner line
represents the value of cross-correlation of the code sequence with circular shift applied
thereto after a code sequence portion having the length X— L is removed. A darker/thicker
L 5 line represents values of code sequence to which circular shift is applied prior to removing
the code sequence portion corresponding to the length X - L. More specifically, Figure 7
illustrates a graph where L is 75 and X is 79 which is the smallest prime number greater than
75. Moreover, Figure 8 illustrates a graph where L 8 is 225 and X is 227 which is the
smallest prime number greater than 225.
! 0 In Figures 18 and 19, if the value of circular shift is 0 or put differently, if there is no
shift, then high correlation value is indicated only when the auto-correlation value of the
code sequence corresponds and in other cases, moderate correlation is maintained. On the
contrary, if the code sequence has a section corresponding to length X — L is removed and
thereafter applied circular shift, severe fluctuations occur with correlation values, resulting
in deteriorated correlation characteristics. As such, if cross-correlation is used to analyze the
code sequence, the code sequence according to the embodiments of the present invention
shows superior performance and outcome to that of the conventional code sequence.
Figure 20 is an exemplary diagram illustrating boosting the power of the generated
code sequence. As discussed, the code sequence is generated based on length X, and a
padding portion, whose length corresponds to length L - X, is attached to the code sequence
(e.g., CAZAC sequence). Thereafter, the portion of the code sequence corresponding to
lengthXis used where length L is divided by length X(ZZZ). The result of the division is the
amount of power that can be boosted. Moreover, the amount of power that can be boosted
can be applied to the code sequence whose length is length X. When the receiving end
receives power boosted code sequence, more effective detection performance can be
achieved since interference is reduced.
However, regarding a code sequence generated with a padding portion with cyclic
prefix/postfix attached thereto, there is no need to power boost since all of the code
sequences corresponding to length L are used for acquiring sequence ID information.
In the receiving end, information related to the generated code sequence and length
X used to generate the code sequence is received. From the code sequence, a portion
corresponding to length X is processed to acquire the control information. To this end, it is
important to first receive synchronization information of the received data. Equation 7 and
Equation 8 can be used to acquire synchronization information. Here, Equation 7 relates to
auto-correlation, and Equation 8 relates to cross-correlation.
[Equation 7]
R M N (d) = ∑c(k,M,X)* c (mod(k + d),X);M,X)
A=O
[Equation 8]
R Mι Ml N {d) = Y J c{k,M x ,X)*c {moά{k + d\X);M 2 ,X)
A=O
Equation 7 is used to acquire auto-correlation value(s) from the received code
sequence whose sequence ID is M. Further, the acquired auto-correlation value d, which is a
value other than 0, is used to achieve synchronization.
Equation 8 is used to acquire cross-correlation value(s) of a code sequence whose ID
is M.% from the received code sequence whose sequence ID is M 1 . Through the acquired
value, synchronization can be achieved.
Typically, if the wireless communication system is a synchronous network, auto-
correlation is used to acquire synchronization information, and if the system is an
asynchronous network, cross-correlation is used to acquire synchronization information.
However, according to the embodiments of the present invention, synchronization
information can be acquired using any one or at least one of the correlation schemes.
After the synchronization information of the received code sequence is acquired, the
receiving end analyzes the received code sequence to acquire the sequence ID, as shown in
Equations 9 and 10.
[Equation 9]
σc(k;M ,X) = c(k + \;M ,X)*c (k;M ,X) (for k = 0,1,- -,L - 1)
[Equation 10]
σc(k;M,X) = c(k + l;M,X)*c * (k;M,X) (for k = 0,1, — ,X-I)
5 In Equations 9 and 10, σc(k;M, X) denotes difference sequence of the received
sequences. Equation 9 is used to acquire the ID information of the received sequence using
the differential sequence corresponding to the total length of the received sequence.
Equation 9 can also be used to acquire the ID information of the code sequence which has
been generated with the cyclic prefix/postfix padded portion. Equation 10 is used to acquire
L 0 the ID information of the received sequence using the smallest prime number corresponding
to length X
As discussed, if the differential sequence of the CAZAC sequence is calculated
using Equations 9 or 10, A: of the sequence index is generated, and the result therefrom is
transformed by the Fourier transform scheme, to show a single peak value. Thereafter, by
.5 detecting the peak value, the ID information of the sequence can be acquired.
The discussion of above regarding a code sequence or a code sequence set can be
applied to 3 rd Generation Partnership Project (3GPP) system or 3GPP2 system as well as a
Wibro system or a Wimax system.—
It will be apparent to those skilled in the art that various modifications and variations
0 can be made in the present invention without departing from the spirit or scope of the
inventions. Thus, it is intended that the present invention covers the modifications and
variations of this invention provided they come within the scope of the appended claims and
their equivalents.
Next Patent: METHOD OF FORMING SILICON FILM BY TWO STEP DEPOSITION