Method for setting cyclic shift considering frequency offset

Procedure to establish cyclic shift considering frequency offset

Cross reference to related requests

The present application claims the benefit of the Korean patent application No. 10-2007-0011772, filed on February 5, 2007 and the Korean patent application No. 10-2007-00102563, filed on October 11, 2007.

The present application also claims the benefit of the US provisional application with serial number 60 / 883,754, filed on January 5, 2007, of the US provisional application with serial number 60 / 884,398, filed on January 10, 2007, of US provisional application with serial number 60 / 915,096, filed on April 30, 2007 and the US provisional application with serial number 60 / 941,562, filed on June 1, 2007.

Background of the invention

Field of the Invention

The present invention is about a sequence of a wireless communication system and, more particularly, about a method for establishing a cyclic shift in consideration of characteristics of a CAZAC sequence to solve the problem of a frequency offset.

Exhibition of related technique

A null constant amplitude automatic correlation sequence (CAZAC) is representative of various sequences that have been set out in detail in 3GPP LTE.

Generally, the channels extract a variety of identifiers (IDs) or information using the CAZAC sequence; for example, synchronization channels (for example, a primary SCH, a secondary SCH and a BCH) for downlink synchronization, other synchronization channels (for example, a RACH) for uplink synchronization and pilot channels (for example, a data pilot and a channel quality pilot). In addition, the aforementioned CAZAC sequence has been used to carry out the mixing.

Two types of procedures have been used for the CAZAC sequence; specifically, a first procedure to change one root index to another and use the changed root index, and a second procedure to perform a cyclic shift (CS) in a sequence of a single root and use the result of CS.

If a current root index is changed to a new root index, a low cross correlation occurs between the current root index and the new root index; however, there is no limitation in the design of sequence uses.

In the case of cyclic shift, there is a zero cross correlation between the current root index and the new root index, so that the two root indexes are used when each of the root indexes requires a high suppression ratio. Specifically, when time-frequency resources are shared in the same cell and data / control signals are transmitted, the two aforementioned root indices are adapted to discriminate between different signals or different UEs.

A representative example of CAZAC sequences is a Zadoff-Chu (ZC) sequence, and the Zadoff-Chu sequence can be defined by the following Equation 1:

[Equation 1]

(juπ (n +1) J

u () = exp for odd NZC

xn N

ZC

(juπn2 J

u () = exp for NZC pair

xn

NZC

wherein "n" is indicative of a sampling index, "NZC" is indicative of the length of the ZC sequence and "u" is indicative of the root index of the ZC sequence.

However, if the offset occurs in a frequency domain in the same manner as in the case where the CAZAC sequence is transmitted using the OFDM scheme, there may be excessive performance deterioration and an increase in the frequency of false alarms.

Specifically, if the cyclic shift (CS) is applied to the CAZAC sequence, frequency shift or temporal shift occurs excessively, so that it is difficult to discriminate between sequences.

LG Electronics: “RACH Design under Frequency Offset”, 3GPP Draft; R1-063162, 3rd Generation Partnership Project (3GPP), Mobile Competence Center; 650, Route des Lucioles; F-06921 Sophia-Antipolis Cedex; France, vol. 5 RAN WG1, no. 47, Riga, Latvia; November 1, 2006 (2006-11-01), XP050103617, considers various forms of RACH design under frequency offset. For this, the following procedures are analyzed: the use of a 0.5 ms RACH; the use of a 1.0 ms RACH; repetition of the preamble, in which a preamble is repeated within a RACH; use of a long sequence, in which, in one implementation, some CAZAC indexes use ZCZ sequences and some do not, or, in another implementation, a high-speed UE uses a special slot of

10 RACH with a long RACH period and a low speed UE uses another special RACH with a short RACH period.

Summary of the Invention

Accordingly, the present invention is directed to a method for establishing a cyclic shift (CS) considering a frequency shift that substantially obviates one or more problems due to limitations and disadvantages of the related art.

fifteen An object of the present invention is to provide a method for establishing a cyclic shift (CS) to prevent against a frequency shift so that it can easily prevent a sequence (for example, a CAZAC sequence) from deteriorating under the condition in which frequency shift occurs.

Advantages, objects and additional features of the invention will be set forth in part in the description that follows and in part will be apparent to those who have a normal mastery of the technique after studying what follows or what follows.

twenty which can be learned from the implementation of the invention. The objectives and other advantages of the invention can be realized and achieved by means of the structure indicated in particular in the written description and in the claims herein, as well as in the accompanying drawings.

The objects of the present invention are achieved by means of the materials of the independent claims.

According to an embodiment of the present invention, a method for establishing a

25 cyclic shift to be applied to a given sequence against an effect of a high Doppler frequency greater than a predetermined value. According to the method, the method comprises: acquiring a first variable (du) of a cyclic shift corresponding to a Doppler shift of a subcarrier separation using a root index (u) of the given sequence; acquire secondary variables comprising a number of groups (G) comprised in the given sequence, a length (S) of each group and a number (P) of

30 cyclic shifts per group using the first variable (du); and establish the cyclic offset value to be applied to the given sequence according to the secondary variables.

Preferably, the secondary variables further comprise a number of additional cyclic shifts that are applicable to the given sequence not based on the group (R).

Preferably, the given sequence is a Zadoff-Chu (ZC) sequence, and the first variable is acquired by an equation.

=

du

u − 1 mod NZC, 0 ≤ (u − 1 mod NZC) <NZC

2

- (u), NZC

2 ≤ (u − 1 mod NZC) <NZC

NZC

−1 mod NZC

where "u" indicates the root index of the ZC sequence and "NZC" corresponds to a length of the ZC sequence.

And, in this case, the secondary variables are acquired differently according to an interval of the first variable (du), and the interval of the first variable is divided by a criterion corresponding to 1/3 of the length of the given sequence ( NZC / 3).

40 And, if the interval of the first variable (du) is NCS du <(NZC / 3), the secondary variables are acquired by the equations

P =

du

NCS

S

S = 2 ⋅du + P⋅ NCS

G =

NZC

R = max ((NZC −2 ⋅du −G ⋅S)

NCS, 0)

in which "NCS" is a predetermined cyclic shift parameter, "P" corresponds to the number of cyclic shifts per group, "S" corresponds to the length of each group, "G" corresponds to the number of groups and "R" corresponds to the number of additional cyclic shifts.

On the other hand, if the interval of the first variable (du) is (NZC / 3) du (NZC-NCS) / 2, the secondary variables are acquired by the equations

P = du

NCS

S = NZC −2 ⋅du + P⋅NCS

G = du

S

R = min max (d − G ⋅S)

, 0, P

((

or

NCS))

in which "NCS" is a predetermined cyclic shift parameter, "P" corresponds to the number of cyclic shifts per group, "S" corresponds to the length of each group, "G" corresponds to the number of groups and "R" corresponds to the number of additional cyclic shifts.

And, preferably, said setting of the cyclic offset value (Cv) is performed as an equation

Cν = ⋅S ν

P + (νmod P) ⋅NCS, ν = 0.1, or, (PG R 1.)

⋅+−

And the given sequence can be to generate a random access preamble.

In another aspect of the invention, there is provided a method for establishing a cyclic offset to be applied to a given sequence, the method comprising: determining whether the cyclic offset has to be established according to restricted sets, restricted due to a Doppler offset; and establish the cyclic offset to be applied to the given sequence considered a cyclic offset corresponding to a Doppler offset of a subcarrier separation when it is determined that the cyclic offset is established according to the restricted sets.

Preferably, when it is determined that the cyclic displacement is established according to the restricted sets, said establishment of the cyclic displacement to be applied to the given sequence comprises: acquiring a first variable (du) indicating the cyclic displacement corresponding to a Doppler displacement of a separation of a subcarrier using a root index (u) of the given sequence; acquire secondary variables comprising a number of groups (G) comprised in the given sequence, a length (S) of each group, a number (P) of cyclic shifts per group using the first variable (du) and a number (R) of additional cyclic shifts that is applicable to the given sequence not based on the group and establish the cyclic shift that has to be applied to the given sequence according to the secondary variables.

Preferably, the given sequence is a Zadoff-Chu (ZC) sequence, and the first variable is acquired by an equation

−−

u1 mod NZC, 0 ≤ (u1 mod NZC) <NZC

2

du = -

−

NZC - (u1 mod NZC), NZC

two ≤ (u1 mod NZC) <NZC

where "u" indicates the root index of the ZC sequence and "NZC" corresponds to a length of the ZC sequence.

And the secondary variables are acquired differently according to an interval of the first variable (du), and the interval of the first variable is divided by a criterion corresponding to 1/3 of the length of the given sequence (NZC / 3).

In this case, if the interval of the first variable (du) is NCS du <(NZC / 3), the secondary variables are acquired by the equations

P = du

NCS

S = 2 ⋅du + P⋅ NCS

G = NZC

S

R = max ((NZC - 2 ⋅du - G ⋅S)

NCS, 0)

in which "NCS" is a predetermined cyclic shift parameter, "P" corresponds to the number of cyclic shifts per group, "S" corresponds to the length of each group, "G" corresponds to the number of groups and "R" corresponds to the number of additional cyclic shifts.

On the other hand, if the interval of the first variable (du) is (NZC / 3) du (NZC-NCS) / 2, the secondary variables are acquired by the equations

P = (NZC - 2 ⋅du)

NCS

S = NZC - 2 ⋅du + P⋅ NCS

G = du

S

R = min max ((u

(d− G ⋅S)

NCS, 0), P)

in which "NCS" is a predetermined cyclic shift parameter, "P" corresponds to the number of cyclic shifts per group, "S" corresponds to the length of each group, "G" corresponds to the number of groups and "R" corresponds to the number of additional cyclic shifts.

And, preferably, the cyclic shift (Cv) is carried out according to the following equation:

ν⋅ NCS, ν = 0.1, or, (NZC

NCS −1,) for unrestricted sets

Cν = S ⋅ν

P + (ν mod P) ⋅ NCS, ν = 0.1, or, (PG + R −1,) for constrained sets.

⋅

10 And the given sequence can be to generate a random access preamble.

In another aspect of the present invention, there is provided a method for establishing a cyclic shift that has to be applied to a given sequence, the method comprising: (a) acquiring a variable du by means of an equation

u − 1 mod NZC, 0 ≤ (u − 1 mod NZC) <NZC

2

= du NZC - (u − 1 mod NZC), NZC

2 ≤ (u − 1 mod NZC) <NZC

wherein "u" indicates a root index of the given sequence and "NZC" corresponds to a given sequence length; (b) 15 acquire variables G, S, P and R through the equations

P = du

NCS

S = 2 ⋅du + P⋅ NCS

G = NZC

S

R = max ((NZC - 2 ⋅du - G ⋅S)

NCS, 0)

when the interval of the first variable (du) is NCS du <(NZC / 3), and acquire variables G, S, P and R through the equations

P = (NZC - 2 ⋅du)

NCS

S = NZC - 2 ⋅du + P⋅ NCS

G = du

S

R = min max (((d− G ⋅S)

, 0), P)

or

NCS

when the interval of the first variable (du) is (NZC / 3) du (NZC-NCS) / 2, in which "NCS" is a predetermined cyclic shift parameter; (c) establish the cyclic shift (Cv) through the equation

ν⋅ NCS, ν = 0.1, or, (NZC

NCS −1,) for unrestricted sets

C = ν S ⋅ν

ν (⋅ 1,

P + (mod P) ⋅ N, ν = 0.1, or, PG + R -) for sets with restriction

CS

wherein the restricted sets are sets of restricted cyclic shifts due to a Doppler shift, and the unrestricted sets are sets of unrestricted cyclic shifts due to a Doppler shift.

In another aspect of the present invention, there is provided a method for transmitting a random access preamble using cyclic shift, the method comprising: acquiring, from system information, a root index (u) of a sequence for the access preamble random; establish the cyclic offset to be applied to the sequence, in said establishment, when it is determined that the cyclic offset is established according to the constrained sets due to a Doppler offset, the cyclic offset to be applied to the sequence is established considering a cyclic offset corresponding to a Doppler offset of a subcarrier separation; generate the sequence according to the root index (u) with the established cyclic shift; and transmit the sequence with the cyclic shift as the random access preamble.

Preferably, when it is determined that the cyclic displacement is established according to the restricted sets, said establishment of the cyclic displacement to be applied to the sequence comprises: acquiring a first variable (du) indicating the cyclic displacement corresponding to the Doppler displacement of a separation of a subcarrier using the root index (u) of the given sequence; acquire secondary variables comprising a number of groups (G) comprised in the sequence, a length (S) of each group, a number (P) of cyclic shifts per group using the first variable (du) and a number (R) of Additional cyclic shifts that are applicable to the given sequence are not based on the group and set the cyclic shifting to be applied to the sequence according to the secondary variables.

Preferably, the given sequence is a Zadoff-Chu (ZC) sequence, and the first variable is acquired by an equation

u − 1 mod NZC, 0 ≤ (u − 1 mod NZC) <NZC

2

du = NZC - (u − 1 mod NZC), NZC

2 ≤ (u − 1 mod NZC) <NZC

where "u" indicates the root index of the ZC sequence and "NZC" corresponds to a length of the ZC sequence.

Preferably, the secondary variables are acquired differently according to an interval of the first variable (du), and the interval of the first variable is divided by a criterion corresponding to 1/3 of the length of the given sequence (NZC / 3) .

More specifically, if the interval of the first variable (du) is NCS du <(NZC / 3), the secondary variables can be acquired by the equations

P = du

NCS

S = 2 ⋅du + P⋅ NCS

G = NZC

S

R = max ((NZC - 2 ⋅du - G ⋅S)

NCS, 0)

in which "NCS" is a predetermined cyclic shift parameter, "P" corresponds to the number of cyclic shifts per group, "S" corresponds to the length of each group, "G" corresponds to the number of groups and "R" corresponds to the number of additional cyclic shifts.

On the other hand, if the interval of the first variable (du) is (NZC / 3) du (NZC-NCS) / 2, the secondary variables are acquired by the equations

P = (NZC - 2 ⋅du)

NCS

S = NZC - 2 ⋅du + P⋅ NCS

G = du

S

R = min max ((u

(d− G ⋅S)

NCS, 0), P)

in which "NCS" is a predetermined cyclic shift parameter, "P" corresponds to the number of cyclic shifts per group, "S" corresponds to the length of each group, "G" corresponds to the number of groups and "R" corresponds to the number of additional cyclic shifts.

And, preferably, the cyclic shift (Cv) is performed as the following equation:

ν⋅ NCS, ν = 0.1, or, (NZC

NCS −1,) for unrestricted sets

Cν = S ⋅ν

P + (ν mod P) ⋅ NCS, ν = 0.1, or, (PG + R −1,) for constrained sets.

⋅

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 additional explanation of the invention as claimed.

The present invention can easily establish a cyclic shift interval (CS) at a specific location that has no overlap considering a channel response of a reception sequence (Rx) and an overlap location of this reception sequence (Rx), even if a reception signal (Rx) is displaced by a frequency offset regardless of categories of a domain that generates a sequence, so you can greatly reduce the number of detection errors and the frequency of false alarms.

And, if a cyclic shift (CS) sequence is assigned to a cell that has a frequency offset of more than a predetermined level, the present invention can minimize the influence of a frequency shift in a high mobility cell.

Brief description of the drawings

The accompanying drawings, which are included to provide a further understanding of the invention, illustrate embodiments of the invention and, together with the description, serve to explain the principle of the invention.

In the drawings:

FIG. 1 is a conceptual diagram illustrating the influence of a frequency shift caused by a pulse conformation in a frequency domain when a correlation is established between a sequence and a subcarrier according to the present invention; FIG. 2 is a conceptual diagram illustrating different situations of frequency shifting existing in a plurality of cells according to the present invention; FIG. 3 it is a conceptual diagram illustrating a sequence assignment procedure when a sequence is a CAZAC sequence according to the present invention;

FIG. 4 is a conceptual diagram illustrating overlaps that occur in a temporal domain channel response of a reception sequence due to frequency offset according to the present invention; FIG. 5 it is a conceptual diagram illustrating a procedure for establishing an application unit of a new cyclic shift (CS) by adding an additional margin to an application unit of an old CS according to the present invention; FIGURES 6 and 7 are conceptual diagrams illustrating examples of application of the additional margin of FIG. 5 with the proviso that a sequence index is low according to the present invention; FIGURES 8 and 9 are conceptual diagrams illustrating exemplary additional margins of FIG. 5 with the proviso that a sequence index is high according to the present invention; FIG. 10 shows an example of a single group consisting of P sets of cyclic shifts according to the present invention; FIG. eleven it is a conceptual diagram illustrating a procedure for establishing a group of application of a cyclic shift (CS) and a range of application of a CS of each group according to the present invention; FIG. 12 shows locations where pulses occur due to interference when the CAZAC index is contained in the range of N / 3 ~ N / 2 according to the present invention; FIG. 13 it is a flow chart illustrating a restricted set of cyclic shifts according to an embodiment of the present invention; FIG. 14 is a conceptual diagram illustrating a procedure for establishing a variable (du) of a cyclic shift corresponding to the Doppler shift associated with the separation of 1 subcarrier when the restricted set of cyclic shifts is established according to the present invention; FIG. fifteen it is a conceptual diagram illustrating a specific case in which the variable (du) is smaller than a basic unit NCS to which the cyclic shift (CS) according to the present invention is applied; FIG. 16 is a conceptual diagram illustrating a procedure for calculating a variable that sets the cyclic shift within the NCS du <(NZC / 3) interval according to the present invention; FIG. 17 it is a conceptual diagram illustrating a procedure for calculating a variable that establishes the cyclic shift within the interval (NZC / 3) du <(NZC - NCS) / 2 according to the present invention; FIGURES 18 and 19 are conceptual diagrams illustrating a procedure for reducing the number of ZCZ preamble sequences due to an overlap response in the case of NZC = 839, NCS = 100 and du = 155 according to the present invention; FIG. twenty it is a conceptual diagram illustrating the increasing proportion of a restricted cyclic displacement available after the restriction of a cyclic displacement start site is eliminated in the case of NZC = 839 according to the present invention; FIG. 21 is a conceptual diagram illustrating an exemplary cyclic shift in the case of NZC = 839, NCS = 40 and du = 150 according to an embodiment of the present invention; FIG. 22 it is a conceptual diagram illustrating an exemplary cyclic shift in the case of NZC = 839, NCS = 40 and du = 399 according to an embodiment of the present invention; FIG. 23 is a conceptual diagram illustrating an exemplary cyclic shift in the case of NZC = 839, NCS = 40 and du = 150 according to another embodiment of the present invention; and FIG. 24 it is a conceptual diagram illustrating an exemplary cyclic shift in the case of NZC = 839, NCS = 40 and du = 399 according to another embodiment of the present invention.

Detailed description of 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. Whenever possible, the same reference numbers from beginning to end of the drawings will be used to refer to identical or similar parts.

Before describing the present invention, it should be noted that most of the terms disclosed in the present invention correspond to the general terms well known in the art, but the applicant has selected some terms as necessary and will be disclosed in hereinafter in the following description of the present invention. Therefore, it is preferable that the terms defined by the applicant are understood based on their meanings in the present invention.

For the sake of the description and a better understanding of the present invention, general structures and devices well known in the art will be omitted or denoted by a block diagram or a flow chart. Whenever possible, the same reference numbers from beginning to end of the drawings will be used to refer to identical or similar parts.

The present invention provides a method of configuring a cyclic shift (CS) to prevent against frequency shift, so that it can easily prevent the performance of a sequence from deteriorating (i.e., the CAZAC sequence). To this end, the present invention will disclose the procedure for the application of the cyclic shift to the CAZAC sequence and the influence of the frequency shift of the CAZAC sequence.

The cyclic shift can be applied to the CAZAC sequence according to two schemes; specifically, a first scheme to carry out the cyclic shift in the sequence and a procedure to multiply an exponential function of other areas by a time or frequency domain sequence and carry out the cyclic shift in the multiplied result.

The cyclic shift "d" is applied to the frequency index "k" in the frequency domain. If the sequence index M and the sequence of length N are represented by c (k; d, M, N), a procedure for carrying out the cyclic shift in the sequence can then be represented by Equation 2:

[Equation 2]

ck (; d, M, N) = c (mod (k −d, N); M, N)

wherein "d" is indicative of the amount of cyclic shift and "mod" is indicative of a module operator.

A procedure to apply the cyclic shift by multiplying an exponential function by the sequence can then be represented by Equation 3:

[Equation 3]

(j 2πdk

ck (; d, M, N) = f (mod (k −d, N); M, N) = exp (()

J FFT ck; d, M, N) N

On the other hand, although each of the above Equations 2 and 3 shows an exemplary cyclic shift applied in the frequency domain, the cyclic shift can be applied in the sampling index "n" of the temporal domain sequence in the temporal domain. In this case, an example of the application of cyclic displacement can then be represented by Equation 4:

[Equation 4]

xn = x (n + C)

() (mod NZC)

u vu v

in which "Cv" is indicative of the degree of cyclic shift, "n" is indicative of a sampling rate, "NZC" is indicative of the length of the sequence ZC and "u" is indicative of a root index of the sequence ZC

CAZAC sequences can be distinguished from each other on the condition that different root indices are used; however, it should be noted that a difference in cross correlation occurs between the CAZAC sequences.

However, in the case of at least two CAZAC sequences associated with the cyclic shift, the cross-correlation value between the CAZAC sequences is zero, so that the aforementioned CAZAC sequences are used when a high suppression ratio is required for two CAZAC sequences.

Specifically, the CAZAC sequences associated with cyclic shift share the time-frequency resources within the same cell, so that they can be used to discriminate between different signals / UEs during the transmission of data / control signals.

However, if the frequency shift occurs in the frequency domain in the same manner as in the case where the CAZAC sequence is transmitted using the OFDM scheme, the present invention may find an excessive deterioration in performance and in the frequency of false alarms.

The following description will disclose an example in which the sequence is transmitted in the frequency domain, and another example in which the sequence is transmitted using the OFDM scheme.

FIG. 1 is a conceptual diagram illustrating the influence of a frequency shift caused by a pulse conformation in a frequency domain when a correlation is established between a sequence and a subcarrier according to the present invention.

As shown in FIG. 1, a correlation of each sequence sample with the subcarrier is established. If a receiving end carries out the sampling of the signal due to the frequency shift, as denoted by the "Interference" site, the signals from the neighboring subcarriers are mixed within a single sample. In other words, if the pulse-forming function is p (x), the response of an arbitrary subcarrier can then be represented by Equation 5:

[Equation 5]

N − 1 () = pkw0 -) ()

rk, fdes L (nw0 + fdes cn

n = 0

wherein "r (k, fdes)" is indicative of a frequency reception (Rx) response at the sub-carrier location k-th if the frequency offset is fdes, "c (n)" is indicative of a CAZAC sequence correlated with the subcarrier by the user equipment (UE), "p (f)" is indicative of a pulse-forming function in a frequency domain and W0 is indicative of a subcarrier separation.

In the case that fdes = 0, the previous Equation 5 produces only the value c (k). If not, in the case that fdes = 0, the signal from the neighboring subcarrier may enter the receiving end, so that a deterioration in performance arises. Due to the performance deterioration caused by the frequency shift, the probability of finding a detection error at the receiving end increases, and the frequency of false alarms and / or defective detection can inevitably increase at the receiving end.

Specifically, since the cyclic shift is applied in the temporal domain and the CAZAC sequence is transmitted within the frequency domain, it may not discriminate between various sequences. And the problem just mentioned can occur even in a situation where the CAZAC sequence is transmitted within the temporal domain as a form of temporal displacement.

In other words, if the frequency shift or temporal shift occurs, the procedures for the use of cyclic shift must inevitably undergo performance deterioration. In addition, the influence of frequency offset applies equally to a specific case in which the cyclic shift is applied in the temporal domain, as denoted by Equation 4.

Therefore, a technology must be developed again to prevent the sequence performance (ie, from the CAZAC sequence) from deteriorating in the condition in which the frequency shift occurs.

Specifically, in the case of applying the cyclic shift to the CAZAC sequence, frequency offset or temporal shift occurs in excess, so that the present invention has difficulty discriminating between sequences when the frequency or temporal offset occupies at least half of a separation of a single subcarrier.

However, the degree of frequency shift and the degree of Doppler shift may be different in different cells of a cellular mobile communication system.

Therefore, according to one embodiment, the present invention provides different cyclic shift (CS) configuration procedures according to the degree of the frequency deviations of the individual cells, and hereafter a detailed description thereof will be presented.

FIG. 2 is a conceptual diagram illustrating different situations of frequency shifting existing in a plurality of cells according to the present invention.

With reference to FIG. 2, the present invention can determine that a specific cell that has many highly mobile UEs in a cellular mobile communication system that includes many cells has a high frequency offset. It is very likely that a UE contained in a cell that includes residential neighborhoods can be a low speed UE, so that the frequency shift within the cell can be low.

In more detail, FIG. 2 shows cells A and B adjacent to a high-speed rail, and cell C, distant from the high-speed rail.

In the case of cells A and B adjacent to the high-speed rail, there is a high probability that a plurality of high-speed UEs are contained in a corresponding cell, so that the present invention has an advantage, because a sequence can be assigned which is very resistant to frequency shift.

For example, in the case of cell C adjacent to the residential neighborhood distant from the high-speed rail, the probability of including the high-speed UE in a corresponding cell is relatively low, so there is no need to assign only the sequence which is very resistant to frequency shift.

In the case of the available sequence (for example, the CAZAC sequence), the first sequences caused by the root indices of the individual sequences and the second sequences caused by the cyclic shift applied to the first sequences may have different frequency shift characteristics.

Therefore, the present invention establishes the restricted case and the unrestricted case, and provides the cyclic displacement configuration procedures for individual cases.

The restricted case indicates that the influence of the Doppler shift is greater than a predetermined threshold value, so that a limitation occurs in the procedure for establishing a range of application of the cyclic shift (CS).

The unrestricted case indicates that the influence of the Doppler shift is equal to or less than a predetermined threshold value, so there is no limitation in the procedure for establishing a range of application of the CS.

The procedure for establishing the cyclic shift will be described in detail hereafter.

FIG. 3 is a conceptual diagram illustrating a sequence assignment procedure when a sequence is a CAZAC sequence according to the present invention.

The CAZAC sequence may include a root sequence of each root CAZAC sequence and a zero correlation zone (ZCZ) sequence to which different cyclic shifts (also called circular shifts) are applied.

In more detail, FIG. 3 shows the root sequence for each root index in Nt root indexes and the sequence set ZCZ to which L cyclic shifts are applied to each root sequence.

In this case, the ZCZ is indicative of a cyclic shift application interval to which the cyclic shift (CS) is applied, so that node B is able to discriminate between RACH signals.

On the other hand, if the CAZAC sequence is used when the frequency shift exists, the present invention may have difficulty discriminating between ZCZ sequences by frequency offset. Therefore, the present invention may determine that the ZCZ sequence is not used in a predetermined cell that has a frequency offset of more than a predetermined level.

In this way, the threshold value used to decide the degree of frequency offset of each cell can be duly decided according to the number of available sequences of a corresponding system and the degree of frequency offset of each cell.

If it is determined that the cell has the frequency offset of more than the predetermined level, the probability of containing in this cell the high speed UE is very high, as shown in cells A or B.

However, if it is determined that the ZCZ sequence is not used in the cell that has the frequency offset of more than the predetermined level, there can only be Nt indices based on the CAZAC indices, so that the number of available sequences becomes smaller .

If a sequence reuse coefficient becomes smaller, sequences should be assigned according to cell planning. However, this allocation based on cell planning may unexpectedly increase the complexity in the procedure for assigning the sequences to individual cells, so that another solution may also be required, provided that the number of available sequences finds the problem.

In addition, if using only Nt sequences and not using the ZCZ sequence, there may be a problem in estimating the round trip delay or the one-way trip delay while improving the sequence performance of detection. That is, there may be the problem of distinguishing the position of the correlation peak that varies due to the round trip delay or the one-way trip delay and a correlation peak position that varies due to the frequency offset. Therefore, another solution against this problem may also be required.

On the other hand, the aforementioned problem of having difficulty discriminating between ZCZ sequences due to frequency shift is intensified by the condition that the CAZAC index is very high

or that is not very low.

In more detail, if "k" is indicative of a frequency domain index, "N" is indicative of the length of the CAZAC sequence, "M" is indicative of a CAZAC sequence and a transmission signal (Tx) is indicative of "C (k, N, M)", a reception signal (Rx) may then be represented by Equation 6:

[Equation 6]

(2π M⋅d J

Rk (, N, M) = cK, N, M) ⋅exp -

(⋅ k

N

wherein "d" is indicative of the amount of delay in the frequency domain caused by the frequency offset.

As can be seen in Equation 6, if the CAZAC “M” index has a very low value, or if the CAZAC “M” index has the highest value among a total of Nt sequence indices, the influence of the exponential function caused by frequency shifting it is gradually reduced, so that the influence of frequency shifting on the Rx signal is gradually reduced.

If the CAZAC sequence is assigned to the cell that has the frequency offset of more than the predetermined level, the present invention can assign only the root sequence. In the case of using the CAZAC ZCZ sequence due to the insufficient number of root sequences, the present invention may allow the CAZAC sequence to employ a specific sequence that is in a predetermined initial range or in the last predetermined range of the total indexes. In this case, it should be noted that the term "predetermined interval" can be set in different ways depending on the detection performance of the system.

In the case of comparing the aforementioned procedure with the other procedure to allow the ZCZ sequence not to be used in the cell having the high frequency offset, the aforementioned procedure increases the categories or types of available sequences, so there is almost no need to carry out cell planning.

In more detail, if the number of total CAZAC sequences is Nt, as shown in FIG. 3, the sequence to be used in the cell with the high frequency offset can be set to CAZAC indices 0, 1, 2, Nt-2, Nt-1 and Nt.

On the other hand, in the case of using the CAZAC sequence for the cell that has the frequency offset of more than the predetermined level, there is no need to use only indices other than the CAZAC indices just mentioned 0, 1, 2, Nt- 1, Nt-2 and Nt. To reduce interference between the aforementioned CAZAC sequence and the other sequence used for the cell that has the high frequency offset, the present invention may not use the frequency index used for the cell that has the high frequency offset as necessary, resulting in the implementation of high efficiency.

On the other hand, in the case of using the ZCZ sequence to guarantee the number of sequences available in the cell having the high frequency offset and / or to guarantee the performance of the estimation of the time delay occurred in the channel, the present invention sets the cyclic shift interval in the restricted case in consideration of the overlap (that is, the Doppler shift) caused by the frequency offset. Accordingly, the present invention avoids performance deterioration caused by frequency shifting, and a detailed description thereof will be described hereafter.

If the presence of the frequency shift is decided, the frequency response of the Rx signal can be represented by the above Equation 6.

On the other hand, Equation 6 shows that a signal value is transferred from all neighboring subcarriers due to frequency offset. However, in fact, you can establish a specific component that greatly affects the response of the Rx signal channel to a part located on both sides of a corresponding subcarrier, in which the part receives a signal from the neighboring subcarrier.

Therefore, in the case of considering only the first order case, Equation 6 can be represented by three terms, as shown below, in Equation 7:

[Equation 7]

rk, f = p − f − fck − 1 + p − fck + pf − fck + 1

(off) (0 off) () (off) () (off) ()

On the other hand, the receiving end applies a conjugate complex number c (n) to the signal Rx, so that the result of the application can then be represented by Equation 8:

[Equation 8]

(j2πMkJ (j2πMk (+1) J

*

rk (, fc () = α0 + α1 exp - + α

) k1 exp

des -

NN

The pulse-forming function of Equation 7 can easily be denoted by an exponential function of cosine or sine.

For the sake of the description, the pulse-forming function is represented by constants 00, 0-1 and 01.

With reference to Equation 8, the response of the Rx signal channel occurs at three points; specifically, "t", indicative of a target position in the temporal domain, "tM", indicative of a position shifted to the left side, and "t + M", indicative of a position shifted to the right side. It can be recognized that the response of the channel generated in the offset position in M ​​based on the right / left sides corresponds to the overlap of the Rx signal, that is, the Doppler shift component that has the separation of 1 subcarrier.

In FIG. 4 the phenomenon mentioned above is shown in which the overlap occurs in the response of the channel due to the frequency shift.

FIG. 4 is a conceptual diagram illustrating overlaps that occur in a temporal domain channel response of a reception sequence due to frequency offset according to the present invention.

If the cyclic shift is applied to a sequence used in a specific cell that has a high frequency offset of more than a predetermined level, a single channel response occurs at the target position in the Rx channel response of the corresponding sequence, and two additional overlaps may occur in the response of the Rx channel of the corresponding sequence according to the Doppler displacement dimensioned to the separation of 1 subcarrier.

Therefore, if the CS application interval is established regardless of the target position and the overlap positions, an unexpected overlap occurs between the channel response and the Rx sequence overlap due to delay propagation. of the channel and the propagation delay, so that there may be confusion between the position of the target and the position of the overlaps between different sequences of CS application.

Consequently, if the restricted case is decided when the CS application interval is established in the CAZAC sequence, the present invention considers the overlap generated in the channel response, so that it establishes the CS application interval during a specific period. in which the response of the channel of the Rx sequence does not overlap with the overlap of the response of the previous channel.

FIG. 4 shows an exemplary case in which the overlap of size M (where M = sequence index) occurs when the CAZAC sequence is generated in a frequency domain. however, if the CAZAC sequence is generated in the temporal domain, the overlap generation position caused by the Doppler shift of the subcarrier separation can be determined in different ways.

In the following, all the CS application cases used for the individual domains will be described in detail.

For the sake of the description and a better understanding of the present invention, FIGURES 5-11 assume that the cyclic displacement unit is fixed at T0.

FIG. 5 is a conceptual diagram illustrating a procedure for establishing an application unit of a new cyclic shift (CS) by adding an additional margin to an application unit of an old CS according to the present invention.

The present invention generates a cyclic displacement preamble according to said based on the RACH component. However, in the environment where the OFDM frequency offset exists, the receiving end of the present invention can easily confuse a normal sequence with another sequence.

To prevent the aforementioned problem from being generated, the present invention may use an additional range of cyclic displacement, as shown in FIG. 5.

With reference to FIG. 5, the propagation of the delay is indicative of a propagation of the channel delay, and the round trip delay (RTD) is indicative of a continuity time of the propagation of a physical distance between the user equipment (UE) and node B. In the case of using the additional cyclic offset range, the present invention adjusts the size of the margin for each sequence, so that it can reduce the influence of the frequency offset when the sequence is used.

In the case of implementing the frequency offset using the additional margin, the CAZAC sequence function decides the cyclic shift unit. In other words, in association with the sequence CAZAC "M", the cyclic shift unit is then represented by Equation 9:

[Equation 9]

()

TM () = T0 + Tmargen M

in which T0 is indicative of a common unit of cyclic displacement irrespective of the sequence index, and Tmargen (M) is indicative of an additional margin used when the sequence index is M. This margin can be decided by other procedures according to Sequence uses and cyclic shift.

Therefore, although it is preferable that the cyclic displacement unit is at least 2M, this additional margin may be changed to another margin according to the CS application area. The situation mentioned above is shown in FIGURES 6 and 7.

FIGURES 6 and 7 are conceptual diagrams illustrating examples of application of the additional margin of FIG. 5 with the proviso that a sequence index is low according to the present invention.

Here, in the case of Fig. 6, the interval of M due to the frequency shift is smaller than the cyclic shift interval of T0. Even when this interval is used, we can avoid the problem of overlapping with other sequences. However, there may be a problem in estimating the information due to the time delay of the transmitted sequence. Therefore, in an embodiment of this invention, it is preferable not to use this interval in which the range of M due to the frequency shift is less than the cyclic shift range of T0. However, there may be a system that uses this interval according to the system requirement.

The part marked with oblique lines of FIGURES 6 and 7 indicates the opportunity for cyclic displacement.

If the signal that has no influence on the frequency offset is "t", the pulse affected by the frequency offset may occur at a single point on the left loop and may occur at a single point on the right side. If the signal includes T0 used as a basic cyclic shift unit, Tmargen (M) can be set to 2M.

The additional margin is applied to all indices, so that the present invention can define cyclic displacement that is highly resistant to frequency / temporal deviations.

However, the higher the sequence index, the higher the Tmargen value (M). Consequently, the number of cyclic displacements available is reduced to "1". To avoid the reduction of cyclic shifts, the present invention will disclose in detail the case of the high CAZAC index.

FIGURES 8 and 9 are conceptual diagrams illustrating exemplary additional margins of FIG. 5 with the proviso that a sequence index is high according to the present invention.

FIG. 8 shows the case in which the CAZAC index "M" is 2Q0 ~ 3T0, and FIG. 9 shows the case in which the CAZAC index "M" is 3Q0 ~ 4T0. Although the case of FIG. 8 considers the basic unit of cyclic displacement, the set of cyclic displacements denoted by the part marked with oblique lines can be additionally inserted in the intermediate space. The case of FIG. 9 It has a wider space, so that at least two cyclic shifts can be inserted into this larger space.

FIG. 10 shows an example of a single group composed of P sets of cyclic shifts according to the present invention.

With reference to FIG. 10, if the explanation mentioned above is generalized, the slots denoted by the parts marked with oblique lines are defined in the 3M interval, in which the block is constructed by pulses and the interval M is PT0 ~ (P + 1) T0 , it can be recognized that P sets of cyclic shifts are constructed.

For the sake of the description, the unit 3M or 2M + PT0 will be referred to hereinafter as a group of cyclic displacements. A specific sequence to which the cyclic shift is applied includes a predetermined number of cyclic shift groups. The predetermined number of cyclic shifting groups can be applied to each cyclic shifting group, so that the predetermined number of cyclic shifts can be applied to the cyclic shifting component caused by the Doppler shifting.

FIG. 11 is a conceptual diagram illustrating a procedure for establishing a cyclic shift (CS) application group and an application range of a CS of each group according to the present invention.

With reference to FIG. 11, the units of the cyclic shifting groups can be defined in total sequences, and each cyclic shifting group can be defined as shown in FIG. 10. Since the number of cyclic shifting groups is G and the number of cyclic shifts for each group is P, the total number of cyclic shifts available is P * G. As shown in FIG. 11, according to an embodiment of the present invention, it is assumed that the sequence is divided into groups and that each group seeks a restricted cyclic shift available in each group.

In the case of using the scheme mentioned above, all available cyclic shifts are defined in the index range in which the number of cyclic shifting groups is "1". If the length of

sequence is N, this interval that has the sequence length of N corresponds to the indices that oscillate between 1

<dl><dt>-</dt><dd> N / 3 and 2N / 3 - N-1. In this case, the k-th index has the same group of cyclic shifts as that of the (Nk) -th index and the set of cyclic shifts.</dd></dl>

FIG. 12 shows locations where pulses occur due to interference when the CAZAC index is contained in the range of N / 3 ~ N / 2 according to the present invention.

A single square of FIG. 12 indicates the cyclic displacement unit. If the CAZAC index is greater than "N / 3", not all consecutive cyclic shifting positions (ie, cyclic shifting positions defined by T0) can be used, and can be used according to predetermined rules.

Hereinafter, a procedure for establishing the restricted set of cyclic shifts according to an embodiment of the present invention will be described.

FIG. 13 is a flow chart illustrating a restricted set of cyclic shifts according to an embodiment of the present invention.

With reference to FIG. 13, if the restricted set of cyclic shifts is established in a cell that has the frequency offset of more than a predetermined threshold value, the present invention provides a method for establishing the cyclic shift in consideration of the overlap, so that there is no confusion some between a desired channel response and this overlap.

To this end, as shown in step S1301 of FIG. 13, the present invention provides a distance "du" between the response generated by the Doppler shift and a desired channel response using a root index "u" of the given sequence. In this case, the previous distance corresponds to the cyclic shift generated by the Doppler offset corresponding to the separation of 1 subcarrier.

In the following, a detailed description of the variable "du" will be described in detail.

FIG. 14 is a conceptual diagram illustrating a procedure for establishing a variable (du) of a cyclic shift corresponding to the Doppler shift associated with the separation of 1 subcarrier when the restricted set of cyclic shifts is established according to the present invention.

With reference to FIG. 14 (a), if there is no influence of the Doppler frequency, the peak position generated by the correlation operation of the receiving end is denoted by "1401". Due to the propagation of the delay and the round trip delay (RTD), the peak position at the receiving end appears in the cyclic displacement unit NCS (1402) used as the cyclic displacement unit basically decided by the system.

On the other hand, in the case that the presence of the Doppler frequency corresponds to the separation of 1 subcarrier, the peak position caused by the correlation operation of the receiving end is determined according to the sequence indices.

According to the present invention, the distance between the peak position based on the Doppler shift corresponding to the separation 1 of 1 subcarrier and the ideal peak position is called "du".

In other words, FIG. 14 (b) shows the offset of the reception end channel response caused by the Doppler -lƒ frequency. FIG. 14 (c) shows the offset of the reception end channel response caused by the Doppler + lƒ frequency. Based on the aforementioned fact, the value "du" can be considered as the cyclic shift caused by the Doppler shift.

If the restricted cyclic shift is established in consideration of the cyclic shift corresponding to the Doppler shift of the 1 subcarrier separation, the present invention controls that the established restricted cyclic shift does not overlap with the response movement of the channel caused by the Doppler shift.

The present invention excludes reserved areas marked "reserved" in FIGURES 14 (a) and 14 (b) of the established cyclic shift interval, so that it can prevent unexpected confusion between channel responses, even if it occurred the relatively high Doppler shift.

With reference again to FIG. 13, the present invention acquires secondary variables using the acquired variable "du" from the previous step S1301 in step S1302. Specifically, the present invention acquires from the current sequences (for example, ZC sequences) the number (G) of cyclic shift groups, the number

(P) of cyclic shifts applicable to each group and the length (S) of each group.

The secondary variables mentioned above must be established differently according to sequence indices, because the length of the group is changed to another according to the sequence indices. And the variable

"Du" depends on the sequence index, so that the present invention provides a method for establishing secondary variables according to the range of the variable "du".

In addition, the present invention can apply not only the above group-based cyclic shift, but also an additional cyclic shift using a specific area that is not contained in the group of cyclic shifts within the sequence range, and a detailed description of the Same will be described below.

Then, in step S1303, the present invention establishes the cyclic shift using the secondary variables acquired from step S1302.

The mathematical relationship between the detailed variables for the application of cyclic displacement will be described in detail.

The restricted cyclic shift according to the present invention has been proposed to prevent the high Doppler frequency effect from being generated.

In the following, the other displacement “Cdes” of cyclic displacement different from the variable “du” will be described in detail.

The value "CDs" indicates the degree of a shift generated by the Doppler shift.

If the degree of displacement generated by the Doppler displacement is less than half the range of the given sequence, this degree of displacement may have the same meaning as that of the variable du. If not, if the degree of displacement generated by the Doppler displacement is equal to or greater than half the range of the given sequence, the resulting value acquired when the “Cdes” value is subtracted from the total length of the sequence may correspond to the variable du.

The value "CDs" depends on the root index of the sequence used. The preamble can be generated from either the temporary domain or the frequency domain. The relationship between the values ​​"CDs" and "u" depends on the domain generated by the preamble.

If the ZC sequence is generated from the frequency domain and the cyclic shift is applied in the time domain, the present invention may induce the "Cdes" value to use the following procedure, and a detailed description of the following will be described below. same.

It is assumed that the signal energy is propagated by the value transferred from the neighboring subcarrier according to the Doppler frequency. And it is assumed that the transfer from the neighboring carrier occurs only at the position of the subcarrier separated from a current subcarrier by a blank space, and this case is called a first-order case. In this case, the Rx signal on the specific subcarrier is composed of three terms shown, then in Equation 10:

[Equation 10]

sn () = p − fcn + p − w − f () + pw − fcn (+)

(off) () (0 off) cn − 1 (0 off) 1

in which the "p (f)" impulse formation function can be denoted by an exponential cosine or sine function. For the convenience of the description, if the constants c0, c-1, and c1 are set, the value s (n) may be denoted by s (n) = c0c (n) + c-1c (n-1 ) + c1c (n + 1). For the sake of the description, if the sequence conjugate is multiplied by the resulting s (n) value, the following Equation 11 can be acquired:

[Equation 11]

** **

snc () = c () (c () + c (+ 1cn (+1 = 0 + c1cn − 1) cn + 1cn (+1 c ()

() nn0cn1cn − 1) c)) c (() c) n

−−

in Equation 11, if “c (n) = x (n)” is denoted by CAZAC, c (n-1) c * (n) may be represented by the following Equation 12:

[Equation 12]

(j2πun J

* (-) () exp -

xn 1 xn =

NZC

Here, "u" indicates the root index, and "NZC" indicates the length of the sequence.

If Equation 12 is applied to Equation 11, it can be recognized that "s (n)" is composed of three signals. A first term of the value "s (n)" is indicative of a simple DC component, a second term is indicative of a complex exponential wave that has the frequency of u / NZC, and the third term is indicative of a complex exponential wave which has the frequency of -u / NZC.

Therefore, the value "CDs" can be represented by Equation 13 below:

[Equation 13]

= u

CDs, u

On the other hand, if the ZC sequence is generated from the temporary domain and the cyclic shift is generated from the temporary domain, the "Cdes" value can be calculated by means of the following procedure.

If the RACH preamble received without having the frequency offset is set to ar (n), the RACH signal received with the frequency offset may be represented by Equation 14 below:

[Equation 14]

rn =

f () ejΔwnrn ()

in which lW is denoted by lw = 2nlƒ / ƒs, and lƒ indicates the frequency offset denoted by the unit of hertz (Hz), and fs is indicative of a sampling frequency of the RACH preamble.

The autocorrelation of the value f () can be calculated by the equation “r (n) - xu (n)”, in which “u” is indicative of

rn

ZC sequence index.

[Equation 15]

NZC −1 NZC −1 NZC −1

* jΔwn j 2π (Δff s) n

c0 = L rnx () = L e = L e

() f () n

ru n = 0 n = 0 n = 0

In Equation 15, if “Cds, u” is indicative of the range of a frequency offset, the autocorrelation of

f () can be calculated by r (n) = xu ((n + Cdes, u) NZC) of Equation 16 below:

rn

[Equation 16]

NZC −1 NZC −1 (

J

j 2π (u − cdes, u) NZC

n

* NZC

c0 = L rnz () = L e

() f () n

rv n = 0 n = 0

In Equation 16, "() NZC" is indicative of a modular operation of the value "NZC". If Cdes, u '= u * Cdes, u is a root index related to sampling offsets and and is indicative of a resampling relationship that responds to the timing error, the Cdes value, u' may be denoted by Cdes, u = (y-1) NZC.

By Equations 15 and 16, the value y can be denoted by y = 1 + lf / fs.

The position of the channel response is called a main lobe, and the position of the overlap response of a channel affected by the Doppler frequency (+/-) is called a lateral lobe.

In more detail, the main lobe is indicative of the position caused by displacement 0 and is equal to a normal response position of the channel when there is no influence of the Doppler frequency.

The positive side lobe (+) is indicative of the position caused by the positive displacement (+) and is equal to an overlap response position affected by the positive Doppler frequency (+). The negative side lobe (-) is indicative of the position caused by the negative displacement (-) and is equal to an overlap response position affected by the negative Doppler frequency (-).

As can be seen from Equation 16, it can be recognized that the main lobe of the autocorrelation peak occurs in CDs, u = 0 or CDs, u '= 0. By the previous Equation 16, the pair of lateral lobes occurs with the condition of the following Equation 17:

30 [Equation 17]

u = −1

(* CDs, u)

NZC

Therefore, the result of “u * Cdes, u - m * NZC” is equal to “-1”, as represented by “Cdes, u = (m * NZC -1) / u”. In this case, "m" is indicative of the smallest integer capable of allowing the Cdes value, that is, an integer. For example, if the length of the sequence ZC is 839 and the root index is 300, the value "m" is set to 59, and the value Cdes, u is set to 165.

In case of using the ZC sequence in the time domain, the Cdes value can be defined by the following Equation 18:

[Equation 18]

CDs, u = (NZCm − 1) or

In Equation 18, "m" is indicative of the smallest integer capable of allowing the Cdes value to be an integer and "NZC" is indicative of the length of ZC.

All indexes "u" are relative cousins ​​of the NZC value. Therefore, there is the positive integer (uinv = 1 / u) capable of satisfying the equation (u * uinv = 1 mod NZC). Therefore, the value Cdes, u can easily be represented by Equation 19 below:

[Equation 19]

m⋅NZC 1

cdes, u = - = u − 1 mod NZC

uu

In Equation 19, a negative sign (-) is the opposite of the positive sign (+), so that it can be represented by Equation 20 below:

[Equation 20]

CDs, u = (1 u) mod NZC

In short, if the CAZAC sequence is used in the frequency domain, the "u" index of the CAZAC sequence becomes "CDs" without any change. If the CAZAC sequence is used in the time domain, the operation "(1 / u) mod NZC" is performed on the index "u" of the CAZAC sequence, so that the value Cdes can be acquired.

Since the ZC sequence is used in the frequency or temporal area and that the conjugate property between the Cdes and ZC sequences is used, the distance "du" between the main lobe and the temporal lobe can be represented by Equation 21 below:

[Equation 21]

,

2

cdes, uu ≤NZCdu =

,

NZC −cdes, uu&gt; NZC

The present invention provides several methods for the establishment of restricted cyclic shifts; for example, a first procedure for establishing the restricted cyclic displacement without using the fixed cyclic displacement position, and a second procedure for establishing the restricted cyclic displacement using the fixed cyclic displacement position.

The first procedure is associated with restricted cyclic displacement without considering the predefined displacement position. The second procedure is associated with restricted cyclic displacement with consideration of the predefined displacement position.

With regard to the first procedure, there are a variety of procedures; specifically, a procedure to directly use the displacement value of the Va-th restricted cyclic displacement, and procedure to establish the cyclic displacement interval using the displacement value "Cva". Specifically, the

Cyclic shift sequence is converted into xu, v (n) = xu ((n + Cva) mod NZC), as shown in Equation 4.

With regard to the first procedure, there are a variety of procedures that employ a decimal "Va" for use in cyclic displacement; for example, a procedure to establish the cyclic shift interval by calculating the decimal Va of the offset index.

In other words, if the length of the cyclic shift is set to NCS, the cyclic shift index becomes “xu, va (n) = xu ((n + round (VaNCS) mod NZC).” In this case, “ round ”is indicative of a rounding function.

With regard to the second procedure, there are a variety of procedures that employ the entire "Va" for use in cyclic displacement; for example, a procedure to establish the cyclic shift interval by calculating the integer Va of the shift index. Specifically, the cyclic shift sequence is converted to xu, va (n) = xu ((n + vaNCS) mod NZC).

On the other hand, if the cyclic shift is performed by the multiple of NCS, the random access preambles, each of which has a zero (ZCZ) that has no correlation in the u-th root ZC sequence, are defined by xu, v (n) = xu ((n + vNCS) mod NZC). This definition is appropriate for the low / medium cell that has no problem in high frequency offset. However, if restricted cyclic displacement is used in the high mobility cell, the definition mentioned above is inappropriate for the high mobility cell. Specifically, the available "v" value is restricted and the number of available ZCZ preambles is reduced to 1/3 of the ZCZ preambles of a general case.

Embodiments associated with the cases mentioned above will be described in detail below.

Better mode

This embodiment of the present invention will disclose a method for establishing restricted cyclic displacement using only the influence of Doppler displacement, without using the fixed position of cyclic displacement.

The present invention assumes that the preamble is generated using the ZC sequence used as the CAZAC sequence.

The "du" value of Equation 22 below shows a specific case in which the ZC sequence is generated in the frequency domain.

[Equation 22]

u, 0 ≤ u &lt;NZC

2

d = u NZC - u, NZC

2 ≤ u &lt;NZC

In case of generating the ZC sequence in the time domain, the value "du" can be represented by the following Equation 23:

[Equation 23]

(NZC ⋅ m − 1) u, 0 ≤ (u − 1 mod NZC) &lt;NZC

2

du = NZC - (NZC ⋅ m − 1) u, NZC

2 ≤ (u − 1 mod NZC) &lt;NZC

In Equation 23, "m" is indicative of the smallest positive number capable of allowing the value "du" to be an integer, and NZC is indicative of the length of ZC. Equation 23 can also be represented by the following Equation 24:

[Equation 24]

u − 1 mod NZC, 0 ≤ (u − 1 mod NZC) &lt;NZC

2

du = NZC - (u − 1 mod NZC), NZC

2 ≤ (u − 1 mod NZC) &lt;NZC

Therefore, the v-th cyclic shift of the u-root index can be defined by Xu, v (n) = Xu ((n + Cv) mod NZC). In this case, if the general cyclic shift is decided, the value Cv can be represented by Cv = v * NCS. If the restricted cyclic shift is decided, the value Cv can be represented by Equation 25 below.

[Equation 25]

ν⋅N, ν = 0.1, or, (N

N −1,) for unrestricted sets

CS ZC CS

Cν = S ⋅ν

P + (νmod P) ⋅NCS, ν = 0.1, or, (PG + R −1,) for constrained sets

⋅

If the restricted cyclic displacement that has no predefined position of displacement is decided, this case is considered to be a first case (Case 1), and a detailed description thereof will be described below.

The u-th root ZC sequence and the v th random access preamble, each of which has a null correlation area, are defined by "xu, v (n) = xu ((n + Cv) mod NZC) "

In this case, "Cv" is denoted by Equation 25 above.

In other words, in the case of unrestricted assemblies that have a small amount of the influence of the Doppler shift, the present invention may establish the cyclic shift corresponding to an integer multiple of NCS equal to the basic unit of cyclic shift.

However, the case of unrestricted sets less affected by the Doppler shift may establish the cyclic offset corresponding to the integer multiple of NCS.

In association with FIG. 13, the case of restricted sets that are very affected by the Doppler shift can establish the number (G) of cyclic shifting groups, the number (P) of cyclic shifts applicable to each cyclic shifting group and the number (R) of shifts Additional cyclic

The procedure for calculating each secondary variable can be decided differently by the interval of "du", as previously stated in FIG. 13.

During the overlapping distance interval of NCS du &lt;(NZC / 3), the number of cyclic shifts per group is denoted by P = jdu / NCSj, and there are G (G = jNZC / Sj) groups, each of which it has the length S = 2 · du + P · NCS, and the number of additional restricted cyclic shifts is denoted by R = max (j (NZC -2 · du - G · S) / NCSj, 0).

During the overlapping distance interval of (NZC / 3) du (NZC-NCS) / 2, the number of cyclic shifts per group is denoted by P = j (NZC-2 · du) / NCSj, and there are G (G = jdu / Sj) groups, each of which has the length S = NZC-2 · du + P · NCS, and the number of additional restricted cyclic shifts is denoted by R = min (max (j (du - G · S) / NCSj, 0) P).

In the following, the principles for calculating the secondary variables mentioned above will be described in detail.

() d

1u &lt;NCS

FIG. 15 is a conceptual diagram illustrating a specific case in which the variable (du) is smaller than a basic unit NCS to which the cyclic shift (CS) according to the present invention is applied.

The cyclic displacement unit (NCS) is designed in consideration of the propagation of the delay and the RTD that are capable of being generated in the channel. Therefore, if du is less than NCS, a peak caused by the propagation of the delay and / or the RTD within the NCS range may overlap with the other peak caused by the Doppler shift, as shown in FIG. fifteen. Therefore, when the restricted cyclic offset is established, this embodiment does not establish the cyclic offset for the case where the du value is less than the NCS value.

2≤d (

3)

() NCS u &lt;NZC

FIG. 16 is a conceptual diagram illustrating a procedure for calculating a variable that establishes the cyclic shift within the NCS du &lt;(NZC / 3) interval according to the present invention.

As shown in FIG. 16, the area of ​​cyclic shift generated by the Doppler frequency occurs in the NCS du &lt;(NZC / 3) interval. Specifically, the area of ​​cyclic shift appears in the range of a sequence length located on both sides of the intended cyclic shift.

According to this embodiment, the areas of cyclic shift caused by the Doppler frequency of both sides of the cyclic shift can be grouped into a single group. In addition, the present invention determines how many NCS values ​​can be used without overlapping with others within the "du" range. The number of restricted cyclic shifts available for group can be set to P. Specifically, the P value can be calculated by

5 Middle of the following Equation 26:

[Equation 26]

P = du

NCS

The distance between a specific response 1601 of the channel and the overlap 1601a caused by the Doppler shift is denoted by "du". The distance between a specific response 1601 of the channel and the other overlap 1602b caused by the Doppler shift is denoted by "du."

If the P cyclic shifts are applied to each group, the overlaps generated in the left area with

10 based on the response 1601 of the channel are contained in the interval du, and other overlaps generated in the right area based on the response 1601 of the channel may exist outside the interval du.

In this case, in the case of considering all the overlapping operations of P channel responses generated in the right area, a corresponding length corresponds to P NCS (1602).

Therefore, the length (S) of a single group of cyclic shifts can be equal to the sum of the length 15 "du" and the length "P · NCS" and is represented by the following Equation 27:

[Equation 27]

S = 2 ⋅du + P⋅NCS

On the other hand, the number of cyclic shift groups in total sequences can be calculated by dividing the total length (NZC) of the sequence by the length (S) of the group, and can be represented by Equation 28 below:

[Equation 28]

S

G = NZC

On the other hand, as shown in FIG. 16, a specific area 1603 may be shorter than the length (S)

twenty of the group. The length of the area "1603" corresponds to "NZC - G · S", NZC being the length of the total sequence, G being the number of groups and S being the length of the group.

If NZC-G · S -2du is greater than NCS, the additional cyclic offset can also be applied to the aforementioned area 1603, and a detailed description thereof is shown in the area "1604" of FIG. 16. Therefore, since the number of cyclic shifts that are not based on the shifting group

25 Cyclic is R, the R value can be represented by the following Equation 29:

[Equation 29]

R = max −2 ⋅d − G ⋅S

NCS, 0)

((NZC u)

3) ≤du &lt;(NZC −NCS)

2

()

3 (NZC

FIG. 17 is a conceptual diagram illustrating a procedure for calculating a variable that sets the cyclic shift within the interval (NZC / 3) du &lt;(NZC - NCS) / 2 according to the present invention.

In the area of ​​(NZC / 3) du, unlike the aforementioned case (2) (that is, the aforementioned case (2) of NCS du &lt;(NZC / 3)), the response positions of the channel and of the overlap caused by the

30 Doppler shift exceeds the total length NZC of the sequence, so that overlap can occur between the response of the ideal case channel and the du interval.

For example, at the peak located in the "1701" position of FIG. 17 overlap can occur at positions 1701a and 1701b by the Doppler shift (+/-). Therefore, the number of cyclic shifts applicable to a single group of cyclic shifts in this case (3) is decided by the area “NZC-2du” (1702) located in the

35 center of FIG. 17, so that the number P of restricted cyclic shifts applicable to each group can be calculated by the following Equation 30:

30 [Equation 30]

P = (NZC −2 ⋅du)

NCS

In this case (3), the length S of each group of cyclic shifts can be represented by the following Equation 31:

[Equation 31]

S = NZC −2 ⋅du + P⋅NCS

The variable S is equal to the sum of the length of the area 1702 (NZC-2du) and the length of the area 1703, which corresponds to the length "P · NCS". The length "P · NCS" is variable with the number of cyclic shifts applicable to each real group located on the right side.

On the other hand, the aforementioned case (3) determines the number of cyclic shifting groups in a given ZC sequence considering how many lengths (S, where S = the length of a specific group) will be allowed in the interval du (1704), while the aforementioned case (2) has determined the number of cyclic shifting groups in such a given ZC sequence considering how many lengths (S) will be allowed in the total length NZC of the sequence.

The separation between a specific channel response and two overlaps of this channel response exceeds the total sequence interval, so that the present invention controls the individual overlaps so that they do not overlap each other within the du interval. The number of cyclic shift groups may be represented by Equation 32 below:

[Equation 32]

G = du

S

Finally, the group of cyclic shifts is established in the interval du (1704) as described above, and the area 1705 that is shorter in length than that of the cyclic shifting group may remain. This length of area 1705 corresponds to "du -G · S". If the length of the area 1705 is greater than NCS, the additional cyclic offset can be applied to this length.

Therefore, the number R of additional cyclic shifts can be represented by max (j (duG • S) / NCSj).

If the length (S) of each group of cyclic shifts is greater than "P", additional cyclic shifts corresponding to the number of more than "P" may overlap with the overlapping area (+/-) in the right area. Therefore, this embodiment may indicate the number R of additional cyclic shifts as shown in Equation 33 below:

[Equation 33]

R = min max ((d − G⋅S)

NCS, 0), P)

(

or

2 ≤du

4()

() NZC −NCS

With reference to FIG. 17, the area NZC-2du (1702) located in the central part must be greater than NCS, so that the cyclic offset can be applied to each group. Specifically, this requirement can be represented by NZC-2du&gt; NCS.

If the requirement mentioned above is represented in different ways based on the value du, it can be recognized that the equation NZC-NCS&gt; 2du (that is, (NZC-NCS) / 2&gt; du) must be satisfied. Therefore, this embodiment does not establish the restricted cyclic shift in the interval (NZC-NCS) / 2 du.

Based on the explanation of the individual intervals mentioned above, a detailed description of only the restricted set contained in Equation 25 will be disclosed below. First, the restricted set of Equation 25 may be represented by Equation 34 below:

[Equation 34]

Cν = ⋅S ν

P + (νmod P) ⋅N, ν = 0.1, or, PG R)

(⋅+− 1

CS

In the following, the individual terms for use in the previous cyclic shift will be described.

In Equation 34, S · jv / Pj is indicative of a starting point of each group of cyclic shifts. If the value v is less than the number P of cyclic shifts for each group, S · jv / Pj is indicative of “0”. If the value v is greater than the number P of cyclic displacements for each group and is less than "2P", S · jv / Pj is indicative of "S", corresponding to the length of a single group of cyclic displacements.

If the value v is greater than "2P" and is less than 3P, S · jv / Pj is indicative of "2S", corresponding to the length of two groups of cyclic shifts.

(v mod P) .NCS is indicative of the position of the cyclic shift applied to each group (or of the position of an additional cyclic shift). In other words, the value v is shifted to another position by a predetermined NCS distance in intervals of time P.

The value v of Equation 34 (or Equation 25, which includes Equation 34) does not discriminate between groups or group components, and is indicative of the total number of cyclic shifts. Consequently, the total number of cyclic shifts can be represented by P · G + R.

Modified Examples

In the following several modified examples applicable to the present invention will be described.

Although the best mode mentioned above has disclosed the specific case in which there is no restriction at the starting point of cyclic displacement, the present invention can be applied not only to the aforementioned case, but also to other restricted cases.

In the following, not only the best mode mentioned above will be described, but also all the embodiments capable of being applied more generally.

The position in which the overlapping by the Doppler (+) frequency occurs is denoted by the "+ offset" position, and the position in which the overlapping by the Doppler (-) frequency occurs is denoted by the "-displacement" position. .

FIGURES 18 and 19 are conceptual diagrams illustrating a procedure for reducing the number of ZCZ preamble sequences due to an overlap response in the case of NZC = 839, NCS = 100 and du = 155 according to the present invention.

The cyclic shift of FIG. 18 can start in any position. The cyclic shift of FIG. 19 can be carried out only in the multiple position of NCS. The NCS value of FIG. 18 is the same as in FIG. 19; however, the starting positions of the individual cyclic shifts are different in FIGURES 18 and

19.

In conclusion, the case of FIG. 18 can build many more cyclic shifts than those in FIG. 19. In more detail, the case of FIG. 18 eliminates the restriction of the starting position of the cyclic displacement, so that it can acquire the additional restricted cyclic displacement.

FIG. 20 is a conceptual diagram illustrating the increasing proportion of a restricted cyclic displacement available after the restriction of a cyclic displacement start site is eliminated in the case of NZC = 839 according to the present invention.

The removal of the restriction at the beginning of cyclic displacement may not increase the complexity of physical support.

Therefore, restricted cyclic displacement that has no consideration in the predefined position of displacement is preferred, and the best mode mentioned above is established with the premise mentioned above.

However, the present invention can also be applied to the restricted cyclic displacement having the predefined displacement position, so that the following description will disclose the two classes mentioned above.

First, the case of restricted cyclic displacement (i.e., Case 1) that has no consideration in the predefined position of the displacement will be described in the following.

Equation 21 indicates the overlap distance, regardless of the preamble's generation domain. The number of restricted cyclic shifts available per root ZC sequence is decided differently according to the root index and the NCS value, so different equations are required for use in different overlapping distance intervals.

Specifically, there are two intervals of overlapping distances in which there is no discrimination between overlapping responses. The range in which the restricted cyclic shift can be used is set to NCS du (NZC-NCS) / 2. In this interval, the cyclic shift interval and two overlap intervals do not overlap each other.

In this case, if the preamble is generated in the frequency domain, the value "du" is set to "u", as denoted by du = u. If the preamble is generated in the time domain, the value "du" is set to "1 / u mod NZC", as denoted by du = 1 / u mod NZC. The number of restricted cyclic shifts can be represented by means of the following Equation 35:

[Equation 35]

P⋅G + R, for NCS ≤du ≤ (NZC −NCS)

2

()

Ndesp du =

0, in the other cases

In Equation 35, "P" is indicative of the number of restricted cyclic shifts per group, "G" is indicative of the number of groups generated in a single preamble sequence and "R" is indicative of the number of additional restricted cyclic shifts that do not It is based on the additional group.

The available range of the restricted cyclic shift is denoted by NCS du (NZC-NCS) / 2. This “NCS du (NZC-NCS) / 2” interval can be divided into “NCS -du &lt;(NZC / 3)” and “(NZC / 3) du (NZC-NCS) / 2” based on NZC / 3 .

The reason why the overlap distance interval is divided into “NZC du &lt;(NZC / 3)” and “(NZC / 3) du (NZC-NCS) / 2” based on NZC / 3 has already been released.

Therefore, “NCS du (NZC-NCS) / 2” is decided differently based on “NZC / 3”. Next, the NCS du &lt;(NZC / 3) interval and the (NZC / 3) du (NZC-NCS) / 2 interval will be described.

If the starting position of the first group is set to “0”, the Va-th restricted range of cyclic displacement is defined by [Cva, start, Cva, end] in Equations 36 and 37.

[Equation 36]

gS pN CS, for v Pg pp = 0.1, or, P − 1, g = 0.1,,

⋅ + ⋅ = ⋅ + KG −1

C =

I saw

and

⋅ + ⋅ CS, for v Pgr r = 0.1, or, R − 1

GS rN = ⋅ +

[Equation 37]

Cva, end = Cva, start + NCS −1

The overlap occurs in the positions of Equations 38 and 39 below:

[Equation 38]

±

Fd = (C ±)

() d

Go, start grape, start u

NZC

[Equation 39]

±

Fd = (C ±

() d)

Go, end grape, end u

NZC

In Equation 39, "() NZC" is indicative of a modular operation.

First, the NCS du &lt;(NZC / 3) interval of the overlap distance (i.e., interval 1 of the overlap distance) has a number G = jNZC / Sj of groups. Each group includes a P = jdu / NCSj number of restricted cyclic shifts. The length of each group is denoted by S = 2 · du + P · NCS. If he

Additional cyclic shift available is a positive number (+), the value R is denoted by R = j (NZC-G · S2 · du) / NCSj.

FIG. 21 is a conceptual diagram illustrating an exemplary cyclic shift in the case of NZC = 839, NCS = 40 and du = 150 according to an embodiment of the present invention. Each group has three cyclic shifts, and there are two additional cyclic shifts in the remaining intervals. In this example, the total number of restricted cyclic shifts is "5".

According to one embodiment, the present invention applies the number of calculated groups, the number of restricted cyclic shifts per group and the length of the group to Equations 36 and 37, and then sets the application range of the cyclic shift in consideration of the mentioned parameters in the foregoing.

Then, in the interval (NZC / 3) du (NZC-NCS) / 2 of the overlap distance (i.e., interval 2 of the overlap distance), the number of cyclic shifts available per group is denoted by P = [(NZC-2 · du) / NCS], the length of each group is denoted by S = NZC-2 · du + P · NCS, and there are G groups (where G = jdu / Sj).

The additional cyclic offset is selected from the central part and the residual part of the right side. In this case, the selected cyclic shifts should be the least number of cyclic shifts. Specifically, if the value R is a positive number, the number of additional cyclic shifts is denoted by R = min (j (du-G · S) / NCSj, P). The starting position of the Va-th restricted cyclical displacement is calculated by applying the above-mentioned parameters to Equations 36 and 37.

FIG. 22 is a conceptual diagram illustrating an exemplary cyclic shift in the case of NZC = 839, NCS = 40 and du = 399 according to an embodiment of the present invention. Each group

There are four groups, each of which has a single cyclic shift and a single additional cyclic shift. In this example, the total number of restricted cyclic shifts is 5.

According to this embodiment, the present invention applies the number of calculated groups, the number of restricted cyclic shifts per group and the length of the group to Equations 36 and 37, and then sets the range of application of the cyclic shift in consideration of the mentioned parameters in the foregoing.

In fact, the equal sign (=) between two overlapping distance intervals may have no meaning or relatively small importance. For example, if using the ZC sequence that has the length of 839, the value (NZC / 3) is equal to 279.67 (ie (NZC / 3) = 279.67), so that the divided intervals NCS du &lt;(NZC / 0) and (NZC / 3) du (NZC-NCS) / 2 can have the same results as those of the divided intervals NCS du (NZC / 3) and (NZC / 3) &lt;du (NZC -NCS) / 2.

Then, in the following, the restricted cyclic displacement (ie, Case 2) will be described considering the predefined position of displacement.

A procedure for generating the restricted cyclic shift using the predefined offset position is changed to another procedure. Each interval of the overlapping distance includes not only G groups, each of which has P cyclic shifts, but also an additional first cyclic shift between the R1 groups.

In case of using the predefined position of displacement, the present invention has a particular additional cyclic displacement, unlike the other case in which there is no predefined position of displacement in the area 2 of the overlap distance range.

In area 2 of the overlap distance range, the main region generally appears in the forward samples of the sequence, and the overlapping regions generally appear in the subsequent samples of the sequence. However, according to Case 2, the main region appears in the subsequent samples of the sequence and the overlapping regions appear in the front samples of the sequence.

The second additional cyclic shift is denoted by R2. The second additional cyclic shift does not appear in interval 1 of the overlap distance. The total number of restricted cyclic shifts can be represented by means of the following Equation 40:

[Equation 40]

P⋅G + R1 + R2, for NCS ≤du ≤ (NZC −NCS)

2

()

Ndesp du =

0, in the other cases

Since the starting position of the first group is "0", the Va-th restricted cyclic shift is defined in [Cva, start, Cva, end], as denoted by Equations 41 and 42:

[Equation 41]

gS pN CS, for vPg p, p = 0.1, or, P −1, g = 0.1,,

⋅ + ⋅ = ⋅ + KG −1

⋅ + ⋅ = ⋅ + r 11

C, = GS r1 NCS, for v PG 1, r1 = 0.1, or, R -

vinicio

and

N dPN + (G -) S)

i (- + ⋅ 1

N rN, for vPG R r, r = 0.1, or, R −1

N l⋅ + ⋅ = ⋅ ++

ZCu CS CS CS 2 CS 122 2

[Equation 42]

Cva, end = Cva, start + NCS −1

Related overlap occurs in the positions of Equations 43 and 44 below:

[Equation 43]

±

Fd = (C ±)

() d

Go, start grape, start u

NZC

[Equation 44]

F ± d = C ± d

() () NZC

Go, end grape, end u

In Equations 43 and 44, () NZC is indicative of a modular operation.

In the NCS interval du &lt;(NZC / 3) of the overlap distance (i.e., the interval 1 of the overlap distance), there are G groups (where G = jNZC / Sj), there are P restricted cyclic shifts (where P = jdu / NCSj), and the length of the group is denoted by S = ([2du / NCSl + P · NCS). If the value R1 is a positive number (+), the number of additional cyclic shifts is denoted by R1 = j (NZC-G · S-2 · du) / NCSj.

FIG. 23 is a conceptual diagram illustrating an exemplary cyclic shift in the case of NZC = 839, NCS = 40 and du = 150 according to another embodiment of the present invention. In FIG. 23, each group includes three cyclic shifts and two additional cyclic shifts. In this example, the total number of cyclic shifts is "5".

According to this embodiment, the present invention applies the number of calculated groups, the number of restricted cyclic shifts per group and the length of the group to Equations 41 and 42, and then sets the application range of the cyclic shift in consideration of the mentioned parameters in the foregoing.

Then, in the interval (NZC / 3) du (NZC-NCS) / 2 of the overlap distance (i.e., interval 2 of the overlap distance), the number of cyclic shifts available per group is denoted by P = [(NZC-2 · du) / NCS], the length of each group is denoted by S = ([NZC-2 · du) / NCSl + P) · NCS, and there are G groups (where G = jdu / Sj ).

The first additional cyclic shift is calculated by the same procedure as that of interval 1 of the overlapping distance. If the value R1 is a positive number, the number of additional cyclic shifts is denoted by R = min (j (du-G · S) / NCSj, P).

If the value R1 is equal to "0" (ie, R1 = 0), the presence or absence of a second additional cyclic shift must be determined. The shape of the second additional cyclic displacement is the opposite of the conventional cyclic displacement form, as shown in the last cyclic displacement of FIG. 2. 3.

The present invention determines whether the overlap interval of the second additional cyclic shift is an available range (ie, du- [P · NCS, + (G-1) · S] NZC-2du + NCS) and determines whether the interval Cyclic scrolling is available (ie, X + NCS: 2du). If it is determined that the cyclic shift interval is available (i.e., X + NCS 2du),

FIG. 24 is a conceptual diagram illustrating an exemplary cyclic shift in the case of NZC = 839, NCS = 40 and du = 399 according to another embodiment of the present invention. In FIG. 24, each group includes three cyclic shifts and no additional cyclic shifts (that is, a first zero additional cyclic shift). And each group also includes a single additional cyclic shift in which a relative position of the main region is opposite that of the overlapping region. This second cyclic shift does not occur when the fixed cyclic offset position is not used, as shown in FIG. 22. In this example, the total number of cyclic shifts is "4".

2 2

According to this embodiment, the present invention applies the number of calculated groups, the number of restricted cyclic shifts per group and the length of the group to Equations 41 and 42, and then sets the application range of the cyclic shift in consideration of the mentioned parameters in the foregoing.

According to another embodiment, a specific system with the fixed cyclic offset can determine the cyclic offset according to the following procedure.

First, the total sequence interval is divided by the value of the cyclic shift.

Next, the present invention looks for the interval (± uo ± (m * NZC-1) / u) in which the interference caused by displacement occurs in the first interval (ie, n = 1). In this case, there is a plurality of intervals, each of which has interference.

For example, if only the first interference is considered, the maximum number of interference generation intervals can be set to "4".

Then, if the first interval does not overlap with all the interference intervals caused by the displacement, the first interval is set to an available interval, and the remaining intervals caused by the displacement are set to a restricted interval (also called the interval of prohibition).

The present invention goes to the next interval (ie, n = n + 1), and repeatedly searches for the interval in which the interference is generated by the displacement.

Although the present invention seeks the interference generation interval in the nth interval, if an observation interval, several intervals caused by displacement, a pre-established available interval and pre-established prohibition intervals do not overlap each other, The present invention determines that a current interval is an available interval and determines that the various previous intervals caused by the displacement associated with the current interval are prohibition intervals. If the above-mentioned procedure is repeated until the last interval is reached, the present invention can determine the cyclic displacement in the system, including the fixed cyclic displacement.

According to another additional embodiment, the present invention can apply the range of application of the established cyclic shift mentioned above only to the high mobility cell in a mobile communication system that includes several cells.

In this case, the present invention can determine if a corresponding cell has high mobility by determining if the frequency offset associated with the cell is greater than a predetermined level after acquiring the cell information. In this case, the predetermined level is indicative of a frequency offset value, which can be easily decided or modified by those skilled in the art.

Preferably, the present invention can control node B or the UE to determine if the corresponding cell is the high mobility cell. However, the UE has difficulty estimating the frequency offset value of each of the other UEs contained in the cell. Therefore, it is more preferable that node B determines whether the corresponding cell is the high mobility cell in consideration of several UEs of the cell and that it transmits the resulting signal through the broadcasting channel.

On the other hand, if it is determined that the corresponding cell is not indicative of the high mobility cell, the present invention may include a method for assigning an unassigned sequence to the high mobility cell.

The following description shows that the equations are modified in others with the same condition as in the best way, and a detailed description of this will be described in the following.

In association with the best mode, the equations mentioned above may also be denoted by the following expression.

If Cv = S · [v / P] + (v mod P) · NCS, v = 0.1,…, (P · G + R-1) and E = jdu / NCSj, F = j (NZC-2du ) / NCSj in the interval NCS du &lt;(NZC / 3) of the overlap distance, the values ​​P and G are denoted by P = E, S = 2du + E · NCS, G = [F · NCS / S].

If Cv = S · [v / P] + (v mod P) · NCS, v = 0.1,…, (P · G + R-1) and E = jdu / NCSj, F = j (NZC-2du ) / NCSj in the interval (NZC / 3) du (NZCNCS) / 2 of the overlapping distance, the values ​​P, S, G and R are denoted by P = F, S = NZC-2du + F · NCS, G = jE · NCS / Sj, R = min (j (du-G · S) / NCSj, F).

Next, the following will describe the case of restricted cyclic displacement using other equations considering the predefined position of displacement (Case 2).

The ZC sequence of u-th root having the null correlation region, that is, the v-th random access preamble, is defined by xu, v (n) = xu ((n + Cv) mod NZC). In this case, the value Cv is defined by Equation 45:

35 [Equation 45]

ν⋅NCS, ν = 0.1, or, (NZC

NCS −1,) for low / medium mobility cell

Cν = S ⋅ν

ν (⋅ + - 1,

P + (mod P) ⋅N, ν = 0.1, or, PG R) for high mobility cell

CS

in which CP⋅ = X, if R2 = 1, for the high mobility cell.

G + R1 + R2 -1

In this case, the parameters of the high mobility cell can be defined by the following explanation.

In more detail, in the NCS du &lt;(NZC / 3) overlap interval the value P is denoted by P = jdu / NCSj, the value S is denoted by S = ([2du / NCS + P) · NCSl, and the G value is denoted by G = jNZC / Sj. A first additional cyclic shift R1 is denoted by R1 = max (j (NZC-2 · du-G · S) · NCSj, 0), and a second additional cyclic shift R2 is denoted by R2 = 0.

In the interval (NZC / 3) du (NZC-NCS) / 2 overlap, the value P is denoted by P = j (NZC-2 · du) / NCSj, the value S is denoted by S = ([NZC- 2 · du) / NCS + P) · NCSl, and the value G is denoted by G = jdu / Sj. A first additional restricted cyclic shift is denoted by R1 = min (max (j (du-G · S) / NCSj, 0), P), a second additional restricted cyclic shift R2 is denoted by R2 = 1 in the case of R1 = 0 and "X - NCS &lt;2 du". In this case, the value X is denoted by X = [(NZC-du + P · NCS + (G-1) S) / NCSl · NCS.

In the case of restricted cyclic phase shift of xu, v (n) = xu ((n + Cv) mod NZC), the procedure for directly using the displacement value of the 10th restricted cyclic displacement has been disclosed. Unlike the procedure, another method can be applied to the present invention to employ the Va value for the Va-th restricted cyclic shift. In more detail, the similar cyclic shift can be generated using the equation xu, va (n) = xu ((n + round (vaNCS)) mod NZC).

In case of generating the cyclical displacement using the procedure mentioned above, the basic concept is the same as those of the procedures mentioned above. However, different equations apply to the procedures mentioned above.

The case (Case 1) of the restricted cyclic displacement that has no consideration of the predefined displacement position will be described using other equations.

The index (v) of the cyclic shift is represented by the following Equation 46:

[Equation 46]

gS ⋅ +,

⋅ + p, for v = Pg pp = 0.1, or, P −1, g = 0.1, KG −1

v =

⋅ + r, for v = Pg rr = 0.1, or, R −1

GS ⋅ +

In the NCS du &lt;(NZC / 3) overlap interval, the P value is denoted by P = jdu / NCSj, the S value is denoted by S = 2du / NCS + P, and the G value is denoted by G = jNZC / (S · NCS) j, and the additional restricted cyclic shift R is denoted by R = max (j (NZC-2 · du) / NCS-G · Sj, 0).

In the interval (NZC / 3) du (NZC-NCS) / 2 overlap, the value P is denoted by P = j (NZC-2 · du) / NCSj, the value S is denoted by S = (NZC-2du ) / NCS + P, the value G is denoted by G = jdu / (S · NCS) j and the value R is denoted by R = min (max (jdu / NCS-G · Sj, 0), P).

If E = jdu / NCSj, F = j (NZC-2du) / NCSj, the expression mentioned above can be represented in other ways. In more detail, in the NCS du &lt;(NZC / 3) overlap interval, the P value is denoted by P = E, the S value is denoted by 2du / NCS + E, the G value is denoted by G = jF / Sj and the value R is denoted by R = min (j (NZC-2 · du) / NCS-G · Sj, E).

In the interval (NZC / 3) du (NZC-NCS) / 2 overlap, the P value is denoted by P = F, the S value is denoted by S = NZC / NCS-2du / NCS + P, the G value is denoted by G = jE / Sj and the value R is denoted by R = min (jdu / NCS-G · Sj, F).

Next, the case of restricted cyclic displacement (Case 2) will be described using other equations considering the predefined position of displacement.

The index (v) of the cyclic shift is represented by the following Equation 47:

Four. Five [Equation 47]

gSp, paravPgp p = ,,, P−, g = ,,,

⋅ + = ⋅ + 01o 101o G − 1

v GSr, paravPGr r = ,,, or

= ⋅ + = ⋅ + 01R − 1

1 111

= GRr r 01o

Xr +, paravP⋅ + +, = ,,, R − 1

2 122 2

In the NCS du &lt;(NZC / 3) overlap interval, the P value is denoted by P = jdu / NCSj, the S value is denoted by S = ([2du / NCSl + P), and the G value is denoted by G = jNZC / (SNCS) j, and the additional restricted cyclic shift R1 is denoted by R1 = max (j (NZC-G · SNCS-2 · du) / NCSj, 0).

In the interval (NZC / 3) du (NZC-NCS) / 2 overlap, the value P is denoted by P = j (NZC-2 · du) / NCSj, the value S is denoted by S = ([(NZC -2 · du) / NCS + P) l, the value G is denoted by G = jdu / (SNCS) j and the value R1 is denoted by R1 = min (max (jdu - G · SNCS) / NCSj, 0), P).

If R1 = 0 and X · NCS + NCS 2du, the value R2 can be represented by R2 = 1. In this case, the value X is denoted by X = [(NZC - du + P · NCS + (G-1) SNCS) / NCSl.

If E = jdu / NCSj; s = du mod NCS; E '= [2s / NCSl and F = j (NZC-2du) / NCSj; t = (NZC-2du) mod NCS; E = [t / NCSl, in the NCS du &lt;(NZC / 3) overlap interval, the P value is denoted by P = E, the S value is denoted by S = 2F + F ', the G value is denoted by G = jE / Sj, and the R2 value is denoted by R2 = min (EG · S, F).

If R1 = 0 and X · NCS 2du-NCS, the R2 value may be represented by R2 = 1. In this case, the value X is denoted by X = [X '+ F + (G-1) Sl, X' = (NZC - du) / NCS.

As described above, according to the aforementioned embodiments, in case of implementing the cyclic shift sequence using the CAZAC sequence, the present invention can define the set of cyclic shifts capable of eliminating the ambiguity of shifting caused by the frequency or temporary displacement.

Furthermore, in case of accessing the non-synchronized channel, the frequency shift or the temporal shift are not adjusted to this non-synchronized channel, so that the present invention can increase the intensity of this channel.

Depending on the range of influence of the pulse-forming filter, the present invention can define the set of cyclic shifts in which first order interference, order interference and major order interference are considered.

It should be noted that most of the terminology disclosed in the present invention is defined in consideration of functions of the present invention, and may be determined differently according to the intent of those skilled in the art or usual practices. Therefore, it is preferable that the terminology mentioned above be understood based on all the content disclosed in the present invention.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the scope of the invention. Thus, it is intended that the present invention encompass the modifications and variations of this invention insofar as they are within the scope of the appended claims.

As is evident from the above description, the present invention can easily establish a cyclic shift interval (CS) at a specific location that has no overlap considering a channel response of a reception sequence (Rx) and a location of overlapping of this reception sequence (Rx), even if a reception signal (Rx) is displaced by a propagation delay of the channel or a propagation delay regardless of the categories of a domain that generates a sequence, so that it greatly reduces the number of detection errors and the frequency of false alarms.

And, if a cyclic shift sequence (CS) is assigned to a cell that has a frequency offset of more than a predetermined level, the present invention can minimize the influence of a frequency shift in a high mobility cell.

The present invention is related to a first procedure of assigning a sequence to each cell in consideration of the characteristics of the CAZAC sequence, and a second procedure for the establishment of the cyclic shift to be applied to the first procedure. Therefore, the present invention can be applied to a wireless communication system (for example, a UE and a node B).

Although preferred embodiments of the present invention have been disclosed for illustrative purposes, those skilled in the art will appreciate that various modifications, additions and substitutions are possible without departing from the scope of the invention as disclosed in the appended claims.