RU2419234C1 - Method to generate reference signal sequence with application of grouping - Google Patents

Abstract

FIELD: information technologies.

SUBSTANCE: one or several basic sequences are received, having alternating length, a cyclic shift is applied, corresponding to the alternating value of the cyclic shift, to one or more basic sequences to generate a reference signal sequence, a reference signal sequence is sent to a receiving side.

EFFECT: minimisation of noise stipulated for by signals having another length and arriving from other cells.

16 cl, 5 dwg, 32 tbl

Description

Technical field

The present invention relates to a method for generating a reference signal sequence and, in particular, to a method for grouping sequences having a variable length corresponding to one or more resource block sizes, a method for generating a reference signal sequence, and a method for generating a reference signal sequence using a Zadov-Chu sequence (ZC )

State of the art

The following explanation is considered mainly with respect to the 3GPP LTE system, but the present invention is not limited to this system, and the illustrative 3GPP LTE system is intended only for one skilled in the art to understand the present invention.

There is a lot of sequences used for transmitting signal, but in 3GPP LTE system (3 ^rd Generation Partnership Project Long Term Evolution) sequence CAZAC (Constant Amplitude Zero Auto-Correlation , constant amplitude, zero autocorrelation) forms a base sequence for transmitting signals. The CAZAC sequence can be used on various channels to retrieve ID or control information, for example, uplink / downlink (SCH) channels, including P-SCH (primary SCH) and S-SCH (secondary SCH), a pilot channel for transmitting the reference signal. In addition, the CAZAC sequence can be used in scrambling.

As CAZAC sequences, basically two types of CAZAC sequences are used, i.e. the CAZAC GCL sequence and the CAZAC Zadova-Chu sequence. The two types of CAZAC sequences are linked together by a complex conjugation ratio. Thus, the CAZAC GCL sequence can be obtained by calculating the complex conjugate CAZAC Zadova-Chu sequence. The CAZAC Zadova-Chu sequence is defined as follows.

[Equation 1]

(для нечетных N)

(for odd N)

[Equation 2]

(для четных N),

(for even N),

where k represents the index of the sequence component, N represents the length of the generated CAZAC sequence, and M represents the sequence ID or sequence index.

When the CAZAC Zadova-Chu sequence given by Equations 1 and 2 and the CAZAC GCL sequence, which is obtained by complex conjugation of the CAZAC Zadova-Chu sequence, are represented by c (k; N, M), these sequences can have the following three attributes.

[Equation 3]

[Equation 4]

[Equation 5]

R _{M1, M2; N} (d) = p (for all M ₁ , M ₂ and N)

Equation 3 means that the CAZAC sequence always has size 1, and Equation 4 indicates that the autocorrelation function of the CAZAC sequence is expressed by a delta function. In this case, autocorrelation is based on circular correlation. In addition, Equation 5 indicates that cross-correlation is always constant.

Of these two kinds of CAZAC sequences, the following explanation is mainly focused on the Zadov-Chu sequence (hereinafter “ZC sequence”).

In a 3GPP LTE system using this ZC sequence as a reference signal sequence, the length of the ZC sequence must be equal to the size of the resource block. In addition, you can use not only a sequence of one resource block size, but also a reference signal sequence having a length corresponding to several resource block sizes.

In a single cell environment, the reference signals are transmitted using the local FDM (frequency division multiplexing) method for multiplexing signals from multiple instances of user equipment (UE). However, in a multi-cell environment, reference signals are transmitted using an additional CDM (Code Division Multiplexing) method to distinguish signals from signals of neighboring cells. With this multiplexing, two methods can be used. One is a CDM method using ZC sequences having different root indices, and the other is a CDM method using ZC sequences having the same root index (M) but using a different cyclic shift.

When the length of the reference signals using these varieties of ZC sequences is the same, the cross-correlation values are small in both cases. However, if reference signals having a different length came as interference from neighboring cells and were transmitted using the same frequency range or an overlapping frequency range, the cross-correlation value would be significant.

Disclosure of invention

Object of the invention

Accordingly, the present invention is intended to provide a method for generating a reference signal sequence that minimizes interference caused by signals having a different length coming from neighboring cells.

For this method, the present invention also provides a method for efficiently grouping sequences, according to which each group consists of sequences having a high cross-correlation value, and supports variable length sequences for use as reference signals.

In addition, the present invention provides a method for generating a reference signal sequence based on the grouping described above.

Problem solving

To solve these problems and achieve other advantages, and according to the purpose of the invention, implemented and widely described here, a method for grouping sequences having a variable length corresponding to one or more sizes of a resource block is provided. According to one embodiment, the method comprises the steps of grouping sequences into groups so that each group contains at least one sequence of each length, the grouped sequence being the base sequence that is used to apply the cyclic shift corresponding to the variable value of the cyclic shift , and the base sequence with a cyclic shift is used as a reference signal sequence.

Preferably, the number of groups is 30.

In addition, grouping can be carried out so that each of the groups contains one base sequence of each length corresponding to from 1 to 5 sizes of the resource block, and two basic sequences of each length corresponding to 6 or more sizes of the resource block.

Preferably, a base sequence having a length corresponding to 3 or more resource block sizes is specified using a Zadova-Chu (ZC) sequence, and a base sequence having a length corresponding to 1 or 2 resource block sizes is specified using a sequence other than the sequence ZC.

In another aspect of the present invention, a method for generating a reference signal sequence is provided. In one embodiment of this aspect, the method comprises the steps of setting one or more base sequences having a variable length corresponding to one or more resource block sizes; and applying a cyclic shift corresponding to a variable value of the cyclic shift to a given base sequence, the base sequences being divided into groups and each of the groups containing at least one base sequence of each length.

), заданную наибольшим простым числом, которое меньше размера соответствующей опорной сигнальной последовательности. Кроме того, базовую последовательность можно задать усечением последовательности ZC, имеющей длину (

), заданную наименьшим простым числом, которое больше размера соответствующей опорной сигнальной последовательности. Кроме того, в этом варианте осуществления, предпочтительно, количество групп равно 30.In this case, the base sequence can be specified by cyclic extension of the sequence ZC having a length (

) given by the largest prime number that is less than the size of the corresponding reference signal sequence. In addition, the base sequence can be specified by truncating the ZC sequence having a length (

) defined by the smallest prime number that is larger than the size of the corresponding reference signal sequence. In addition, in this embodiment, preferably, the number of groups is 30.

In addition, each of the groups may contain one base sequence of each length corresponding to 1 to 5 resource block sizes, and two base sequences of each length corresponding to 6 or more resource block sizes.

In addition, a base sequence having a length corresponding to 3 or more resource block sizes can be defined using a Zadova-Chu sequence (ZC) with a specific index (q) of the ZC sequence, and a base sequence having a length corresponding to 1 or 2 block sizes resources can be specified using a sequence other than the ZC sequence.

In addition, preferably, the specific index (q) of the sequence ZC is a function of the index (u) of the group and the index of the number of the base sequence (v) in the group.

Furthermore, a predetermined cyclic shift base sequence may be used for the uplink reference signal sequence.

Furthermore, according to the above-described embodiments, the resource block size may correspond to the size of 12 subcarriers in the frequency domain.

In another aspect of the present invention, there is provided a method for generating a reference signal sequence using a Zadoff-Chu (ZC) sequence. In one embodiment of this aspect, the method comprises the steps of defining a specific base sequence using the qth root sequence ZC, the base sequences being divided into groups, and “q” being a function of the group index (u) and index (v) of the base number sequence in the group; and applying a cyclic shift corresponding to the cyclic shift variable to a predetermined base sequence to generate a reference signal sequence.

In one case, the specific index (q) of the sequence ZC can be determined by one of the equations

Where

илиWhere

or

Where

- длина, заданная наибольшим простым числом, которое меньше размера соответствующей опорной сигнальной последовательности,

- размер соответствующей опорной сигнальной последовательности, "round(z)" является функцией округления до ближайшего целого числа, ближайшего к z, и "floor(z)" является функцией, возвращающей наибольшее целое число, не превышающее z. Однако

может быть длиной, заданной наименьшим простым числом, которое больше соответствующей опорной сигнальной последовательности согласно другому варианту осуществления этого изобретения.Where

- the length specified by the largest prime number that is less than the size of the corresponding reference signal sequence,

is the size of the corresponding reference signal sequence, "round (z)" is a function of rounding to the nearest integer closest to z, and "floor (z)" is a function that returns the largest integer not exceeding z. However

may be the length given by the smallest prime number that is greater than the corresponding reference signal sequence according to another embodiment of this invention.

In another case, the specific index (q) of the sequence ZC can be determined by one of the equations,

Where

илиWhere

or

Where

- длина, заданная наибольшим простым числом, которое меньше размера соответствующей опорной сигнальной последовательности,

- размер соответствующей опорной сигнальной последовательности, "round(z)" является функцией округления до ближайшего целого числа, ближайшего к z, и "floor(z)" является функцией, возвращающей наибольшее целое число, не превышающее z. Однако

может быть длиной, заданной наименьшим простым числом, которое больше соответствующей опорной сигнальной последовательности, согласно другому варианту осуществления этого изобретения.Where

- the length specified by the largest prime number that is less than the size of the corresponding reference signal sequence,

is the size of the corresponding reference signal sequence, "round (z)" is a function of rounding to the nearest integer closest to z, and "floor (z)" is a function that returns the largest integer not exceeding z. However

may be the length given by the smallest prime number that is greater than the corresponding reference signal sequence, according to another embodiment of this invention.

In one specific embodiment of this invention, the maximum number of indices (v) of the base sequence number in each group can be set to 2, and then the specific index (q) of the ZC sequence can be determined by one of the equations,

Where

илиWhere

or

Where

or one of the equations

Where

илиWhere

or

Where

можно задать равным 31 или 37, но для

можно задать и другие значения.Preferably,

can be set to 31 or 37, but for

You can specify other values.

Advantages of the Invention

According to these embodiments of this invention, since the base sequence for applying the cyclic shift is grouped and each group contains at least one base sequence of each length, the UE can use sequences of different lengths as a reference signal sequence when a particular group is allocated to one cell or Node B.

In addition, since each group contains base sequences having high cross-correlation, if each group is allocated to one cell or Node B, inter-cell interference can be minimized.

Description of drawings

FIG. 1 is a generalized diagram illustrating a truncated sequence generation method.

FIG. 2 is a generalized diagram illustrating a generation method using a filling part.

FIG. 3-5 are generalized sequence grouping schemes according to one embodiment of this invention.

The implementation of the invention

Further, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. It should be understood that the detailed description, which will be disclosed in conjunction with the accompanying drawings, is intended to describe illustrative embodiments of the present invention and is not intended to describe a unique embodiment that can perform the present invention.

Further, the detailed description includes some details to fully understand the present invention. However, it will be apparent to those skilled in the art that the present invention may be practiced without these details. In order to avoid ambiguity of the concept of the present invention, structures and devices known in the art will be omitted or shown in a block diagram based on the basic functions of each structure or device. In addition, to the extent possible, the same positions will be used in the drawings and description of the invention to refer to the same or similar parts.

As indicated above, the present invention is intended to provide a method for generating a reference signal sequence that minimizes interference caused by signals having a different length coming from neighboring cells.

To do this, consider the length of the CAZAC sequence.

Currently, in a 3GPP LTE system, a resource block size (RB) for transmitting any kind of OFDM symbol, including a reference signal symbol, corresponds to a size of 12 subcarriers. Therefore, when a ZC is generated for the uplink reference signal sequence, the size of the ZC sequence will correspond to the size of 12 subcarriers.

In the case of a CAZAC sequence, the number of indices (M) of CAZAC sequences that can be distinguished from each other is determined by the number of mutually prime numbers mutually prime with the length (N) of the sequence. Therefore, when generating a ZC sequence of length 12, the number of ZC sequences having different sequence indices is 4. However, if the ZC sequence is generated based on a length equal to a prime number (N), the number of ZC sequences having different sequence indices may be N- 1, which maximizes the number of ZC sequences. Thus, various methods for generating a CAZAC sequence based on a length equal to a prime are provided.

First, we describe a method for generating a truncated sequence.

In FIG. 1 shows a generalized diagram illustrating a truncated sequence generation method.

According to FIG. 1, when the required length of the CAZAC sequence is “L”, a CAZAC sequence having a length equal to the prime number “X” (where X> L) is generated. In addition, the generated CAZAC GCL sequence having a length of "X" is truncated to a length of "L", i.e. the portion of the sequence having the length "X-L" is truncated.

According to this method, the number of CAZAC sequences reaches a maximum. But since part of the generated sequence is cut off, the auto / cross-correlation properties of the CAZAC sequence, defined by equations 4 and 5, are somewhat impaired. In addition, with the exception of sequences having low correlation properties, the actual number of sequences decreases. In addition, due to truncation, the good PAPR property of the CAZAC sequence may also be degraded.

Thus, another type of method for generating a CAZAC sequence based on a length equal to a prime is presented. One of these methods involves generating a CAZAC sequence having a length equal to the prime number “X” (where X <L), and adding components having a length “L-X” to the generated CAZAC GCL sequence. These components added to the generated sequence can be called the filling part, so this method can be called the generation method using the filling part.

In FIG. 2 is a generalized diagram illustrating a generation method using a fill part.

According to FIG. 2, when the required CAZAC sequence length is “L”, a CAZAC sequence of length “X” is generated, which is the maximum prime number less than “L”. A fill portion having a length of "L-X" is added to the generated sequence.

According to one method for such methods, the filling part may consist of zeros. According to this method, the number of CAZAC sequences can be maximized. Furthermore, CAZAC sequence auto / cross-correlation properties can be maintained when the sequence is distinguished with respect to the length “C1” shown in FIG. 2.

Preferably, the filling portion may be a cyclic extension of the CAZAC sequence. Thus, the filling part (C2) can be generated by cyclic copying of the first part of the generated CAZAC GCL sequence and adding to the generated sequence. Due to this, the resulting sequence can have good auto / cross-correlation properties, even when the sequence is distinguished in relation to the total sequence length (L). Therefore, this method has an additional advantage over the above method, using a filling part consisting of zeros.

The present invention for generating a reference signal sequence using a CAZAC sequence is mainly based on a generation method using a fill portion generated by the above-described cyclic extension. However, it is not necessary to be limited to this generation method, i.e. the present invention can be based on a truncated sequence generation method and a generation method using a filler part consisting of zeros.

On this basis, we consider the inter-cell interference caused by the use of sequences having different lengths.

When the CAZAC sequence is used as a reference signal sequence, the inter-cell interference is proportional to the cross-correlation value between the two sequences. Thus, in the following examples, the cross-correlation value due to the overlap between the original reference signal transmitted by a certain resource region and incoming sequences coming from neighboring cells whose length is different from the length of the original reference signal and transmitted by the same resource region is considered regarding the sequence index ZC.

In particular, in the following examples, sequences having a length of 1 RB, 2 RB, and 3 RB are considered. Assume that sequences having a length of 1 RB and 2 RB are generated by cyclically expanding a ZC sequence having a length specified by a largest prime number that is less than the corresponding resource block size. In addition, suppose that a sequence having a length of 3 RB is generated by the truncated sequence generation method. In this way, sequences having an appropriate resource block size can be generated based on a length equal to a prime number by one of the 3 aforementioned generation methods.

First, consider the case where a sequence having a length of 1 RB and a sequence having a length of 2 RB overlap in the same resource area. A sequence having a length of 1 RB and a sequence having a length of 2 RB can be expressed as follows.

[Equation 6]

Here, s ₁ and s ₂ denote indices that are coprime with a sequence length (N or 2N). In this example, for sequences having a length of 1 RB and a length of 2 RB, are generated using the cyclic extension method, s ₁ may be 1, 2, ..., 10, and s ₂ may be 1, 2, ..., 22. N ₁ may be equal to 11, and N ₂ may be equal to 23.

), генерируемое, когда последовательность длиной 1 RB перекрывается с последовательностью длиной 2 RB в области первых 12 поднесущих последовательности длиной 2 RB, можно выразить следующим образом.On this basis, the value of cross-correlation (

) generated when a sequence of 1 RB length overlaps with a sequence of 2 RB length in the region of the first 12 subcarriers of a sequence of 2 RB length can be expressed as follows.

[Equation 7]

For d = 0,

приближается к нулю, последовательности, указанные этими индексами последовательности, дают высокую взаимную корреляцию.From equation 7 it follows that if the combination of sequence indices (s ₁ and s ₂ ) satisfies the condition that the expression

approaches zero, the sequences indicated by these sequence indices give a high cross-correlation.

приближается к нулю.Thus, one embodiment of the present invention provides for the grouping of sequences into groups so that the sequences contained in each group have a high cross-correlation with each other. If we consider a sequence of length 1 RB and a sequence of length 2 RB, it is possible to group a combination of indices of a sequence that satisfies the condition that the expression

approaching zero.

However, to determine a more general condition for grouping sequences, consider some other examples.

двух последовательностей можно выразить следующим образом.When a 1 RB sequence is overlapped in the region of the last 12 subcarriers of a 2 RB sequence, the cross-correlation value

two sequences can be expressed as follows.

[Equation 8]

For d = 0

According to equation 8, we can also conclude that if the combination of sequence indices (s ₁ and s ₂ ) satisfies the condition that the expression

approaches zero, the sequences indicated by these sequence indices give a high cross-correlation. Therefore, if we consider a sequence of 1 RB length and a sequence of 2 RB length, the position where the overlap occurs does not change the grouping condition.

Now consider the case where a 1 RB sequence and a 3 RB sequence overlap in the same resource area.

First of all, a sequence of length 1 RB and a sequence of length 3 RB can be expressed as follows.

[Equation 9]

Here s ₁ and s ₃ denote indices that are coprime with the length of the sequence (N or 3N). In this example, when sequences of length 1 RB are generated using the cyclic extension method and a sequence of length 3 RBs are generated using the truncated sequence generation method, s ₁ may be 1, 2, ..., 10, and s ₂ may be 1, 2 , ..., 36. N ₁ may be equal to 11, and N ₂ may be equal to 37.

On this basis, if a sequence of 1 RB length overlaps in the region of the first 12 subcarriers of a sequence of 3 RB length, the cross-correlation value between the two sequences can be expressed as follows.

[Equation 10]

For d = 0,

приближается к нулю, последовательности, указанные этими индексами последовательности, дают высокую взаимную корреляцию. Таким образом, если рассматривать последовательность длиной 1 RB и последовательность длиной 3 RB, предусмотрено группирование комбинации индексов последовательности, которая удовлетворяет условию, что выражение

приближается к нулю.From equation 10 it follows that if the combination of sequence indices (s ₁ and s ₃ ) satisfies the condition that the expression

approaches zero, the sequences indicated by these sequence indices give a high cross-correlation. Thus, if we consider a sequence of length 1 RB and a sequence of length 3 RB, it is possible to group a combination of indices of a sequence that satisfies the condition that the expression

approaching zero.

To confirm the relationship at the position where the overlap occurs, consider the case where a sequence of 1 RB length overlaps in the region of the second 12 subcarriers of a sequence of 3 RB length. In this case, the value of cross-correlation between these two sequences can be expressed as follows.

[Equation 11]

For d = 0,

When a 1 RB sequence is overlapped in the region of the last 12 subcarriers of a 3 RB sequence, the cross-correlation value can be expressed as follows.

[Equation 12]

For d = 0,

приближается к нулю, последовательности, указанные этими индексами последовательности, дают высокую взаимную корреляцию. Поэтому, если рассматривать последовательность длиной 1 RB и последовательность длиной 3 RB, позиция, где происходит перекрывание, не изменяет условие группирования.According to equations 11 and 12, we can also conclude that if the combination of sequence indices (s ₁ and s ₃ ) satisfies the condition that the expression

approaches zero, the sequences indicated by these sequence indices give a high cross-correlation. Therefore, if we consider a sequence of length 1 RB and a sequence of length 3 RB, the position where the overlap does not change the grouping condition.

According to the above examples, this embodiment provides for grouping sequences so that two sequence indices from all sequence indices grouped into the same group satisfy the condition that the expression (s ₂ / N ₂ - s ₁ / N ₁ ) approaches zero if we consider two sequences having a length of N ₁ and N ₂ . Here, N ₁ and N ₂ can be maximal mutually prime numbers that are less than the resulting reference signal sequence. In addition, s ₁ and s ₂ denote the root indices of the ZC sequences and can be selected in the ranges 1 ~ (N ₁ -1) and 1 ~ (N ₂ -1), respectively.

Based on this concept, we consider a more general grouping method taking into account sequences of different lengths.

In FIG. 3-5 show generalized sequence grouping schemes according to one embodiment of this invention.

According to this embodiment, from sequences of different lengths, for example 1 RB, 2 RB, 3 RB ... as shown in FIG. 3, sequences whose indices satisfy the aforementioned high cross-correlation condition can be grouped into the same group. Each of the groups of sequences can be allocated to the same cell or Node B.

In general, FDM is performed for each cell or Node B, which minimizes interference due to the use of sequences having different lengths of one cell or Node B. Thus, by highlighting sequences having high cross-correlation, the same cell or Node B, inter-cell interference due to the use of sequences of different lengths can be minimized.

Another variant implementation of this invention provides for the implementation of grouping so that each of the groups contains at least one sequence of each length. Thus, the allocation of a group of sequences to the same cell or Node B allows the UE located in this cell or Node B to use reference signal sequences of different lengths. However, a particular grouping method can be defined in different ways.

First, the number of sequences allocated to one group may be proportional to the number of RBs, which corresponds to the length of the reference signal sequence. According to FIG. 3 one sequence for a 1 RB sequence, two sequences for a 2 RB sequence, 3 sequences for a 3 RB sequence, etc. are grouped.

Secondly, the number of sequences allocated to one group can be a constant number. According to FIG. 4, one sequence for a sequence of each length is grouped into the same group.

This embodiment may include grouping so that the number of sequences allocated to one group is neither proportional to the length of the sequence, nor constant. In FIG. 5 shows an example of grouping sequences so that one sequence for a 1 RB sequence, 2 sequences for a 2 RB sequence, 2 sequences for a 3 RB sequence and 3 sequences for a 4 RB sequence, etc. grouped into one group.

Similar to the above, if each group contains at least one sequence of each length, you can specify the maximum number of sequences per group. When the maximum number of sequences per group is specified, the method for selecting the index of the root of the sequence ZC, taking into account the limit on the number of sequences, can be specified as follows.

If for each length of the sequence one sequence is selected and if one particular sequence with index s ₁ and length N _{1 is} already selected for this group, you can select one sequence (having index s ₂ ) for each length whose index makes the expression (s ₂ / N ₂ -s ₁ / N ₁ ) closest to zero, where N ₂ is the length of the sequence corresponding to the length in question. If, for each specific sequence length, 2 sequences are selected and if one particular sequence with index s ₁ and length N _{1 is} already selected for this group, two sequences for this length can be selected so that the expression (s ₂ / N ₂ -s ₁ / N ₁ ) tended to zero. This can be further generalized to the maximum number of x sequences for each length.

Another grouping method can be defined as follows. If one sequence is selected for each sequence length and if one particular sequence with index s ₁ and length N _{1 is} already selected for this group, we first select a certain number (y) of sequences from sequences for which the expression (s ₂ / N ₂ -s ₁ / N ₁ ) approaches a certain value, and then select one sequence of y sequences that has a high cross-correlation with a sequence having index s ₁ . If, for each specific sequence length, 2 sequences are selected and if one particular sequence with index s1 and length N1 is already selected for this group, we first select a certain number (y) of sequences from sequences for which the expression (s ₂ / N ₂ -s ₁ / N ₁ ) approaches a certain value, and then we select two sequences from y sequences that have a high cross-correlation with a sequence having index s ₁ . This can be further generalized to the maximum number of x sequences for each length.

In the above examples, one particular sequence is first selected with index s ₁ and length N ₁ , which becomes a support for selecting the remaining sequences. This reference sequence can be defined as a 1 RB sequence, a 2 RB sequence, a 3 RB sequence, etc. However, in the following explanation, suppose that the reference sequence is a 3 RB sequence. Since the number of sequence indices for a length of 3 RB is 30, the number of groups for grouping sequences according to this embodiment of the invention may be 30.

In the case where the number of root indices for a sequence of 3 RB length is 30, the number of root indices selected for a particular group can be determined as follows.

[Equation 13]

round (the number of root indices for a sequence of a certain length / 30)

Here, "round (z)" is a function of rounding to the nearest integer closest to z.

According to equation 13, for a length of 3 RB and 4 RB, 1 sequence can be selected. For a length of 5 RB ~ 6 RB, 2 sequences can be selected. In addition, for lengths greater than 6 RB, 3 or more sequences can be selected, respectively. In addition, according to one embodiment of this invention, a sequence of length less than 3 RB lengths can be set differently without using a ZC sequence. Because of this, the number of sequences selected for a length of 1 RB and a length of 2 RB can be set to 1.

As a result, according to this embodiment, the number of sequences per group can be set as follows.

[Equation 14]

{1RB, 2RB, 3RB, 4RB, 5RB, 6RB, 8RB, 9RB, 10RB, 12RB, 15RB, 16RB, 18RB, 20RB, 24RB, 25RB, ...} = {1, 1, 1, 1, 2, 2, 3 , 3, 4, 5, 6, 6, 7, 8, 9, 9, ...}

On this basis, the following tables 1-5 show an example of grouping sequences in which each group contains the number of sequences according to equation 14, and the sequences selected for each group satisfy the above condition of high cross-correlation.

In tables 1-5, sequences of length 1 RB and 2 RB are not shown, since sequences of length 1 RB and 2 RB are specified differently.

In another example, in the case where the number of root indices for a sequence of 3 RB is 30, the number of root indices selected for a particular group can be determined as follows.

[Equation 15]

floor (number of root indices for a sequence of a certain length / 30)

Here floor (z) is a function that returns the largest integer not exceeding z.

According to equation 15, 1 sequence can be selected for a length of 3 ~ 5 RB. For a length of 6 ~ 8 RB, 2 sequences can be selected. In addition, for lengths over 9 RB lengths, 3 or more sequences can be selected, respectively. In addition, according to one embodiment of this invention, a sequence of length less than 3 RB lengths can be set differently without using a ZC sequence. Because of this, the number of sequences selected for a length of 1 RB and a length of 2 RB can be set to 1.

As a result, according to this embodiment, the number of sequences per group can be set as follows.

[Equation 16]

{1RB, 2RB, 3RB, 4RB, 5RB, 6RB, 8RB, 9RB, 10RB, 12RB, 15RB, 16RB, 18RB, 20RB, 24RB, 25RB, ...} = {1, 1, 1, 1, 1, 2, 2 , 3, 3, 4, 5, 6, 7, 7, 9, 9, ...}

On this basis, the following tables 6-8 show an example of grouping sequences in which each group contains the number of sequences according to equation 16, and the sequences selected for each group satisfy the above condition of high cross-correlation.

Like tables 1-5 in tables 6-8, sequences of 1 RB and 2 RB are not shown, since sequences of 1 RB and 2 RB are specified differently.

In yet another embodiment of this invention, the maximum number of sequences per group can be predetermined for various reasons. The following tables 9 and 10 show an example of a case where the maximum number of sequences per group is limited to 5 sequences.

In another example, the maximum number of sequences can be predetermined equal to 4. This case is shown in the following tables 11 and 12.

In another example, the maximum number of sequences can be predetermined equal to 3. This case is shown in the following tables 13 and 14.

In another example, the maximum number of sequences can be predetermined equal to 2. This case is shown in the following tables 15 and 16.

In another example, the maximum number of sequences can be predetermined equal to 1. This case is shown in the following table 17.

Given the complexity of the configuration and the flexibility to support the ability of the UE to use a variable length reference signal sequence, one embodiment of this invention provides for grouping such that each group contains one base sequence of each length corresponding to a length of 1 to 5 RB and two base sequences of each length corresponding to a length of 6 RB or more. This corresponds to tables 15 and 16.

Here, the base sequence is the ZC sequence indicated by the root index, which is used to apply a cyclic shift corresponding to different values of the cyclic shift. The base sequence with a cyclic shift can be used as a reference signal sequence.

In the above tables 1-17 reflects the case of selecting the index (s) of the root using the expression (s ₁ / N ₁ -s ₂ / N ₂ ). However, in yet another embodiment of this invention, the root index (s) can be selected by calculating the actual cross-correlation value. The following tables 18-0 correspond to tables 6-8, but the root indices are selected by calculating the actual cross-correlation value.

In this case, if the maximum number of sequences for each group is predetermined to be 5, grouping can be performed according to the following tables 21 and 22. Tables 21 and 22 also reflect the case of selecting the base sequences by calculating the actual cross-correlation.

In another example, if the maximum number of sequences for each group is predetermined equal to 4, grouping can be performed according to the following tables 23 and 24. Tables 23 and 24 also reflect the case of selecting the base sequences by calculating the actual cross-correlation.

In another example, if the maximum number of sequences for each group is predetermined to be 3, grouping can be performed according to the following tables 25 and 26. Tables 25 and 26 also reflect the case of selecting the base sequences by calculating the actual cross-correlation.

In another example, if the maximum number of sequences for each group is predetermined equal to 2, grouping can be performed according to the following table 27. Table 27 also reflects the case of selecting the base sequences by calculating the actual cross-correlation.

In another example, if the maximum number of sequences for each group is predetermined equal to 1, grouping can be performed according to the following table 28. Table 28 also reflects the case of selecting the base sequences by calculating the actual cross-correlation.

For the above cases, the tables can be reorganized according to the allocated number of sequences for each group and each length.

For another example of this invention, the above tables can be extended to a length of 100 RB, and the following tables show this example. In this example, the maximum root index number (v) for a length of 5 RB or less is set to 1, and the maximum root index number (v) for a length greater than 5 RB is set to 2.

Based on these concepts, the present invention provides the following method for generating a reference signal sequence using the ZC sequence.

To generate a reference signal sequence, one embodiment of the present invention defines a particular base sequence for applying a cyclic shift. In this embodiment, the base sequence is specified using a ZC sequence with a specific root index (hereinafter “q”). A particular base sequence is selected from groups of base sequences, and each group of base sequences contains base sequences having high cross-correlation, as described above. Therefore, if you want to select a specific base sequence with the index "q", "q" should be selected taking into account the index of the group (hereinafter "u") and the index of the number of the basic sequence in each group (hereinafter “v"). Thus, q must be a function of u and v.

After selecting a particular base sequence with the root index “q”, a cyclic shift corresponding to various values of the cyclic shift can be applied to the selected base sequence.

If we consider the relationship between “q”, “u” and “v” more specifically, “q” can be obtained according to the following equations. The following equations 17 and 18 allow you to choose the index "q" that meets the condition that the expression (s ₁ / N ₁ -s ₂ / N ₂ ) approaches zero.

[Equation 17]

Where

[Equation 18]

Where

- длина генерации корневой последовательности ZC, используемой при генерации q-й корневой последовательности ZC, и

задается наибольшим простым числом, которое меньше размера соответствующей опорной сигнальной последовательности. Таким образом, базовая последовательность генерируется методом циклического расширения.Here

is the generation length of the root sequence ZC used to generate the qth root sequence ZC, and

is given by the largest prime number that is less than the size of the corresponding reference signal sequence. Thus, the base sequence is generated by the cyclic extension method.

- это длина, заданная наибольшим простым числом, которое меньше размера опорной последовательности, например длины 3 RB. Если группирование основано на длине 3 RB, то

равно 31. "round(z)" является функцией округления до ближайшего целого числа, ближайшего к z, и "floor(z)" является функцией, возвращающей наибольшее целое число, не превышающее z.Besides,

is the length given by the largest prime number that is less than the size of the reference sequence, for example, a length of 3 RB. If the grouping is based on a length of 3 RB, then

equals 31. "round (z)" is a function of rounding to the nearest integer closest to z, and "floor (z)" is a function that returns the largest integer not exceeding z.

можно задать наименьшим простым числом, которое больше размера соответствующей опорной сигнальной последовательности. В этом случае

может представлять собой длину, заданную наименьшим простым числом, которое больше размера опорной последовательности, например длины 3 RB. Если группирование основано на длине 3 RB, то

может быть равно 37.According to another embodiment of this invention, if the ZC sequence is generated by truncation, then

can be specified with the smallest prime number that is larger than the size of the corresponding reference signal sequence. In this case

may be the length given by the smallest prime number that is larger than the size of the reference sequence, for example, a length of 3 RB. If the grouping is based on a length of 3 RB, then

may be equal to 37.

) можно выразить следующим образом.According to equations 17 and 18, the “m” th element of the “q” th sequence ZC (

) can be expressed as follows.

[Equation 19]

означает, что если "y" имеет значение 0.5 или более в своем десятичном разряде,

Since "round (y)" and "floor (y + 0.5)" are actually equivalent, equations 17 and 18 have the same meaning. In equations 17 and 18, the expression

means that if "y" has a value of 0.5 or more in its decimal place,

можно вычислить как "-1". Поэтому

can be calculated as "1", and if "y" has a value less than 0.5 in its decimal place,

can be calculated as "-1". therefore

или любыми другими эквивалентными выражениями, имеющими один и тот же смысл.can be replaced

or any other equivalent expressions that have the same meaning.

может быть равно 31. Кроме того, при генерации последовательности ZC методом усечения

может быть равно 37. Когда группирование осуществляется для длины свыше 4 RB на основании длины 4 RB и при генерации последовательности ZC на основании метода циклического расширения,

может быть равно 47. Кроме того, при генерации последовательности ZC методом усечения,

может быть равно 49. Это можно легко применить к группированию на основании другой длины.In the above examples, when grouping is performed for a length of more than 3 RB based on a length of 3 RB and when generating a ZC sequence based on a cyclic extension method,

may be 31. In addition, when generating a ZC sequence by truncation

may be equal to 37. When grouping is performed for a length of more than 4 RB based on a length of 4 RB and when generating a ZC sequence based on the cyclic extension method,

may be 47. In addition, when generating a ZC sequence by truncation,

may be 49. This can easily be applied to grouping based on a different length.

The above tables can be obtained from equations 17 and 18. The following examples demonstrate the choice of the root index according to equations 17 and 18.

First, if N _{reference, zc} = 31, the method of selecting the first group ("u" = 0), when 1) N _zc = 47, 2) N _zc = 71, 3) N _zc = 211, is as follows. In the following examples, equation 18 is used.

Therefore, for a length of 4 RB, the number of first base sequences (v = 0) in the first group (u = 0) is 2 (q = 2).

Therefore, for a length of 6 RB, the number of first base sequences (v = 0) in the first group (u = 0) is 2 (q = 2).

Therefore, for a length of 18 RB, the number of first base sequences (v = 0) in the first group (u = 0) is 7 (q = 7).

For the above cases, the selected root indices (q) correspond to the data in Tables 6-8, which are generated based on a length of 3 RB.

In another example, if N _{reference, zc} = 47 (based on the length of 4 RB), the method of choosing the second group (u = 1), when 1) N _zc = 59, 2) N _zc = 107, 3) N _zc = 139 carried out as follows. In the following examples, equation 18 is used.

Therefore, for a length of 5 RB, the number of first base sequences (v = 0) in the second group (u = 1) is 3 (q = 3).

Therefore, for a length of 9 RB, the number of first base sequences (v = 0) in the second group (u = 1) is 5 (q = 5).

Therefore, for a length of 12 RB, the number of first base sequences (v = 0) in the second group (u = 1) is 6 (q = 6).

In yet another embodiment of the present invention, equations 17 and 18 can be replaced as follows.

[Equation 20]

Where

[Equation 21]

Where

Since “round (y)” and “floor (y + 0.5)” are actually equivalent, equations 20 and 21 have the same meaning.

In yet another embodiment of the present invention, equations 17 and 18 can be replaced as follows.

[Equation 22]

Where

[Equation 23]

Where

These equations correspond to various grouping methods explained in connection with the above tables.

If the maximum number of sequences that can be grouped into one group is predetermined equal to 2, equations 17-18, 20-21, and 22-23 can be simplified as follows.

[Equation 24]

Where

[Equation 25]

Where

[Equation 26]

Where

[Equation 27]

Where

[Equation 28]

Where

[Equation 29]

Where

Equations 17-18, 20-21 and 22-23 allow you to choose the root index ZC, corresponding to the condition that the expression (s ₁ / N ₁ -s ₂ / N ₂ ) approaches zero. Equations 24-29 allow you to choose the root index ZC, when the maximum number of sequences of each length per group is limited to 2.

However, if we generalize these equations so that the expression (s ₁ / N ₁ -s ₂ / N ₂ ) approaches the specific value (T), we can obtain the following equations. In this case, the value of "T" may be 0, 1/2, -1/2, 1/3, -1/3. However, the value of "T" may have a different meaning.

In the following equations, equations 30 and 31 allow you to select the root index ZC when the maximum number of sequences of each length per group can have a maximum value. Equations 32 and 33 allow you to choose the root index ZC, when the maximum number of sequences of each length per group is limited to 2.

[Equation 30]

Where

[Equation 31]

Where

[Equation 32]

Where

[Equation 33]

Where

Specialists in the art it is obvious that the present invention can be implemented in other specific forms, without going beyond the essence and distinguishing features of the invention. Thus, the above embodiments should be considered in all respects as illustrative, but not restrictive. The scope of the invention should be determined by a reasonable interpretation of the claims, and all changes falling within the equivalent scope of the invention should be included in the scope of the invention.

Industrial application

According to embodiments of the present invention, inter-cell interference due to the use of variable length sequences can be minimized. If each grouped base sequence is allocated to a specific cell or Node B, the UE may use a variable length sequence as a reference signal.

These methods are suitable for use in the 3GPP LTE (3rd Generation Partnership Project Long Term Evolution) system. However, it will be apparent to those skilled in the art that these methods can be applied to any wireless communication system using sequences of different lengths as reference signal sequences.

Claims (16)

1. The method of transmitting a reference signal sequence on the transmitting side, the method comprises the steps of obtaining one or more base sequences having a variable length corresponding to one or more sizes of a resource block, applying a cyclic shift corresponding to a variable value of a cyclic shift to one or more base sequences for generating a reference signal sequence, and transmit the reference signal sequence to the receiving side, and bases stems sequence obtained from a plurality of base sequences, divided into groups of sequences, and each group of sequences comprises at least one base sequence of each length. 2. The method according to claim 1, in which the size of the resource block corresponds to the size of 12 subcarriers in the frequency domain. 3. The method according to claim 1, in which one or more basic sequences are obtained by cyclic extension of the ZC sequence having a length

given by the largest prime number that is less than the corresponding length of the reference signal sequence. 4. The method according to claim 1, in which one or more basic sequences are obtained by truncating a ZC sequence having a length

given by the smallest prime number that is greater than the corresponding length of the reference signal sequence. 5. The method according to claim 1, in which the number of sequence groups is 30. 6. The method according to claim 5, in which each of the groups of sequences contains one base sequence of each length corresponding to from 1 to 5 sizes of the resource block, and two basic sequences of each length corresponding to 6 or more sizes of the resource block. 7. The method according to claim 6, in which the base sequence having a length corresponding to 3 or more resource block sizes is obtained using a Zadova-Chu sequence (ZC) with a specific index q of the ZC sequence, and in which the base sequence having a length, corresponding to 1 or 2 resource block sizes are obtained using a different sequence than the ZC sequence. 8. The method according to claim 7, in which the specified index q of the sequence ZC is a function of the index u of the group and the index v of the number of the base sequence in the group. 9. A method for transmitting a reference signal sequence using a Zadoff-Chu (ZC) sequence on the transmitting side, the method comprises the steps of obtaining a basic sequence using the qth root sequence ZC from a plurality of base sequences, and a set of base sequences is divided into groups, and q is a function of index u of the group of sequences and index v of the number of the basic sequence in the group of sequences;
apply the cyclic shift corresponding to the variable value of the cyclic shift to the obtained base sequence to generate a reference signal sequence and transmit the reference signal sequence to the receiving side. 10. The method according to claim 9, in which the number u is 30 and the number v is set based on the length of the base sequence. 11. The method according to claim 9, in which a given index q of the sequence ZC is determined according to

Where

u∈ {0,1, ..., 29},

Where

- the length of the generation of the root sequence ZC used to generate the qth root sequence ZC,

is set by the largest prime number that is less than the length of the obtained base sequence,

is a given reference prime, and floor (z) is a function that returns the largest integer not exceeding z. 12. The method according to claim 9, in which a specific index q of the sequence ZC is determined according to

Where

u∈ {0,1, ..., 29},

Where

- the length of the generation of the root sequence ZC used to generate the qth root sequence ZC,

is set by the smallest prime number that is greater than the length of the obtained base sequence,

is a given reference prime, and floor (z) is a function that returns the largest integer not exceeding z. 13. The method according to claim 9, in which the maximum number of base sequences in each group of sequences is 2 and in which the index q of the sequence ZC is determined according to

Where

u∈ {0,1, ..., 29}, ν∈ {0,1},
Where

- the length of the generation of the root sequence ZC used to generate the qth root sequence ZC,

is set by the largest prime number that is less than the length of the obtained base sequence,

is a given reference prime, and floor (z) is a function that returns the largest integer not exceeding z. 14. The method according to claim 9, in which the maximum number of base sequences in each group of sequences is 2 and
in which the given index q of the sequence ZC is determined according to

Where

u∈ {0,1, ..., 29}, ν∈ {0,1}
Where

- the length of the generation of the root sequence ZC used to generate the qth root sequence ZC,

is set by the smallest prime number that is greater than the length of the obtained base sequence,

is a given reference prime, and floor (z) is a function that returns the largest integer not exceeding z. 15. The method according to claim 11 or 13, in which the specified reference prime number is 31. 16. The method according to item 12 or 14, in which the specified reference prime number is 37.

Images