KR20040006405A - Switching method for combining and dividing multi-beam in a smart antenna system - Google Patents

Abstract

PURPOSE: A switching method for synthesizing and distributing multi-beam in a smart antenna system is provided to operate an adaptive sector system by optimizing each sector. CONSTITUTION: Vectors of P and Q are defined(S120). Groups of M number are set up by dividing the vector Q into M and a maximum index of the vector Q of each group is set up as an initial value of Am(S140). A vector SID_G1 is set up by using a vector BID_G1 of the first groups of M number(S170). Vectors SID_Gm of m group are set up clockwise(S180). The vectors BID_Gm are compared to the SID_m(S200). Vectors BID_GM are compared to vectors SID_GM(S210). Vectors SIDk are set up by using SID_G1 and SID_Gm(S220). The vector P and the vectors SIDk are rotated counterclockwise(S230). A value R of a sector ratio is calculated(S240). An optimum value of the sector ratio and the SID are decided(S250).

Description

Switching method for combining and dividing multi-beam in a smart antenna system

The present invention relates to a smart antenna, and more particularly, to a method for constructing an optimal sector by synthesizing and distributing multiple beams so that the call volume of each sector can be evenly distributed in a mobile communication system.

1 schematically shows a three sector antenna conventionally used.

The existing mobile communication base station uses the three sector antenna 10 as shown in FIG. 1 to increase the capacity of the cell. The three-sector antenna 10 divides one cell into three sectors, and has sector antennas in each of the sectors 20, 30, and 40 to transmit and receive signals of each sector. Since the direction and width are fixed, the amount of interference can be reduced to 1/3 compared to a single cell when the user distribution is uniform.

However, when the number of users increases in one cell of each of the sectors 20, 30, and 40, it is necessary to solve the problem by increasing the capacity of the cell by increasing the frequency. There is a problem that the efficiency of resource use is poor.

In order to overcome the above problems, a smart antenna system has emerged as a new method that can greatly increase the capacity of the system. The smart antenna system uses a series of array antennas to maintain the antenna beam pattern in a desired user direction to receive a maximum of a desired user signal and to suppress other interference signals by suppressing a signal of another user.

An example of a smart antenna system is disclosed in patent application No. 2001-76330 filed by the applicant of the present invention under the name multi-beam synthesis and distribution system and method thereof, which is presented by reference of the present invention. Patent application No. 2001-76330 discloses a method of determining a sector beam pattern based on the maximum value of channel resources when estimating the amount of channel resources transmitted or received in multiple beams.

However, according to the conventional method, there is a problem in that an optimal sector beam pattern may not be found, which may lower the utilization efficiency of radio channel resources.

The present invention has been made in view of the above problems, and an object of the present invention is to provide a method for constructing an optimal sector so that the currency amount of each sector can be evenly distributed.

In the smart antenna system according to the present invention for achieving the above object, a switching method for multi-beam synthesis and distribution includes a) a vector P and a vector P obtained by rotating a call amount vector E used in N multiple beams clockwise k times; Defining a cumulative sum vector Q of; b) dividing the vector Q into M groups and setting the maximum index of the vector Q belonging to each group to an initial value of A _m , respectively; c) comparing the vector BID_G ₁ composed of the candidate sector numbers of the first group among the M groups and setting the SID_G ₁ ; d) when the SID_G ₁ is set, the SID_G _m of the m-th group having a range of 2 or more (M-1) or less are automatically set sequentially in the clockwise direction; e) comparing the vector BID_G _m composed of candidate sector numbers of the m th group among the M groups with SID_G _m ; f) comparing the vector BID_G _M consisting of the candidate sector numbers of the M-th group with SID_G _M ; g) setting a sector number vector SID _k of multiple beams using SID_G _m having the range of 1 or more M or less; h) rotating the vector P and SID _k counterclockwise k times; i) calculating a value of a sector ratio coefficient R according to the amount of currency of each sector; And j) determining the optimum value and the SID of the sector ratio coefficient R by executing the steps a) to i) N times so that k has a range of 1 or more and N or less. .

Therefore, according to the present invention, the call transmission capacity can be increased by reducing the load ratio of the base station through the optimal sector division.

1 schematically illustrates a conventional three sector antenna pattern.

2 is a general configuration diagram of a mobile communication system using multiple beams.

3 is a flow chart for switching for multi-beam synthesis and distribution in accordance with the present invention.

4 shows the distribution of the vector Q according to the invention.

<Explanation of symbols for the main parts of the drawings>

100: N multiple beams

200: multiple 1: D switches

300: synthesizer / distributor

400: M sector base station

500: control unit

600: call volume estimation device

Hereinafter, the present invention will be described in more detail with reference to the accompanying drawings.

Since the smart antenna system can be implemented in various structures according to the design method, an embodiment of the present invention will be described under the following assumption.

[Assumption 1] The base station configures M sectors with N multiple beams pointing in different directions to provide mobile communication services.

[Assumption 2] The number of N multiple beams is sequentially set from 1 to N in the clockwise direction.

[Assumption 3] The number of M sectors is set sequentially from 1 to M in the clockwise direction.

Assumption 4 Each beam has a service area of 2π / N and there is no interference signal from an adjacent beam.

Assumption 5 The call volume used in each beam can be measured.

[Assumption 6] Each beam is set to one of the D sectors by a 1: D (where D ≦ M) switch.

Assumption 7 Each sector is composed of at least two beams to provide communication services.

It will be appreciated by those skilled in the art that the present invention may be applicable to smart antenna systems of various structures when the assumptions 6 and 7 are modified.

2 is a configuration diagram illustrating a smart antenna system to which a switching method for multi-beam synthesis and distribution is applied in a smart antenna system according to an embodiment of the present invention.

Referring to FIG. 2, the smart antenna system to which the present invention is applied includes N multiple beams 100, a plurality of 1: D switches 200, a synthesizer / divider 300, an M sector base station 400, a controller ( 500, and the call amount estimating apparatus 600.

Each of the N multiple beams 100 points in the direction (2i-1) π / N (where 1 ≦ i ≦ N) and has a 2π / N beam width. The N multiple beams 100 are connected to one of the D sectors configurable by the N 1: D switches 200 and then to the M sectors 400 by the synthesizer / divider 300. Is assigned.

The control unit 500 collects the call amount data used in the N multiple beams from the M sector base station 400 or the call amount estimating apparatus 600, and the call amount is equally distributed to each sector according to the present invention. 1: D switch 200 is controlled.

In case that only the call volume data of M sectors can be collected by the M sector base station 400 or the call volume estimating apparatus 600, the call volume distribution used in each beam constituting the sector is assumed to be equal or non-uniform. N multi-beam call volume data can be estimated and used from the call volume data.

The call amount estimating apparatus 600 measures a call amount used in each of the N beams (or M sectors).

3 is a flowchart illustrating a switching method for multi-beam synthesis and distribution of the system shown in FIG.

First, E, BID, and SID are defined, respectively (S100). E is a call amount vector, which is a normalized call amount for each of the N multiple beams 100, and is defined as in Equation 1, and the sum thereof is 1.

E = [e (1), e (2), ..., e (N-1), e (N)] ^T

Further, the i-th (where 1 ≦ i ≦ N) beam is connected to D sectors through a 1: D switch and is set in one of the D sectors. Accordingly, the BID (i) is a vector consisting of the D candidate sectors for the i th (where 1 ≦ i ≦ N) beam, and is defined as Equation 2 below, and the SID (i) is the i th Sector number to be connected to the beam.

[Equation 2] BID (i) = [b _i (1), b _i (2), ..., b _i (D-1), b _i (D)] ^T

Subsequently, in step S110, the vector BID including the call amount vector E and the candidate sector number is rotated k times in the clockwise direction (S110). When N beams sequentially defined from clockwise 1 to N times are rotated k times (but 1kN) in the clockwise direction, the corresponding call amount vector E is defined as Equation 3 below.

E _k = [e (N−k + 1), ..., e (N), e (1), ..., e (Nk)] ^T

If the BID of each beam is also rotated k times in the clockwise direction, BID _k (i) is equal to BID (N-k + i) when ik, and BID (ik) when i> k.

The vector E _k rotated k times in step S120 is defined as a vector P as shown in Equation 4 below. The cumulative sum vector of the vector P is defined as shown in Equation 5, and the vector Q is calculated from Equation 6. do.

[Equation 4] E _k = [p (1), p (2), ..., p (N-1), p (N)] ^T

Equation 5 Q = [q (1), q (2), ..., q (N-1), q (N)] ^T

[Equation 6] (Where 1≤m≤N)

It is assumed that the sector configuration always starts at p (1) and proceeds clockwise. The assumption is made by rotating the vector E N times so that p (1) has the opportunity of sector construction starting from each N beams.

In order to distribute the call evenly over M sectors composed of N multi-beams, it is like dividing the vector P with M to an optimum condition. Therefore, first depending on the vector Q value, 0-1 / M, 1 / M-2 / M, ..., (M-2) / M- (M-1) / M, (M-1) / M- As in 1, divided into M groups (S130).

After dividing into M groups, an initial value of the index A _m (where 1 ≦ _m ≦ M) of the vector Q is set (S140). In each group, set the maximum indices of array Q to A ₁ , A ₂ , ..., A _M-1 , A _M , respectively. Thus, the first group is the initial setting from the first beam to the _first beam A, m (stage, 2≤m≤M-1) th groups A _m to the _m-th beam from the _A-1 +1 second beam. Of Figure 4 is 12 (N = 12) of the multi-beam 3 (M = 3), separated by sectors k times in a clockwise direction (however, 1≤k≤12) when rotating hayeoteul A _1, A _2, A ₃ This is an example of setting the initial value to 4, 9, 12.

If A _m , the index of the vector Q, does not satisfy [Assumption 7], the value of A _m is forcibly changed to have two beams as shown in Equations 7 to 9 below.

Equation 7 When A ₁ <2, A ₁ = 2

Equation 8 When A _m -A _m-1 <2, A _m = A _m-1 +2 (where 2mM-1)

Equation 9, if A _M -A _M-1 <2 A _M-1 = A _M -2

The candidate sector numbers of the vector BID_G _{1 including} the candidate sector numbers of the first group among the M groups divided by M in step S130 are compared (S150). Each beam is therefore of the vector BID _k of D x 1 size consisting candidate sector number of the first group A ₁ gae beam of vectors BID _k D x A ₁ size of the matrix consisting of _{BID_G 1 = [BID k (1} ), BID _k (2), ..., BID _k (A ₁ -1), BID _k (A ₁ )].

SID_G ₁ is defined as a sector number for which the first group is to be set. D-th element of any BID _k (m) (However, 1≤m≤A ₁₎ Having a candidate sector number that matches d (stage, 1dD) th element of _{BID k (1) BID k (} 1) is SID_G _{Becomes 1} The number of SID_G ₁ is defined as I ₁ .

If the value of I ₁ is zero (0), A ₁ must be reset (S160). Since the sector configuration always starts at p (1) and proceeds clockwise, all BID _k (m) (where 1≤m≤A ₁ ', A ₁ '<A ₁ ) is the d (where BID _k (1) , 1≤d≤D) a ₁ in the case with the sector number of candidates that match the first element, set the candidate sector is maximized in SID_G ₁ and the a ₁ a ₁ at that time, it is reset to. In one embodiment, when BID_G ₁ is [3 3 1 1 2; 1 1 2 2 3], BID _k (1) = [3 1] BID _{k of} beams belonging to the first element 3 of ^T and BID_G ₁ ( m) (where 1≤m≤5) has two consecutively matching beams from the left, and the second element 1 of BID _k (1) and four consecutively matching beams from the left are four. ₁ is reset from 5 to 4 and the SID_G ₁ is set to 1.

In addition, when the reset A ₁ does not satisfy [Assumption 7], the value of A ₁ is forcibly changed to have two beams. That is, if A ₁ <2, change it to A ₁ = 2.

If the I ₁ value is 1, the candidate sector number at that time is set to SID_G ₁ of the first group (S170). For example, the _{BID_G 1 [3 3 1 1 1} ; 1 1 2 2 2] is ₁ SID_G is set to 1 when.

When the value of I ₁ I ₁ is ≥2 and executes the algorithm to set to SID_G ₁ of the first group in turn to each _one SID_G repeated. For example, if BID_G ₁ is [3 3; 1 1], SID_G ₁ is set to 3 and 1 in order.

When SID_G ₁ of the first group is set, SID_G _m of the mth group is automatically set sequentially in the clockwise direction (where 2 ≦ _m ≦ _M ) (S180).

The m groups has the (A _m -A _m-1) pieces of the beam vector BID _k D x (A _m -A _m-1) the size of the matrix comprising the BID_G _m. If all BID _k (m) (A _m-1 + ₁ ≦ _m ≦ A _m ) of the mth group have candidate sector numbers that match SID_G _m , the number I _m of sectors that can be set in the mth group is 1, otherwise 0. However, here, the range of m value is 2 <= m <= M-1) (S200).

If I _m = 0, all BID _k (m) _{(However, A m-1 + 1≤m≤A m} ', A m'<A m) is A _m to have the candidate sector number that matches the SID_G _m To reset to A _m '(S190). For example, if BID_G _m is [3 3 1 1 1; 1 1 3 3 2] and SID_G _m is 3, A _m is reset from 5 to 4.

In addition, when the reset A _m does not satisfy [Assumption 7], the value of A _m is forcibly changed to have two beams as shown in Equations 10 and 11 below.

Equation 10 When A _m -A _m-1 <2, A _m = A _m-1 +2 (where 2≤m≤M-1)

[Equation 11] when A _M -A _M-1 <2 A _M-1 = A _M -2

If I _m = 1, A _m is set as it is and the above step S180 is repeated for the m + 1th group.

The M group is (A _M -A _M-1) has a dog D x (A _M -A _M-1) size of the matrix _M consisting of BID_G vector BID _k of the beams. If all BID _k (m) of the Mth group (where A _M-1 + ₁ ≦ m ≦ A _M ) have candidate sector numbers that match SID_G _M , the number of sectors I _M that can be set in the Mth group is If it is 1, otherwise it is 0 (S210).

If I _M = 0, the process returns to step S110 of the present algorithm, rotates k + 1 times, and repeats the present algorithm. However, if the number has more than one of SID_G _1, after resetting the SID_G ₁ sequentially executes the present algorithm repeatedly.

When I _M = 1, since the sector number SID_G _m (where 1 ≦ _m ≦ M) of the M groups is set, the sector number of the multiple beam of the vector P is set as shown in Equation 12 below (S220).

[Equation 12]

SID _k = [ones (A ₁ ) * SID_G ₁ , ones (A ₂ ) * SID_G ₂ , ..., ones (A _M-1 ) * SID_G _M-1 , ones (A _M ) * SID_G _M ,]

Where ones (A _k ) is a vector of 1 × A _k size consisting of 1 elements.

Since the vector P is the result of rotating the vector E clockwise k times, the sector number of the multiple beam corresponding to the vector E can be obtained by rotating the vector SID _k k times in the counterclockwise direction (S230). That is, SID (i) (where 1 ≦ i ≦ N) is SID _k (i + k) when i ≦ Nk, and SID _k (i + kN) when i ≧ Nk.

Since the sector number to which the N multiple beams are to be set is set, the sector ratio coefficient R according to the call volume of each sector can be calculated by the following equation (13).

[Equation 13] R = [1-MAX (E)] / [(N-1) * MAX (E)]

In order to quantify the degree of distribution of currency in each sector evenly, the sector ratio coefficient R is defined as the ratio of the sum of the currency of the sector with the most currency and the sum of the remaining sectors of currency. The sector ratio coefficient has a value ranging from 0 to 1, and the closer the sector ratio is to 1, the more equal the distribution of the call volume is. The closer to 0, the more the call volume is concentrated in any one sector.

When k≤N, the process returns to the step S110 of the flowchart and rotates k + 1 times to repeatedly execute the algorithm. If the number has more than one of SID_G _1, after resetting the SID_G ₁ sequentially executes the present algorithm repeatedly.

After rotating N times, the maximum value is selected from the calculated sector ratio coefficients (S250). The vector SID at that time is composed of sector numbers of the respective beams for optimally equalizing the distribution of call volume between sectors. As a result, the present algorithm is to find the optimal sector number SID from the candidate sector vector BID of each beam so that the sector ratio coefficient from the call volume vector E is maximized.

Embodiment of the present invention is only one embodiment, of course, many modifications and variations of the components of the present invention without departing from the gist of the present invention, of course, the present invention is not limited to the embodiment. .

As described in detail above, according to the switching method for multi-beam synthesis and distribution in the smart antenna system of the present invention, the adaptive sector system is efficiently operated by configuring an optimal sector so that the call volume of each sector can be evenly distributed. The capacity can be increased by reducing the load rate of the base station.

Claims (2)

a) defining a vector P obtained by rotating the call amount vector E used in N multiple beams clockwise k times and a cumulative sum vector Q of the vector P; b) dividing the vector Q into M groups and setting the maximum index of the vector Q belonging to each group to an initial value of A _m , respectively; c) comparing the vector BID_G ₁ composed of the candidate sector numbers of the first group among the M groups and setting the SID_G ₁ ; d) sequentially setting the SID_G _m of the m-th group according to the clockwise direction when the SID_G ₁ is set; e) comparing the vector BID_G _m composed of candidate sector numbers of the m th group among the M groups with SID_G _m ; f) comparing the vector BID_G _M composed of the candidate sector numbers of the M-th group with SID_G _M ; g) setting the sector number vector SID _k of the multiple beam using the SID_G ₁ and the SID_G _m , wherein the m value has a range of 2 or more (M-1) or less; h) rotating the vector P and SID _k counterclockwise k times; i) calculating a value of a sector ratio coefficient R according to the amount of currency of each sector; And j) performing the steps a) to i) N times so that k has a range of 1 or more and N or less to determine the optimal value and the SID of the sector ratio coefficient R. Switching method for multi-beam synthesis and distribution The method of claim 1, wherein the call volume of each of the multiple beams is calculated by dividing the call volume of the sector by the number (N / M) of the multiple beams forming each sector. Switching method.