Background
With the increasing demand for multimedia digital services, mobile communication systems have also placed higher demands on spectral efficiency and energy efficiency. To meet these needs, the 5G key technology includes several aspects as follows: the frequency spectrum efficiency of the system is improved by utilizing a large-scale antenna technology; adopting the bandwidth of a millimeter wave spectrum spreading system; and a multi-layer dense network is deployed to improve the frequency spectrum multiplexing degree of the system area. Among them, a communication system using a large-scale antenna array is called a large-scale MIMO system. The method not only can compensate the serious attenuation of millimeter wave band signal propagation, but also is an important means for realizing the wireless backhaul and interference control of the multilayer dense network. Therefore, the large-scale antenna technology is one of the key basic technologies of the 5G mobile communication system, and can lay an important basic role for the successful application of the 5G.
However, since the total number of users in the large-scale antenna system is always greater than the number of orthogonal pilots, the pilots need to be multiplexed between cells, and the pilot pollution caused by the pilot multiplexing between cells becomes a performance bottleneck of the large-scale antenna system. Therefore, reasonable multiplexing of pilots is needed between cells to mitigate pilot interference. At present, no effective, low-complexity pilot pollution mitigation solution close to real-world applications has emerged.
On the other hand, the existing research shows that the modulation mode of the user signal has an important influence on the system error code performance, and therefore, the degree of pilot frequency multiplexing is also influenced.
In view of the above situation, the present invention considers the modulation scheme of the user signal as another dimension for the whole pilot allocation.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a pilot frequency distribution method and a pilot frequency distribution system for combining modulation orders in a multi-cell system. Firstly, the invention divides users into two categories according to a dividing threshold which has parameters such as modulation order and base station density, and the like, wherein the two categories are respectively as follows: class I users that fully multiplex pilots and class II users that cannot fully multiplex pilots. Then, a random pilot frequency distribution strategy is adopted for class I users, and a user pair capable of multiplexing pilot frequency is found for class II users according to a certain principle so as to avoid pilot frequency pollution, improve system capacity, reduce system error rate and have lower complexity.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a pilot frequency distribution method combined with modulation orders in a multi-cell system comprises the following steps:
step S1, dividing the user types into I type users which can fully multiplex pilot frequency and II type users which can not fully multiplex pilot frequency;
step S2: dividing a pilot frequency set into two subsets which do not overlap with each other, wherein the number of orthogonal pilot frequencies of the first subset is equal to the number of I-type users of a reference cell;
step S3, randomly distributing orthogonal pilot frequency of the first subset for I-type users of the reference cell, wherein the I-type users of other cells and the I-type users of the reference cell share the orthogonal pilot frequency of the first subset;
step S4, randomly allocating orthogonal pilots of the second subset to the class II users of the reference cell, allocating class II users of other cells to a pilot multiplexing pair, and sharing one orthogonal pilot with the class II users of the reference cell.
Preferably, in step S1, if ρ is greater than the cell radius R, all users are class I users.
Preferably, step S1 is specifically as follows: classification of user categories: first, M users in each cell are classified into two categories according to ρ in formula (1). Users with the distance less than rho from the service base station are pilot frequency complete multiplexing users, and are called I-type users; and the users with the distance greater than or equal to rho from the service base station are called II-type users. Wherein,
it is assumed here that in a multi-cell massive MIMO system, there are L cells, and the number of M orthogonal pilots. Each user sends an uplink pilot signal, and the base station estimates a channel after receiving the pilot signal. Thereafter, the user processes the data signal to be transmitted using a Quadrature Amplitude Modulation (QAM) scheme. N represents QAM modulation order adopted by a user to send data signals, R represents cell radius, alpha represents path loss index, and lambda represents base station density.
Preferably, step S2 is specifically as follows: the pilot frequency set is divided into two subsets which do not overlap with each other:
and
wherein, aggregate
The number of orthogonal pilots in (b) is equal to the number of class I users of the reference cell.
Preferably, step S3 is specifically as follows: pilot allocation for class I users: randomly assigning sets for class I users of a reference cell
Orthogonal pilot in (1), shared set of class I users of other cells and class I users of reference cell
The orthogonal pilot in (1).
Preferably, step S4: random allocation of pilot frequency set for II type users of reference cell
Pilot frequency in (1); when the class II users of other cells are in the range of C and not in
Can be within the range of the reference cell, can be in contact with the reference user UE of the reference cell
0The same pilot is used. Wherein C denotes the base station BS in cell l
lCentered on r
0Is the inside of a circle of radius r
0Indicating that the user in cell l, who can multiplex the pilot, is far from its serving base station BS
lThe farthest distance of (c).
Representing reference user UE
0The pilots used are prohibited from being multiplexed within the area,
is a base station BS
0As the center of circle, in r
2Being the inside of a circle of radius. r is
0Calculated according to equation (2):
wherein r is1=D(BSl,UE0) And D (·,) represents the distance between two points. r is2Calculated from equation (3):
BS0a reference base station representing a reference cell.
The invention also discloses a pilot frequency distribution system of the combined modulation order in the multi-cell system, which comprises the following modules:
a user category division module: dividing user types into I type users which can completely multiplex pilot frequency and II type users which can not completely multiplex pilot frequency;
a pilot set segmentation module: dividing a pilot frequency set into two subsets which do not overlap with each other, wherein the number of orthogonal pilot frequencies of the first subset is equal to the number of I-type users of a reference cell;
the I-type user pilot frequency distribution module: randomly distributing orthogonal pilot frequency of a first subset for I-type users of a reference cell, wherein the I-type users of other cells and the I-type users of the reference cell share the orthogonal pilot frequency of the first subset;
a II-type user pilot frequency distribution module: and randomly distributing the orthogonal pilot frequency of the second subset for the II type users of the reference cell, and distributing the II type users of other cells into a pilot frequency multiplexing pair to share one orthogonal pilot frequency with the II type users of the reference cell.
The invention has the following beneficial effects:
the pilot frequency distribution technical scheme of the multi-cell large-scale antenna system reduces the influence of pilot frequency pollution, improves the system throughput and reduces the error rate.
Detailed Description
The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, but the present invention is not limited to the following embodiments.
The method mainly comprises three main stages of user classification, pilot frequency set division and pilot frequency distribution in the implementation process. The details are as follows:
example one
As shown in fig. 1, the pilot allocation system for combining modulation orders in a multi-cell system of this embodiment includes the following modules:
a user category division module: classification of user categories: first, M users in each cell are classified into two categories according to ρ in formula (1). Users with the distance less than rho from the service base station are pilot frequency complete multiplexing users, and are called I-type users; and the users with the distance greater than or equal to rho from the service base station are called II-type users. Wherein,
it is assumed here that in a multi-cell massive MIMO system, there are L cells, and the number of M orthogonal pilots. Each user sends an uplink pilot signal, and the base station estimates a channel after receiving the pilot signal. Thereafter, the user processes the data signal to be transmitted using a Quadrature Amplitude Modulation (QAM) scheme. N represents QAM modulation order adopted by a user to send data signals, R represents cell radius, alpha represents path loss index, and lambda represents base station density.
If ρ is larger than the cell radius R, all users are class I users.
A pilot set segmentation module: the pilot frequency set is divided into two subsets which do not overlap with each other:
and
wherein, aggregate
The number of orthogonal pilots in (b) is equal to the number of class I users of the reference cell.
The I-type user pilot frequency distribution module: pilot allocation for class I users: randomly assigning sets for class I users of a reference cell
Orthogonal pilot in (1), shared set of class I users of other cells and class I users of reference cell
The orthogonal pilot in (1).
A II-type user pilot frequency distribution module: random allocation of pilot frequency set for II type users of reference cell
Pilot frequency in (1); class II when other cell lUser, in the range of C and not in
Can be within the range of the reference cell, can be in contact with the reference user UE of the reference cell
0The same pilot is used. Wherein C denotes the base station BS in cell l
lCentered on r
0Is the inside of a circle of radius r
0Indicating that the user in cell l, who can multiplex the pilot, is far from its serving base station BS
lThe farthest distance of (c).
Representing reference user UE
0The pilots used are prohibited from being multiplexed within the area,
is a base station BS
0As the center of circle, in r
2Being the inside of a circle of radius. r is
0Calculated according to equation (2):
wherein r is1=D(BSl,UE0) And D (·,) represents the distance between two points. r is2Calculated from equation (3):
BS0a reference base station representing a reference cell.
Example two
As shown in fig. 2-3, for a multi-cell massive MIMO system, the pilot allocation method for combining modulation orders in a multi-cell system of this embodiment includes the following steps:
step S1: classification of user categories: first, M users in each cell are classified into two categories according to ρ in formula (1). Users with the distance less than rho from the service base station are pilot frequency complete multiplexing users, and are called I-type users; and the users with the distance greater than or equal to rho from the service base station are called II-type users. Wherein,
it is assumed here that in a multi-cell massive MIMO system, there are L cells, and the number of M orthogonal pilots. Each user sends an uplink pilot signal, and the base station estimates a channel after receiving the pilot signal. Thereafter, the user processes the data signal to be transmitted using a Quadrature Amplitude Modulation (QAM) scheme. N represents QAM modulation order adopted by a user to send data signals, R represents cell radius, alpha represents path loss index, and lambda represents base station density.
Step S2: the pilot frequency set is divided into two subsets which do not overlap with each other:
and
wherein, aggregate
The number of orthogonal pilots in (b) is equal to the number of class I users of the reference cell.
Step S3: pilot allocation for class I users: randomly assigning sets for class I users of a reference cell
Orthogonal pilot in (1), shared set of class I users of other cells and class I users of reference cell
The orthogonal pilot in (1).
Step S4: random allocation of pilot frequency set for II type users of reference cell
Pilot frequency in (1); when the other is smallClass II users of a zone, within the scope of C and not
Can be within the range of the reference cell, can be in contact with the reference user UE of the reference cell
0The same pilots are used, such as: UE3 in fig. 2 is within this range and may multiplex pilots with UE 0. UE1 is not in range of C and cannot multiplex pilots with UE 0; UE2 is in
Cannot multiplex pilots with the UE 0; in FIG. 2, the small circle has C, r inside
0Indicating that the user in cell l, who can multiplex the pilot, is far from its serving base station BS
lThe farthest distance of (c). The inside of the large circle is
Representing reference user UE
0The pilots used are prohibited from being multiplexed within the area,
is centered at base station BS0 and r
2Being the inside of a circle of radius. r is
0Calculated according to equation (2):
wherein r is1=D(BSl,UE0) And D (·,) represents the distance between two points. r is2Calculated from equation (3):
BS0a reference base station representing a reference cell.
In summary, the present invention discloses a pilot frequency allocation technique method for combining modulation orders in a multi-cell system, which is applicable to a large-scale MIMO system and belongs to the technical field of communication. The invention firstly divides users into two types according to a division threshold with parameters such as modulation order, base station density and the like, wherein one type is a type I user which can completely multiplex pilot frequency; the other is class II users that cannot fully multiplex pilots. The pilot set is then divided into two non-overlapping subsets according to the number of class I users. And then allocating the pilot frequency in a pilot frequency subset for the class I user. And finally, according to a certain rule, finding out the II-type users which can multiplex the pilot frequency mutually. When designing a pilot frequency distribution scheme for realizing a multi-cell large-scale MIMO system, the technical scheme provided by the invention can effectively avoid the influence of pilot frequency pollution on the system capacity and reduce the error rate of the system.