CN105763309B - Pilot pollution mitigates method in the extensive mimo system of chain network - Google Patents

Pilot pollution mitigates method in the extensive mimo system of chain network Download PDF

Info

Publication number
CN105763309B
CN105763309B CN201610151157.3A CN201610151157A CN105763309B CN 105763309 B CN105763309 B CN 105763309B CN 201610151157 A CN201610151157 A CN 201610151157A CN 105763309 B CN105763309 B CN 105763309B
Authority
CN
China
Prior art keywords
cell
pilot
user
users
pilot sequence
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201610151157.3A
Other languages
Chinese (zh)
Other versions
CN105763309A (en
Inventor
张翠芳
曾桂根
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Inner Mongolia Feiqi Technology Development Co.,Ltd.
Original Assignee
Nanjing Post and Telecommunication University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing Post and Telecommunication University filed Critical Nanjing Post and Telecommunication University
Priority to CN201610151157.3A priority Critical patent/CN105763309B/en
Publication of CN105763309A publication Critical patent/CN105763309A/en
Application granted granted Critical
Publication of CN105763309B publication Critical patent/CN105763309B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L5/00—Arrangements affording multiple use of the transmission path
    • H04L5/003—Arrangements for allocating sub-channels of the transmission path
    • H04L5/0048—Allocation of pilot signals, i.e. of signals known to the receiver
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/30—Monitoring; Testing of propagation channels
    • H04B17/309—Measuring or estimating channel quality parameters
    • H04B17/336—Signal-to-interference ratio [SIR] or carrier-to-interference ratio [CIR]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/022—Site diversity; Macro-diversity
    • H04B7/024—Co-operative use of antennas of several sites, e.g. in co-ordinated multipoint or co-operative multiple-input multiple-output [MIMO] systems
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H04L25/0204—Channel estimation of multiple channels
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L5/00—Arrangements affording multiple use of the transmission path
    • H04L5/14—Two-way operation using the same type of signal, i.e. duplex
    • H04L5/1469—Two-way operation using the same type of signal, i.e. duplex using time-sharing

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Power Engineering (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention discloses pilot pollutions in the extensive mimo system of chain network to mitigate scheme, that is, is directed to the extensive MIMO multiple cell TDD system feature of chain network, proposes a kind of new pilot pollution mitigation scheme.In scheme by it is each it is small be distinguished into Zuo Qu and You Qu, and orthogonal pilot frequency sequence is divided into three groups.Three groups of pilot frequency sequence groups successively, cyclically distribute to each left and right regions user in chain network, to realize that target user near field user adjacent thereto arranges orthogonal pilot frequency sequence, and target user and zone user multiplexed pilot sequence farther out, to alleviate pilot pollution.Simulation result shows this programme due to saving the uplink channel estimation time, and handling capacity is better than pilot power control method, and this paper scheme, relative to pilot tone enumeration method, calculation amount is small, has certain practical value.

Description

Pilot pollution mitigates method in the extensive mimo system of chain network
Technical field
Present invention relates particularly to pilot pollutions in the extensive mimo system of chain network to mitigate scheme, belongs to wireless communication uplink Channel estimation technique field.
Background technique
Extensive MIMO (" Large-Scale MIMO " or " Massive MIMO ") technology, by directly increasing base station Antenna number can increase power system capacity, reduce user emission power.And with the increase of antenna for base station number, in noise and cell Interference can be ignored.
1. the pilot pollution problem is analyzed
In the uplink channel estimation of extensive MIMO multiple cell TDD (Time-division Duplex) system, Mei Ge little Mutually orthogonal pilot tone is used in area, minizone uses identical pilot frequency sequence group.It is τ that all community users send length simultaneously The pilot tone vector of (τ >=N).If k-th of user of all cells sends pilot tone simultaneously(meet normalization Condition), mean power (during transmitting) is respectively p at base station and userfAnd ρr, the reception of i-th of base station at this time Signal are as follows:
Wherein ginlFor channel vector of the nth user to i-th of cell base station of first of cell, ziFor the mutual of M × τ Independent additive noise.K-th of subscriber channel estimates of parameters of i-th of base station be
WhereinAs can be seen that the channel parameter estimated from the Section 2 of formula (2) There are the interference that other cells estimate Target cell, i.e. pilot pollution (Pilot Contamination).It is generated main The reason is that in uplink channel estimation, when the user of different community uses same set of pilot frequency sequence or non-orthogonal pilot tone sequence Column, causing base station end to carry out the result of channel estimation is not channel between local user and base station, but is sent out by other cells Estimation after sending the user of non-orthogonal pilot sequence to pollute.
Assuming that the upstream data power that each user sends is identical.According to the channel parameter that formula (2) obtains, base station side is to upper Row data carry out MRC (Maximum-ratio combining) and merge.As antenna for base station M → ∞, i-th of base station is to k-th The upstream data s of userkiThe signal interference ratio SIR (Signal to Interference Ratio) of estimated value is
As can be seen thatMainly declined slowly by other cells using the channel of same pilot or non-orthogonal pilot signaling user Fall factor betaiklIt influences.When the interference user of distribution same pilot is remoter to Target cell distance, slow fading coefficient is smaller,It is worth bigger.
It is similarly assumed that the downlink data power that each base station is sent is identical.According to the channel parameter that formula (2) obtains, base station Side carries out linear predictive coding to downlink data.As antenna for base station M → ∞, k-th of user's downlink reception data of i-th of base station The signal interference ratio of estimated value is
As can be seen that as antenna for base station M → ∞,WithIt is different, it and slow fading coefficient βlkiIt is related, It is i.e. closely related at a distance from target user to interfered cell base station.The slow fading coefficient of target user to interfered cell base station is got over It is small,It is worth bigger.
As can be seen from the above analysis, pilot pollution seriously limits the value of signal interference ratio, to also limit extensive The channel capacity that MIMO communication system can achieve.
2, correlative study scheme
For the pilot pollution problem existing for extensive MIMO uplink channel estimation, domestic and international researcher has carried out correlation Research.(the pilot pollution in the extensive MIMO beehive network system of Wang Hairong, Wang Yuhui, Huang Yongming, Yang Lvxi such as Wang Hairong Mitigation method [J] communicates journal, 2014,35 (1): 24-33.) all cells in cellular network are divided into two classes, it is same Each cell inside class uses identical pilot frequency sequence group, and the cell between inhomogeneity uses pilot power control method, mitigates Each district pilots pollution inside same class.The program has certain practicability, but is only applicable to cellular network.Wang Hairong etc. (pilot pollution in the extensive MIMO multiple cell TDD system of Wang Hairong, Wang Yuhui, Huang Yongming, Yang Lvxi mitigates method [J] Signal processing, 2013,29 (2): 171-180.) be staggered neighboring cell pilot sequence sending time in one-dimensional Multi-cell model, To avoid the pilot pollution of neighboring community.Program channel estimation time is longer, and signal interference ratio is limited to send pilot tone simultaneously The interference of cell.(NEUMANN David, the etal..Pilot coordination for large- such as NEUMANN David scale multi-cell TDD systems[C].201418th International ITG Workshop on Smart Antennas (WSA), Erlangen, Germany, 2014:1-6.) according to extensive mimo system data transmission signal interference ratio only With slow fading coefficient in relation to this conclusion, enumerates all pilot tones of all cells and distribute possible situation, calculate every kind of possible feelings The total rate of system under condition.It allows the total rate of system to reach maximum that group of pilot tone distribution, is exactly optimal pilot distribution.The program with The increase of cell number, number of users, algorithm calculation amount increase with it.
From the point of view of the above existing algorithm, any pilot pollution, which mitigates scheme, will all pay a price.And it leads It can not be completely eliminated in frequency pollution real system, pilot pollution can only be mitigated by effective ways.
Summary of the invention
To overcome the shortcomings of existing technologies, mitigate pilot pollution, the present invention proposes to lead in the extensive mimo system of chain network Frequency pollutes the key step of mitigation scheme are as follows:
1, initial scene setting:
1-1) in the extensive mimo system model of Fig. 1 chain network multiple cell, setting system uses tdd mode.
The cell that L non-cooperating 1-2) is arranged uses same frequency band.
1-3) base station is arranged in the heart in each cell, and there is M root antenna in base station.
1-4) centered on base station, each cell setting is divided into Zuo Qu and You Qu.Each cell is arranged N number of single antenna and uses Family is evenly distributed in cell, i.e., each cell Zuo Qu is identical with the number of users in right area, is
2, scheme realizes step:
2-1) Zuo Qu and You Qu that each cell is arranged have identical number of users.The left area user that each cell is arranged compiles Number it is Right area's Customs Assigned Number is
2-2) system carry out uplink channel estimation when, k-th of user use length forPilot frequency sequencePilot frequency sequence meets normalizing conditionIllustrate: in normalizing condition, subscript+table Show the conjugate transposition of matrix, ΨxIndicate the pilot frequency sequence of 1 × τ of x-th of user, ΨyIndicate the pilot tone of 1 × τ of y-th of user Sequence.Because pilot frequency sequence be it is orthogonal,I.e. as x=y, δxy=[1], as x ≠ y, δxy=[0].
The pilot frequency sequence that length is τ 2-3) is divided into three groups to get mutually orthogonal pilot frequency sequence group is arrived
2-4) as shown in Figure 1.Three groups of pilot frequency sequence groups successively, are cyclically allocated.It uses in the right area of i.e. i-th cell Family uses pilot frequency sequence groupThen the left area user of (i+1) cell uses pilot frequency sequence groupIt uses in the left area of i-th of cell The right area user multiplexed pilot sequence group at family and (i+1) cellAnd so on.I.e. target user region adjacent thereto is close Orthogonal pilot frequency sequence is arranged apart from user, and target user and zone user multiplexed pilot sequence farther out.
2-5) as shown in Figure 1.For the user of multiplexed pilot sequence group, the left area the of i-th of cell is arrangedThe right area of a user and (i+1) cell theA user uses identical pilot frequency sequence.Such as the The right area of the 1st, the left area user of i cell and (i+1) cell theA user uses identical pilot frequency sequence.With This analogizes.
Beneficial effect
Length is by this programme MIMO multiple cell TDD system feature extensive to chain network in uplink channel estimationPilot frequency sequence be divided into mutually orthogonal pilot frequency sequence group so that the short distance user that target user is adjacent thereto Orthogonal pilot frequency sequence is arranged, and target user is effectively reduced pilot pollution with compared with remote subscriber's multiplexed pilot sequence.Work as base When station antenna M → ∞, this paper new departureMainly by compared with remote and to be arranged same pilot sequence user small to target The channel slow fading coefficient in area influences,Mainly by target user to the cell base station for being arranged same pilot sequence user Channel slow fading coefficient influences, and effectively increases signal interference ratio.It is more that simulation result shows that this programme is more suitable for community user number The case where, and it is simple and easy, more have significant practical applications.
Detailed description of the invention
Fig. 1 is the extensive mimo system model of chain network multiple cell: each cell base station is located at center of housing estate, and there is M in base station Root antenna.Centered on base station, small Zuo Qu and You Qu are divided into for each.Each cell has N number of single-antenna subscriber, equably divides For cloth in cell, i.e., each cell Zuo Qu is identical with the number of users in right area, is
Fig. 2 is TDD transport protocol in coherence time: having T OFDM symbol within coherence time.In Fig. 2 in coherence time TDD transport protocol is divided into four parts: 1. τupThe transmission of a upstream data；2. the transmission of τ uplink pilot sequence；3. channel is estimated Meter, upstream data and processing of pre-coding matrix etc., amount to a symbol；④τdownThe transmission of a downlink data.Therefore every T=τ in a coherence timeup+1+τ+τdown。
Fig. 3 is cell number, slow fading coefficient is certain, and community user uplink throughput compares (L in the case that number of users is different =5, b1=0.5；B2=0.11；B3=0.0625；B4=0.04).
Fig. 4 is cell number, number of users is certain, and cell uplink throughput ratio is compared with (N=6, L under slow fading coefficient different situations =5, B1:b1=0.4；B2=0.11；B3=0.0625；B4=0.04, B2:b1=0.7；B2=0.21；B3=0.05；B4= 0.01)
Fig. 5 is that slow fading coefficient, number of users are certain, in the case that cell number is different cell uplink throughput ratio compared with (N=6, B1=0.5；B2=0.11；B3=0.0625；B4=0.04).
Fig. 6 is cell number, slow fading coefficient is certain, and community user downlink throughput capacity compares (L=under number of users different situations 5, b1=0.5；B2=0.11；B3=0.0625；B4=0.04).
Fig. 7 is cell number, number of users is certain, and cell downlink throughput ratio is compared with (N=6, L under slow fading coefficient different situations =5, B1:b1=0.4；B2=0.11；B3=0.0625；B4=0.04, B2:b1=0.7；B2=0.21；B3=0.05；B4= 0.01)。
Fig. 8 is slow fading coefficient, number of users is certain, and cell downlink throughput ratio is compared with (N=6, b1 under cell number different situations =0.5；B2=0.11；B3=0.0625；B4=0.04).
Specific embodiment
Technical solution is further detailed below in conjunction with attached drawing.
1, the channel estimation of new departure
Symbol definition used in this programme is as follows: (Α)T、(A)*、(A)+Respectively indicate the transposition of matrix Α, conjugation, Conjugate transposition.Oeprator E { } indicates mathematic expectaion.
System uses tdd mode in Fig. 1.Assuming that there is the cell of L non-cooperating to use same frequency band.Each cell base station is set There are M root antenna in the heart in the cell, base station.Centered on base station, small Zuo Qu and You Qu are divided into for each.Each cell has N number of Single-antenna subscriber is evenly distributed in cell, i.e., each cell Zuo Qu is identical with the number of users in right area, isAssuming that i-th Channel parameter between k-th of user of a cell and the m root antenna of (i-1) a cell is Channel parameter constitutes vector g(i-1)ki=(g1(i-1)ki,g2(i-1)ki,…,gM(i-1)ki)T.Wherein hm(i-1)ki~CN (0,1) (meets Cyclic Symmetry multiple Gauss) it is rapid fading coefficient, the h within coherence timem(i-1)kiKeep constant, i.e. flat fading.Rapid fading coefficient Constitute rapid fading vector h(i-1)ki=(h1(i-1)ki,h2(i-1)ki,…,hM(i-1)ki)T, and h(i-1)ki~CN (0, IM)。It mainly include shadow fading and path fading for slow fading coefficient.Wherein α(i-1)kiIt is obedience pair for constant The shadow fading coefficient of number normal distribution；d(i-1)kiFor i-th cell k-th of user to (i-1) a cell base station away from From；γ is path fading index, and value range is usually 2~6.Assuming that aerial array composed by the M root antenna of same base station It is sufficiently compact, then a certain user has identical slow fading coefficient to the M root antenna of same base station, and think slow fading coefficient It is the constant that anticipated that.As can be seen that the size of slow fading coefficient is closely related at a distance from user to base station, distance is got over Far, slow fading coefficient is smaller.
This programme is discussed by taking left area user as an example.Assuming that all community users send pilot tone simultaneously.Base station and user Locating mean power (during transmitting) is respectively pfAnd ρr.This programme considers in chain network that i-th of cell only receives adjacent thereto (i+1), (i+2), (i-1), (i-2) a cell user send signal, ignore other apart from farther away cell to target The interference of cell.
I-th of base station received signal at this time are as follows:
Wherein ψnlFor the pilot tone vector of the nth user of first of cell；ziFor the mutually independent additive noise of M × τ. If by taking the left area user of i-th of base station as an example, the left area of i-th of base stationA subscriber channel parameter Estimation Value is
Wherein
Similarly, the of i-th of base stationThe area Ge You subscriber channel estimates of parameters is
From formula (6) (7) although second and third in as can be seen that the channel parameter that estimates there are still other cells pair The interference of Target cell user estimation, but interfere both from compared with remote subscriber, it is effectively reduced pilot pollution.
2, the transmitting uplink data analysis of new departure
In transmitting uplink data, herein by taking the left area user of i-th of base station as an example, i.e.,Assuming that first small The nth user in area sends upstream data snl, then N number of user of first of cell sends upstream data Sl=[s1l,…sNl]T。Sl Meet E { Sl}=0,I-th of base station received signal at this time are as follows:
Wherein υi∈M×1It is additive noise, υi~CN (0,1).
S is estimated using MRC (Maximum-ratio combining) method in i-th of base stationkiValue:
In order to calculate as M → ∞, the value of formula (9), we use following lemma 1.
Lemma 1: if x, y ∈ CM×1Two Independent Vectors obey CN (0, cI), then
As antenna for base station M → ∞, since the channel vector of different user is statistical iteration, and hikl~CN (0, IM)。 According to lemma 1, can be obtained
As antenna for base station M → ∞, the channel parameter of each terminal to base station is statistical iteration, and each subscriber channel parameter With noise and independent.According to lemma 1 and formula (12) conclusion, can be obtained in formula (9)
According to the above analysis, available skiEstimated value be
Assuming that each user's transmission upstream data power is identical.I-th of base station is obtained to k-th of user according to formula (17) The upstream data s in (left area)kiThe signal interference ratio of estimated value is
As can be seen that as antenna for base station M → ∞, using the pilot arrangement scheme of this paper,Mainly by relatively remote And it is arranged the channel slow fading coefficient influence of same pilot sequence user to Target cell.
3, the downlink data transmission analysis of new departure
In downlink data transmission, herein by taking the left area user of i-th of base station as an example, i.e.,Downlink data passes When defeated, base station side precoding beam forms vector and is
Due to the reciprocity of TDD system channel, that is, channel parameter when thinking the transmission of uplink and downlink data is identical.Assuming that qnl It is the downlink data that first of cell is sent to nth user, then it is q that first of cell, which sends the downlink data of N number of user,l= [q1l,…qNl]T。qlMeet E { ql}=0,The left area user downlink data q of k-th of i-th of cellkiEstimation Value is
Wherein to user's additive noise z usedkiIt is independent identically distributed, and zki~CN (0,1).It enables and receives the flat of signal Equal power is
Five kinds of situations of lower surface analysis.
1. as l=i,
According to lemma 1 and formula (12) conclusion,
2. as l=i-2,
Because of the right area of k-th of the left area user of i-th of cell and (i-2) cell theA user uses phase Same pilot frequency sequence.According to lemma 1 and formula (12) conclusion,
3. as l=i+1,Analysis method with it is 2. identical, obtain
4. as l=i+2,
According to lemma 1 and formula (12) conclusion, obtain
5. as l=i-1, analysis method with it is 4. identical, obtain
Since the variance of noise is definite value, as antenna for base station M → ∞, the influence of noise is almost nil.By above Analysis, it can be seen that k-th of user (left area) downlink data q of i-th of cellkiEstimated value by formula (23) (25) (26) three Composition.
As antenna for base station M → ∞, using the pilot arrangement scheme of this paper, under k-th of user (left area) of i-th of cell Row data qkiThe signal interference ratio of estimated value is
As can be seen that as antenna for base station M → ∞,WithDifferent, it is mainly by target user to being pacified The channel slow fading coefficient for arranging the cell base station of same pilot sequence user influences.
In above-mentioned analysis, consideration is chain network middle cell.For chain network most both ends cell, due to interfered cell It reduces, pilot pollution can be smaller, and uplink and downlink data signal interference ratio improvement amount can be bigger.
4, scheme measurement index
Assuming that having T OFDM symbol within coherence time, and channel keeps constant in TDD transport protocol.According to fig. 2 TDD transport protocol, the handling capacity of i-th of cell, k-th of user uplink is in this paper scheme
The total throughout of i-th of cell uplink is in this paper scheme
In downlink data transmission, i-th of cell, k-th of user's downlink throughput capacity is in this paper scheme
In downlink data transmission, the total throughout of i-th of cell downlink is in this paper scheme
From formula (31)-(34) as can be seen that there is signal interference ratio of the handling capacity not only with each user in same coherence time It closes, it is also related with channel estimation time.Channel estimation time is longer in same coherence time, and data transmission period accounts for coherence time Ratio it is smaller.
5, simulation result
5.1 simulated conditions
Due to think in this programme i-th of cell of chain network only receive (i+1) adjacent thereto, (i+2), (i-1), (i-2) user of cell sends signal, ignores other interference apart from farther away cell to Target cell.Therefore numerical simulation mould Emphasis considers the case where cell number is 3 and 5 in type.
In numerical simulation, it is assumed that cell number is L, and community user number is N, mean power at base station and user (during transmitting) Respectively Pf=20dB, Pr=10dB.For chain network feature, to slow fading coefficient βlkiNormalized value is taken, i.e., if l=i, directly Meet gain (directgain) βiki=1；If l ≠ i, due to dikl> dikiAnd γ > 1, so cross-gain (crossgain) βikl< 1, and from i-th Target cell is remoter, βiklIt is smaller.It is defined in emulation: whenWhen, βik(i+1)=b1、βik(i+2) =b3、βik(i-1)=b2、βik(i-2)=b4；WhenWhen, βik(i+1)=b2、βik(i+2)=b4、βik(i-1)=b1`、 βik(i-2)=b3.Pilot frequency sequence uses Zadoff-Chu (ZC) sequence with constant amplitude, zero auto-correlation.In order to effectively divide Analyse the performance of various pilot tone transmission schemes, it is assumed that capable or downlink data is only caught in each coherence time.Each OFDM is accorded with The case where number each subcarrier is the same.The case where only considering an OFDM symbol single carrier is emulated herein.
In numerical simulation, emphasis has carried out the relative pilot power control method of this paper scheme, pilot tone enumeration method point Analysis is compared.The characteristics of according to chain network, the uplink channel estimation of k-th of user of i-th of base station of pilot power control method From (i-2), (i+2) two, cell uses same pilot sequence user farther out for interference.Its upstream data skiThe letter of estimated value Dry ratio is
Downlink data q of i-th of the base station of pilot power control method to k-th of userkiThe signal interference ratio of estimated value is
Pilot tone enumeration method is listed under the premise of the optimal pilot allocation plan selected, and k-th of user's of i-th of base station is upper The interference of row channel estimation uses same pilot sequence user from (i-2), (i+2), (i+1), (i-1) cell.Its uplink Data skiThe signal interference ratio of estimated value is
Downlink data q of i-th of the base station of pilot tone enumeration method to k-th of userkiThe signal interference ratio of estimated value is
5.2 transmitting uplink data analysiss of simulation result
1. cell number, slow fading coefficient are certain, in the case that number of users is different
From figure 3, it can be seen that under the same conditions, every community user number N is bigger, this paper scheme user uplink handling capacity is wanted More it is better than pilot power control method.This is because pilot power control method is staggered in chain network, neighboring cell pilot sequence is sent Time, pilot pollution from cell farther out, by formula (18) (35) it is found that itIt is more bigger than this paper scheme.But the side this paper Case uplink channel estimation within each coherence time needsA OFDM symbol time, and pilot power control method needs 2N The OFDM symbol time.This paper scheme savesA OFDM symbol time.Under the same conditions, N is bigger, this paper scheme channel Estimate that the time saved is more, user uplink handling capacity is more better than pilot power control method.For example, work as M=200, when N=6, The big 0.85bit/s/Hz of user uplink throughput ratio pilot power control method of this paper scheme.When every community user number N is smaller, The time that this paper scheme channel estimation is saved becomes smaller with the reduction of N, and because of the pilot pollution ratio of pilot power control method This paper scheme is smaller, therefore the uplink user handling capacity of pilot power control method is slightly better than this paper scheme.For example, work as M=200, N When=2, user uplink throughput ratio this paper scheme of pilot power control method is than bigger 0.2bit/s/Hz.
As can be seen that under the same conditions, when n is large, pilot tone enumeration method is better than this paper scheme from analogous diagram 3.Example Such as, work as M=200, when N=6, the big 0.7bit/s/Hz of user uplink throughput ratio this paper scheme of pilot tone enumeration method.This is because By formula (18) (37) although it is found that pilot tone enumeration method selects optimal pilot allocation plan, its pilot pollution ratio this paper scheme Greatly,It is smaller than this paper scheme, and channel estimation needs N number of OFDM symbol time in each coherence time.And this paper scheme It needsA OFDM symbol time.When N is bigger, it is longer that pilot tone enumeration method uplink channel estimation saves the time, therefore user uplink Handling capacity is bigger.But pilot tone enumeration method need to enumerate all pilot tones of all cells and distribute possible situation, calculate The total rate of system in the possible situation of kind.Such as τ=N=6, L=5, about 2.69 × 10 need to be calculated11In the possible situation of kind The total rate of system, calculation amount is quite big, and practical application is not strong.When N is smaller, pilot tone enumeration method advantage is not obvious.This is Because of every community user a few hours, time phase difference occupied by both scheme channel estimations is little, and the pilot tone of pilot tone enumeration method Damage ratio this paper scheme is big.
2. cell number, number of users are certain, the different situation of slow fading coefficient
N=6, L=5 are set when emulation.It is combined, wherein B1 scheme: being declined neighboring community using two groups of different slow fading coefficients Fall that very fast, slow fading coefficient is smaller；B2 scheme: neighboring community slowly declines, slow fading coefficient is larger.From fig. 4, it can be seen that this Literary scheme and pilot tone enumeration method, the case where being better than using the cell uplink handling capacity of B1 scheme when using B2 scheme.This be because For by formula (18) (37), it is found that B1 scheme neighboring community is used to decline in chain network, very fast, slow fading coefficient is smaller, pilot pollution It is also small,It is just big.Pilot power control method is close using the cell uplink handling capacity of B1, B2 scheme.This is because by formula (35) slow fading coefficient of cell is all from cell farther out, and farther out for the pilot pollution of pilot power control method in chain network known to It is smaller, therefore there is no significant changes using B1, B2 scheme pilot pollution.
When cell number, number of users are certain, when slow fading coefficient changes, the pilot tone of this paper scheme and pilot tone enumeration method is dirty Dye is decreased or increased therewith simultaneously,Also it increases with it or reduces simultaneously.The pilot power control method known to formula (35) Pilot pollution is influenced smaller from cell farther out by slow fading coefficient variation.According to 5.2 analysis 1., this paper scheme is better than leading Frequency power control method is inferior to pilot tone enumeration method.
3. slow fading coefficient, number of users are certain, the different situation of cell number
N=6 is set when emulation.From figure 5 it can be seen that as cell number L=3, the cell uplink handling capacity of every kind of scheme The case where when slightly better than L=5.This is because cell decline is very fast farther out in chain network, slow fading coefficient is all smaller.When cell number When reduction, by formula (18) (35) (37) it is found that the pilot pollution of uplink channel estimation also slightly reduces,Also increase therewith Greatly.Therefore with the reduction of cell number, the cell uplink handling capacity of every kind of scheme is slightly promoted.When cell number is decreased or increased When, the pilot pollution of various schemes is also decreased or increased therewith.According to 5.2 analysis 1., this paper scheme is better than pilot power control Preparation method is inferior to pilot tone enumeration method.
5.3 downlink data transmission analysiss of simulation result
1. cell number, slow fading coefficient are certain, the different situation of number of users
From fig. 6, it can be seen that under the same conditions, N is bigger, this paper scheme user downlink throughput capacity will more be better than pilot tone function Rate control methods.This is because by formula (30) (36) although it is found that the distracter of this paper scheme downlink data estimated value is than pilot tone function Rate control methods are bigger,It is more smaller than pilot power control method, but channel estimation time compares pilot tone within each coherence time Power control method is fewA OFDM symbol time.When N is bigger, the time that this paper scheme channel estimation is saved is more, user Handling capacity is more better than pilot power control method.For example, work as M=200, when N=6, user's downlink throughput capacity ratio of this paper scheme The big 1.2bit/s/Hz of pilot power control method.When every community user number N is smaller, the time of this paper scheme channel estimation saving Become smaller with the reduction of N, therefore user's downlink throughput capacity of pilot power control method is slightly better than this paper scheme.For example, working as M= When 200, N=2, the bigger 0.2bit/s/Hz of user throughput ratio this paper scheme of pilot power control method.
As can be seen that N is smaller from analogous diagram 6, this paper scheme user downlink throughput capacity will more be better than pilot tone enumeration method.Example Such as, work as M=200, when N=6, user's downlink throughput capacity of this paper scheme is 0.8bit/s/Hz bigger than pilot tone enumeration method.Work as M= When 200, N=2, user's downlink throughput capacity of this paper scheme is 2.45bit/s/Hz bigger than pilot tone enumeration method.This is because by formula (4) Know, as antenna for base station M → ∞,With it is closely related at a distance from target user to interfered cell base station.Although pilot tone arranges Act method channel estimation within each coherence time is fewA OFDM symbol time, but by formula (38) it is found that its lower line number The distracter ratio this paper scheme being worth according to estimates is much bigger,It is smaller than this paper scheme.And pilot tone enumeration method when τ=N=6, When L=5, about 2.69 × 10 need to be calculated11The total rate of system in the possible situation of kind, calculation amount are quite big.
2. cell number, number of users are certain, the different situation of slow fading coefficient
N=6, L=5 are set when emulation.Using two groups of different slow fading coefficient combinations.B1 scheme: neighboring community decline compared with Fastly, slow fading coefficient is smaller；B2 scheme: neighboring community slowly declines, slow fading coefficient is larger.From figure 7 it can be seen that the side this paper Case and pilot tone enumeration method, the case where being better than using the cell downlink handling capacity of B1 scheme when using B2 scheme.This is because by Very fast, slow fading coefficient is smaller it is found that B1 scheme neighboring community is used to decline in chain network for formula (30) (38), downlink data estimation When interference value it is also small,Also increase.Pilot power control method is close using the cell downlink handling capacity of B1, B2 scheme.This is Because interference value comes from cell farther out when the downlink data estimation of pilot power control method in chain network, and the slow of cell declines farther out It is all smaller to fall coefficient, by formula (36) it is found that itsVariation less, therefore is not become significantly using B1, B2 scheme pilot pollution Change.
When cell number, number of users are certain, when slow fading coefficient changes, this paper scheme and pilot tone enumeration method downlink data Interference value is decreased or increased therewith simultaneously when estimation.Interference value is from small farther out when the downlink data estimation of pilot power control method Area is influenced smaller by slow fading coefficient variation.According to 5.3 analysis 1., this paper scheme is arranged better than pilot power control method, pilot tone Act method.
3. slow fading coefficient, number of users are certain, the different situation of cell number
N=6 is set when emulation.As can be seen from Figure 8, as cell number L=3, the cell downlink handling capacity of every kind of scheme The case where when slightly better than L=5.This is because according to formula (30) (36) (38) it is found that when in chain network cell number reduce when, target The distracter of the downlink data estimated value of user reduces,It increases with it, the cell throughout of every kind of scheme is slightly promoted.
When cell number is decreased or increased, the pilot pollution of various schemes is also decreased or increased therewith.According to 5.3 point 1. Analysis, this paper scheme are better than pilot power control method, pilot tone enumeration method.
By uplink, downlink data transmission simulation analysis it is found that this paper new departure is handled up as antenna for base station number increases Amount also increases with it, and gradually levels off to definite value, demonstrates formula (18) (30).This paper scheme relative to pilot power control method, The uplink channel estimation time is saved, and is more suitable for the situation more than number of users.This paper scheme is relative to pilot tone enumeration method, letter It is single easy, have more practicability.

Claims (1)

1.链状网大规模MIMO系统中导频污染减轻方法，其特征在于，包括以下步骤：1. A method for mitigating pilot pollution in a massive MIMO system in a chain network, characterized in that it comprises the following steps: 1)初始场景设置：1) Initial scene settings: 1-1)在链状网多小区大规模MIMO系统模型中，设置系统采用TDD模式；1-1) In the chain network multi-cell massive MIMO system model, set the system to adopt the TDD mode; 1-2)设置L个非协作的小区使用相同频带；1-2) Set L non-cooperative cells to use the same frequency band; 1-3)在每个小区中心设置基站,基站有M根天线；1-3) A base station is set in the center of each cell, and the base station has M antennas; 1-4)以基站为中心，每个小区设置分为左区和右区，每个小区设置N个单天线用户，均匀地分布在小区内，即每个小区左区和右区的用户数相同，均为 1-4) With the base station as the center, each cell is divided into a left area and a right area, and each cell is set with N single-antenna users, which are evenly distributed in the cell, that is, the number of users in the left and right areas of each cell same, both 2)方案实现步骤：2) The implementation steps of the scheme: 2-1)设置每个小区的左区和右区有相同的用户数，即设置每个小区的左区用户编号为 右区用户编号为 2-1) Set the left and right areas of each cell to have the same number of users, that is, set the number of users in the left area of each cell to be The user number in the right area is 2-2)在系统进行上行信道估计时，第k个用户采用长度为τ的导频序列导频序列满足归一化条件归一化条件中，上标+表示矩阵的共轭转置，Ψx表示第x个用户的1×τ的导频序列，Ψy表示第y个用户的1×τ的导频序列,因为导频序列是正交的，所以即当x＝y时，δxy＝[1]，当x≠y时，δxy＝[0]；2-2) When the system performs uplink channel estimation, the kth user uses a pilot sequence with a length of τ The pilot sequence satisfies the normalization condition In the normalization condition, the superscript + represents the conjugate transpose of the matrix, Ψ x represents the 1×τ pilot sequence of the xth user, and Ψ y represents the 1×τ pilot sequence of the yth user, because The pilot sequences are orthogonal, so That is, when x=y, δ xy =[1], when x≠y, δ xy =[0]; 2-3)将长度为τ的导频序列分成三组，即得到相互正交的导频序列组 2-3) Divide the pilot sequence of length τ into three groups, that is, obtain mutually orthogonal pilot sequence groups 2-4)将三组导频序列组依次、循环地进行分配，即第i个小区的右区用户采用导频序列组则第(i+1)小区的左区用户采用导频序列组第i个小区的左区用户与第(i+1)小区的右区用户复用导频序列组以此类推，即目标用户与其相邻区域的近距离用户安排正交导频序列，而目标用户与较远区域用户复用导频序列；2-4) Allocate the three groups of pilot sequence groups in turn and cyclically, that is, the users in the right area of the i-th cell use the pilot sequence groups Then the users in the left area of the (i+1)th cell use the pilot sequence group The user in the left area of the i-th cell and the user in the right area of the (i+1)-th cell reuse the pilot sequence group And so on, that is, the target user arranges orthogonal pilot sequences with its nearby users in the adjacent area, and the target user multiplexes the pilot sequence with users in the farther area; 2-5)对于复用导频序列组的用户，令安排第i个小区的左区第k个用户与第(i+1)小区的右区第个用户使用相同的导频序列。2-5) For users of multiplexed pilot sequence groups, let Arrange the kth user in the left area of the i-th cell and the right area of the (i+1)-th cell users use the same pilot sequence.
CN201610151157.3A 2016-03-16 2016-03-16 Pilot pollution mitigates method in the extensive mimo system of chain network Active CN105763309B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610151157.3A CN105763309B (en) 2016-03-16 2016-03-16 Pilot pollution mitigates method in the extensive mimo system of chain network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610151157.3A CN105763309B (en) 2016-03-16 2016-03-16 Pilot pollution mitigates method in the extensive mimo system of chain network

Publications (2)

Publication Number Publication Date
CN105763309A CN105763309A (en) 2016-07-13
CN105763309B true CN105763309B (en) 2019-05-31

Family

ID=56333341

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610151157.3A Active CN105763309B (en) 2016-03-16 2016-03-16 Pilot pollution mitigates method in the extensive mimo system of chain network

Country Status (1)

Country Link
CN (1) CN105763309B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110535794B (en) * 2018-12-29 2022-11-11 中兴通讯股份有限公司 Pilot transmission method, channel estimation method, device and storage medium

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104735792A (en) * 2013-12-23 2015-06-24 华为技术有限公司 Method and device for distributing pilot frequency resources
CN105049098A (en) * 2015-06-15 2015-11-11 西安交通大学 Partial pilot frequency multiplexing method for overcoming pilot pollution in large-scale MIMO system

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9357507B2 (en) * 2012-09-13 2016-05-31 Qualcomm Incorporated Centralized management for pilot pollution mitigation in small cell networks

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104735792A (en) * 2013-12-23 2015-06-24 华为技术有限公司 Method and device for distributing pilot frequency resources
CN105049098A (en) * 2015-06-15 2015-11-11 西安交通大学 Partial pilot frequency multiplexing method for overcoming pilot pollution in large-scale MIMO system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
How Much Training is Needed for Interference Coordination in Cellular Networks;Hans H. Brunner等;《International Itg Workshop on Smart Antennas. IEEE》;20110405;全文

Also Published As

Publication number Publication date
CN105763309A (en) 2016-07-13

Similar Documents

Publication Publication Date Title
Shen et al. Downlink user capacity of massive MIMO under pilot contamination
Marzetta Noncooperative cellular wireless with unlimited numbers of base station antennas
CN104539411B (en) A kind of user choosing method of non-orthogonal multiple access system downlink
Alkhaled et al. A new allocation algorithm for pilot contamination mitigation in TDD massive MIMO systems
CN105681009B (en) Pilot tone optimization distribution joint method for precoding for multi-user&#39;s multiple-input and multiple-output
Li et al. Performance analysis of millimeter wave massive MIMO systems in centralized and distributed schemes
Elijah et al. Mitigating pilot contamination in massive MIMO system—5G: An overview
Xia et al. Large system analysis of resource allocation in heterogeneous networks with wireless backhaul
Kusaladharma et al. Rate and energy efficiency improvements of massive MIMO-based stochastic cellular networks with NOMA
Hu et al. Group successive interference cancellation assisted semi-blind channel estimation in multi-cell massive MIMO-NOMA systems
CN105763309B (en) Pilot pollution mitigates method in the extensive mimo system of chain network
CN115065389A (en) Performance optimization method for cell-free large-scale MIMO-NOMA system
Yang et al. Optimal beamforming for MIMO DFRC systems with transmit covariance constraints
Li et al. User grouping and pilot allocation for spatially correlated massive MIMO systems
CN104753573B (en) A kind of method for realizing space division multiplexing, system, base station and terminal
Panahi et al. Interference alignment and power allocation for multi-user MIMO interference channels
Sun et al. Successive interference cancellation based channel estimation for massive MIMO systems
Hu Beam grouping based user scheduling in multi-cell millimeter-wave MIMO systems
Kusaladharma et al. Rate analysis for NOMA in massive MIMO based stochastic cellular networks with pilot contamination
Tabikh et al. Beamforming design with combined channel estimate and covariance CSIT via random matrix theory
Luo et al. Pilot decontamination via PDP alignment
Zhou et al. Uplink asynchronous fractional pilots scheduling in massive MIMO system
Faisal et al. Efficient beamforming and spectral efficiency maximization in a joint transmission system using an adaptive particle swarm optimization algorithm
CN108599828B (en) Two-layer pre-coding method under multi-cell 3D MIMO scene
Zamarashkin et al. Coordinated zero-forcing precoder with leakage minimization

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
EE01 Entry into force of recordation of patent licensing contract

Application publication date: 20160713

Assignee: Nanjing Nanyou Communication Network Industry Research Institute Co.,Ltd.

Assignor: NANJING University OF POSTS AND TELECOMMUNICATIONS

Contract record no.: X2021980014811

Denomination of invention: Pilot pollution mitigation method in chain network large-scale MIMO system

Granted publication date: 20190531

License type: Common License

Record date: 20211213

EE01 Entry into force of recordation of patent licensing contract
TR01 Transfer of patent right

Effective date of registration: 20250417

Address after: 010040 Building 3, Dahe Baihui Star, West Side, Archives Bureau, Intersection of Zhaowuda Road and South Second Ring Road, Saihan District, Hohhot City, Inner Mongolia Autonomous Region, China

Patentee after: Inner Mongolia Feiqi Technology Development Co.,Ltd.

Country or region after: China

Address before: 210023 9 Wen Yuan Road, Ya Dong new town, Nanjing, Jiangsu.

Patentee before: NANJING University OF POSTS AND TELECOMMUNICATIONS

Country or region before: China

TR01 Transfer of patent right