CN103259578A - Method for selecting collaborative partner of collaborative multiple input multiple output (MIMO) system - Google Patents

Abstract

The invention discloses a method for selecting a collaborative partner of a collaborative MIMO system, and relates to the field of wireless communication. The method resolves the problem that an existing partner selecting method of a WLF algorithm is large in energy grain loss. The method includes the steps: if the number of cell users in a collaborative MIMO system is larger than 2, a base station selects a cell user i who has the poorest channel condition; whether a signal to noise ratio gamma i of a link between the cell user i who has the poorest channel condition and the base station is larger than a first threshold gamma th1is judged, if the signal to noise ratio gamma i of the link between the cell user i who has the poorest channel condition and the base station is not larger than the first threshold gamma th1, a user j who has the best channel condition with the user i is selected in the collaborative MIMO system; whether a signal to noise ratio of a link between the user i and the user j is larger than a preset second threshold gamma th 2 is judged, if the signal to noise ratio of the link between the user i and the user j is larger than a preset second threshold gamma th 2, the user j is used as a collaborative partner, the user i and the user j are deleted from a user list, and collaborative partner selection in the MIMO system is completed at one time. The method is suitable for collaborative partner selection in the collaborative MIMO system.

Description

The cooperation partner selection method of cooperation MIMO system

Technical field

The present invention relates to a kind of wireless communication field.

Background technology

Studies show that the performance of cooperative diversity system depends primarily on the channel quality of link between cooperation way and each node.Therefore, for the wireless network that has a plurality of candidate's cooperative partner, how to choose performance that suitable cooperative partner makes system and reach optimum and have very important significance, this also becomes one of research focus of present cooperative diversity technique.

According to select according to different, the cooperation partner selection method mainly comprises three classes: the system of selection of position-based information, based on the system of selection of average channel condition information and based on the system of selection of instantaneous channel condition information.The system of selection of position-based information need be known each internodal distance; System of selection based on average channel condition information relatively is applicable to continuous transfer of data, and the perhaps situation of fast fading channel, and system can't obtain the situation of instantaneous channel condition information accurately; Relatively be applicable to the transfer of data of burst, the perhaps situation of slow fading channel based on the system of selection of instantaneous channel condition information.

According to the power to make decision difference of selecting, the cooperation partner selection method can be divided into centralized system of selection and distributed system of selection two big classes.Centralized system of selection is applicable to the network at center, as cellular network, Centroid (base station) has between all user position information or all users and the signal to noise ratio information between user and base station, for needing users in collaboration, each selects one or more cooperative partner, according to certain optimization aim, make the overall performance of network reach optimum.More representational centralized partner selection algorithm comprises weight limit matching algorithm, greedy matching algorithm, the preferential matching algorithm of the poorest link etc.Distributed system of selection is applicable to acentric network, and as wireless self-networking, wireless sensor network, each node carries out the selection of cooperative partner independently.

In the classic algorithm of partner selection, weight limit matching algorithm and Greedy matching algorithm all need to carry out within the specific limits exhaustive, computation complexity is all very high, and in real system, need real-time the grouping of finishing the user, the poorest link priority algorithm (WLF) has therefore been proposed, this algorithm energy gain loss is less and can reduce computation complexity, is the partner selection scheme of this primary study.

Traditional WLF algorithm will transmit apart from the principal element of loss as the signal to noise ratio size, namely the near subscriber channel condition in apparent distance base station is better than the subscriber channel condition far away apart from the base station, this just needs the base station to grasp all user position information in the residential quarter.

Summary of the invention

The present invention solves the big problem of energy gain loss of the Partnership Selection Method of the existing WLF of employing algorithm, thereby a kind of cooperation partner selection method of cooperation MIMO system is provided.

The cooperation partner selection method of cooperation MIMO system, it is realized by following steps:

Step 1, whether judge in the cooperation MIMO system residential quarter number of users greater than 2, if judged result is for being that then execution in step two; If judged result is that then execution in step five;

Each community user is judged in the cooperation MIMO system one by one to the channel status of base station in step 2, base station, and therefrom chooses the poorest community user i of channel status; And judge the signal to noise ratio γ of link between community user i that this channel status is the poorest and base station _iWhether greater than default thresholding γ _Th1If judged result is for being that then execution in step five; If judged result is that then execution in step three;

Step 3, in cooperation MIMO system, select and user i between the best user j of channel status, and whether the signal to noise ratio of judging link between user i and the user j greater than default No. two thresholding γ _Th2If judged result is for being that then execution in step four; If judged result is deleted user i, and is returned execution in step one for not from user list;

Step 4, with the cooperative partner of user j as user i, user i and user j are deleted from user list, and return execution in step one;

Step 5, finish the cooperation partner selection of cooperation MIMO system.

In the cooperation partner selection process of cooperation MIMO system, the signal to noise ratio of every section link is according to formula:

$\mathrm{\γ}=\frac{E{\left|\right|H\left|\right|}_F^2}{2N_0}$

Obtain;

Wherein: E is that terminal is launched the required energy of 1 bit information, N ₀Be the one-sided power spectrum density of white Gaussian noise, H is the channel matrix of respective links, Be Frobenius square of norm, that is:

${\left|\right|H\left|\right|}_F^2=\underset{i=1}{\overset{M_t}{\mathrm{\Σ}}}\underset{j=1}{\overset{M_r}{\mathrm{\Σ}}}{\left|H_{j^{\text{'}},i^{\text{'}}}\right|}^2;$

H wherein _{J ', i '}The multiple fading coefficients of the channel of expression from transmitting antenna i ' to reception antenna j ', M _tAnd M _rBe positive integer.

The energy gain loss of Partnership Selection Method of the present invention is little, reduced computation complexity significantly, thereby selection speed is promoted significantly.

Description of drawings

Fig. 1 is signal processing flow schematic diagram of the present invention;

Fig. 2 is the preferential coupling of channel and stochastic selection algorithm performance simulation schematic diagram between the user in the embodiment one;

Fig. 3 is the preferential matching algorithm performance simulation of channel schematic diagram between the user under the different situations in the embodiment one.

Embodiment

The cooperation partner selection method of embodiment one, cooperation MIMO system, it is realized by following steps:

Step 1, whether judge in the cooperation MIMO system residential quarter number of users greater than 2, if judged result is for being that then execution in step two; If judged result is that then execution in step five;

Each community user is judged in the cooperation MIMO system one by one to the channel status of base station in step 2, base station, and therefrom chooses the poorest community user i of channel status; And judge the signal to noise ratio γ of link between community user i that this channel status is the poorest and base station _iWhether greater than default thresholding γ _Th1If judged result is for being that then execution in step five; If judged result is that then execution in step three;

Step 3, in cooperation MIMO system, select and user i between the best user j of channel status, and whether the signal to noise ratio of judging link between user i and the user j greater than default No. two thresholding γ _Th2If judged result is for being that then execution in step four; If judged result is deleted user i, and is returned execution in step one for not from user list;

Step 4, with the cooperative partner of user j as user i, user i and user j are deleted from user list, and return execution in step one;

Step 5, finish the cooperation partner selection of cooperation MIMO system.

In the cooperation partner selection process of cooperation MIMO system, the signal to noise ratio of every section link is according to formula:

$\mathrm{\γ}=\frac{E{\left|\right|H\left|\right|}_F^2}{2N_0}$

Obtain;

Wherein: E is that terminal is launched the required energy of 1 bit information, N ₀Be the one-sided power spectrum density of white Gaussian noise, H is the channel matrix of respective links,

Be Frobenius square of norm, that is:

${\left|\right|H\left|\right|}_F^2=\underset{i=1}{\overset{M_t}{\mathrm{\Σ}}}\underset{j=1}{\overset{M_r}{\mathrm{\Σ}}}{\left|H_{j^{\text{'}},i^{\text{'}}}\right|}^2;$

H wherein _{J ', i '}The multiple fading coefficients of the channel of expression from transmitting antenna i ' to reception antenna j ', M _tAnd M _rBe positive integer.

The present invention proposes to adopt channel matrix H as judging that channel condition carries out choosing of collaboration user, and user position information can be grasped in the base station like this, and channel matrix is included among the CSI as a kind of basic information.Because in the multiaerial system, adopt Space-Time Block Coding again, need know CSIs, therefore can not increase with CSIs and carry out channel estimating because of cooperation and cause overhead.

If the service range loss represents the difference that channel status selects to abandon a single aerial system and multiaerial system, and because desire of the present invention is applied to multiaerial system with this Partnership Selection Method, and analyze its performance in multiaerial system, so the selective channel matrix is used apart from loss and is represented that channel status more can embody the difference of algorithm performance in single antenna and multiaerial system.

On concrete grammar, the present invention has studied a kind of WLF algorithm that has thresholding that is suggested.Thresholding 1 (λ wherein _Th1) be enough to get well the situation appearance that causes cooperating benefit not quite or do not have cooperative gain for avoiding to channel condition between the base station because of source user, the setting of this thresholding also makes selection algorithm be able to a certain degree simplification, namely do not need all users are selected the partner, only the selection for needing has reduced the selection number of times.Thresholding 2 (λ _Th2) setting be effect in order to guarantee to cooperate, if cooperative partner is very bad to the Link State between the base station, and cooperation itself has consumption, then can cause transmission performance to descend, and loses more than gain.Based on above analysis, this novel WLF selects flow process as shown in Figure 1.

A part process is to avoid to channel conditions is enough good, causes the wasting of resources when not needing users in collaboration j to select collaboration user.The matching algorithm of WLF user grouping sources of law in graph theory be not so consider other users are arranged as the situation of the cooperative partner of j again when j is as the cooperative partner of i at this.

The beneficial effect that the present invention obtains: table 1 is between the user under the preferential matching algorithm of channel in the multiaerial system, does not always find the situation of cooperative partner number of users under the different situation of number of users the residential quarter in.These data are the mean value of 1000 simulation calculation.By data in the table as can be seen, this algorithm produces that number of users has nothing to do in the number of users that do not find the partner and this residential quarter.Be when total number of users increases in the residential quarter, not find partner's number of users to remain unchanged substantially, therefore, in the higher residential quarter of user density (being that always number of users is bigger the residential quarter in), the probability that the user can not find the partner is lower.Thereby we consider the performance of algorithm under high density and two kinds of situations of low-density by number in the change residential quarter.The average energy gain that can draw system increases with the increase of number of users in the residential quarter, and table 1 has been explained the reason of this variation tendency.This has illustrated that channel priority algorithm performance in the high density residential quarter is brought into play better between the user, is more suitable for as this high density custom systems such as market, school, apartments.

The total number of users of table 1 not can not find user's number of collaboration user simultaneously

Fig. 2 is the preferential matching algorithm of channel and the performance comparison of stochastic selection algorithm in multiaerial system between the user.Can clearly find out the superiority of the preferential matching performance of channel between the user.How do not consider benefit when selecting because of selection at random, so selection algorithm is simple.And need more energy transmission data when selecting the incorrect meeting of partner to cause same bit error rate.In multiaerial system, because the performance of system itself just is better than with substandard a single aerial system, the limited energy gain of selecting cooperation to produce does not offset the loss of cooperation self at random, causes cooperation not only not bring gain for multiaerial system, and systematic function is descended.From user perspective, in the whole residential quarter, certain customers' cooperative gain is being for just, but the cooperation energy gain of more users is for negative, and therefore the gain of the average energy of whole residential quarter is for negative.This has also illustrated the importance of selection algorithm in cooperation MIMO system.Indiscreet cooperation is run counter to desire on the contrary.The average energy gain of selecting at random among the figure is changing with total number of users in the residential quarter as the preferential matching algorithm of channel between the user not.This is because at random when selecting, will around user situation do not do consideration, just select randomly, so how much can not the impacting it of user density in the residential quarter.

Fig. 3 be multiaerial system thresholding is arranged and do not have average energy gain under the cooperation of the preferential matching algorithm of channel between the user of thresholding with the residential quarter in the variation of number of users, in the total and residential quarter under the different situation of number of users between the user channel preferentially mate the comparison that gains of average energy in many antennas and a single aerial system.Can see significantly among the figure that the selection algorithm that thresholding arranged for system provides higher average energy gain, makes the average energy gain that lifting about 2dB arranged than the selection algorithm that does not have thresholding.

Average energy gain in the multiaerial system is lower than the average energy gain of a single aerial system always.Be that the preferential matching process of channel is that multiaerial system has been brought the average energy gain between the user, but the average energy gain is less than a single aerial system.Cause this result's reason to be, because each user terminal of multiaerial system itself has a plurality of antennas (Jia Ding each terminal has 2 antennas) here, so even do not cooperate between the user terminal in the system, still have diversity gain.By formula (1-1), error rate of system one regularly, send 1bit institute energy requirement during the non-cooperation of terminal use in the multiaerial system and be less than the corresponding institute of a single aerial system energy requirement, and the gain that cooperation itself produces is limited, so be not big to a single aerial system of the average energy gain that brings of multiaerial system with the cooperation of the preferential match selection cooperative partner of channel between the user.

In the simulation process, under the identical simulation times, the multiaerial system performance is more stable, and a single aerial system performance inconsistency is bigger.When being improved an order of magnitude, the simulation times in a single aerial system make performance index stable just now.This point has illustrated that also the good stability of multi-antenna cooperative system is in the single antenna cooperative system.

Claims (2)

1. the cooperation partner selection method of cooperation MIMO system, it is characterized in that: it is realized by following steps:

Step 1, whether judge in the cooperation MIMO system residential quarter number of users greater than 2, if judged result is for being that then execution in step two; If judged result is that then execution in step five;

Each community user is judged in the cooperation MIMO system one by one to the channel status of base station in step 2, base station, and therefrom chooses the poorest community user i of channel status; And judge the signal to noise ratio γ of link between community user i that this channel status is the poorest and base station _iWhether greater than default thresholding γ _Th1If judged result is for being that then execution in step five; If judged result is that then execution in step three;

Step 3, in cooperation MIMO system, select and user i between the best user j of channel status, and whether the signal to noise ratio of judging link between user i and the user j greater than default No. two thresholding γ _Th2If judged result is for being that then execution in step four; If judged result is deleted user i, and is returned execution in step one for not from user list;

Step 4, with the cooperative partner of user j as user i, user i and user j are deleted from user list, and return execution in step one;

Step 5, finish the cooperation partner selection of cooperation MIMO system.

2. the cooperation partner selection method of cooperation MIMO system according to claim 1 is characterized in that in the cooperation partner selection process of cooperation MIMO system, and the signal to noise ratio of every section link is according to formula:

$\mathrm{\γ}=\frac{E{\left|\right|H\left|\right|}_F^2}{2N_0}$

Obtain;

Wherein: E is that terminal is launched the required energy of 1 bit information, N ₀Be the one-sided power spectrum density of white Gaussian noise, H is the channel matrix of respective links,

Be Frobenius square of norm, that is:

${\left|\right|H\left|\right|}_F^2=\underset{i=1}{\overset{M_t}{\mathrm{\Σ}}}\underset{j=1}{\overset{M_r}{\mathrm{\Σ}}}{\left|H_{j^{\text{'}},i^{\text{'}}}\right|}^2;$

H wherein _{J ', i '}The multiple fading coefficients of the channel of expression from transmitting antenna i ' to reception antenna j ', M _tAnd M _rBe positive integer.

Images