WO2007102546A1 - Wireless communication device and scheduling method - Google Patents

Abstract

A group scheduling method is provided for dynamically grouping all users in a multiple-antenna-multiple-user wireless communication system based on average transmission capability which each user has shown. Since, at a certain scheduled time, the scheduling is time-sequentially carried out for each group. At the same time, two average transmission-capability calculating methods and two user-grouping methods are provided. According to the grouping methods, the number of users subjected to scheduling at the same time is reduced and a system feedback overhead is also reduced. Also, according to the two user-grouping methods, a system throughput quantity and user fairness can be effectively improved.

Description

 Specification

 Wireless communication apparatus and scheduling method

 Technical field

 [0001] The present invention relates to a radio communication apparatus and a scheduling method.

 Background art

 [0002] With the fusion of wireless networks and the Internet, demands for the type and quality of wireless communication services and information transmission rates are increasing. How to improve the information transmission rate is one of the major problems in wireless communications. MIMO communication technology is indispensable to solve this problem with limited frequency resources. MIMO communication is a technology that transmits signals using multiple antennas on the transmitting side and receives spatial signals using multiple antennas on the receiving side. MIMO technology can significantly improve channel capacity compared to conventional single-antenna transmission. Therefore, MIMO communication can increase the transmission rate of information.

 [0003] In recent years, with the development of research on MIMO communication, the research theme is shifting from point-to-point single-user MIMO communication to point-to-multipoint multi-user MIMO communication. Furthermore, multi-user diversity technology is one of the core technologies in multi-user MIMO communication. By using multi-user diversity technology, channel capacity in the downlink of multi-user systems can be improved.

[0004] Multi-user diversity technology is one of multi-user scheduling technologies adopted on the transmission side, and data transmission between a base station and a plurality of user terminals is performed in units of a plurality of resources. Sometimes. Each communication unit has a different resource unit. For example, a resource unit is a time in the time domain, a subcarrier in the frequency domain, a transmission antenna in the spatial domain, or the like. Further, in an actual multi-user communication system, the location and environment of each user terminal are different, so that channel characteristics in the same resource unit differ for each user terminal due to fading of radio signals. Multi-user diversity technology is based on this channel characteristic, By performing multi-user scheduling on the transmitting side, that is, comparing channel characteristics for each user in each resource unit, and assigning each resource unit to the user terminal with the best channel characteristics, the channel in the downlink Capacity can be increased.

 FIG. 1 is a diagram showing a configuration of a multiuser MIMO communication system using a conventional multiuser scheduling technique.

 [0006] As shown in FIG. 1, the multiuser MIMO communication system includes a base station 101 as a transmission side and K user terminals (eg, 111-1, 111-K) as a reception side. Base station 101 includes data storage section 102, scheduling section 103, MIMO transmission section 104, and n

 T

 The antenna 105 is provided. Each user terminal k (k = l, 2, ..., K) has n (K)

 R

 Tena 112—1,..., 112—K, MIMO detection 咅 — 1,..., 113—K, and channel estimation unit 114—1,..., 114—K .

 [0007] The data storage unit 102 of the base station 101 stores data to be transmitted to each user terminal k (k = 1, 2,..., K). Scheduling section 103 performs scheduling based on the channel characteristics of each user terminal to which user terminal k (k = l, 2,..., K) force is also fed back, and data storage section based on the scheduling result The corresponding data is extracted from 102 and output to MIMO transmitter 104.

 [0008] When scheduling is performed for each user terminal, the scheduling unit 103 uses current channel characteristic information of each user terminal. The channel characteristic information of each user terminal is represented by matrices H (t), H (t), ..., H (t). H (t) (k = l, 2, ...

 1 2 k k

 This is channel characteristic information between the base station 101 and the user terminal k. The rows and columns indicate n (k) and n, respectively, ie the number of receive antennas and the number of transmit antennas. This channel

R T

 The characteristic information H (t) is the channel estimation unit (114—1,... 114—k of each user terminal on the receiving side.

 K and 124-1,..., 124 -K), and is fed back from each communication terminal k to the base station 101 through a feedback channel 151.

[0009] MIMO transmission section 104 receives n parallel signals from signals input from scheduling section 103.

 T

To a data stream. Each data stream corresponds to one antenna. Then, the MIMO transmission unit 104 handles these parallel data streams. Encoding, interleaving, and modulation processes are performed on each of the n antennas 105.

 T

 Transmit to the other antenna 112-1,… 、 112—K and 122—1,… ゝ 122—K.

[0010] In user terminal k (k = l, 2, ···, K), the channel estimator (114—1,…, 114—K and 124—1,…, 124—K) ) Estimates the channel characteristics from the pilot signal transmitted from the base station 101, and the channel characteristics matrix H (t) (k = l, 2,

 k

 ... get K). Each user terminal then feeds back the channel characteristic matrix H (t) to the feedback channel k.

 The channel 151 is fed back to the scheduling unit 103 of the base station 101. The channel estimator (114-1, 1, ゝ 114—K and 124-1, K1, 124—K) converts the obtained channel characteristic matrix H (t) into the MIMO detector (113-1, 1, ···). ..., 113—K and 123—1,… ゝ 12

 k

 Output to 3—K). The MIMO detectors (113—1,..., 113—K and 123—1,..., 123—K) use the channel characteristic matrix H (t) for MIMO detection of the received signal.

 k

 Disclosure of the invention

 Problems to be solved by the invention

However, in the conventional multiuser scheduling method as described above, the base station compares channel characteristics with each user terminal at each transmission time, and the most appropriate one user terminal or plural user terminals are compared. There are the following problems in order to select the user terminal for communication.

 [0012] First, there is a problem that overhead due to feedback is large. In the conventional scheduling method, since the base station schedules multiple user terminals at each transmission time, it is necessary to grasp the channel characteristics of all user terminals at each transmission time. For this reason, in the conventional scheduling method, every user terminal needs to feed back their channel characteristics to the base station at every transmission time, that is, every scheduling time. .

[0013] Second, there is a problem of high complexity. The processing mainly performed in multiuser scheduling is comparison processing and selection processing. That is, in multi-user scheduling, the base station compares the channel characteristics fed back from each user terminal and selects the user terminal with the best channel characteristics. Therefore, scheduling complexity increases as the total number of user terminals subject to scheduling increases. Get higher. In particular, when performing multi-user scheduling by time division and space division at the same time, the iterative scheduling method is usually used. Therefore, if the number of user terminals is large, the processing complexity becomes higher and scheduling may not be realized.

[0014] Third, there is a problem of fairness. Fairness indicates whether or not the probability that each user terminal is selected in scheduling is equal. In addition to fast fading, factors that affect the channel characteristics of each user terminal include slow fading and the distance between the base station and each terminal. If the number of user terminals is large, the user terminals located at the cell edge or shadowing location have a very low probability of being selected by scheduling, and may not be able to communicate with the base station for a long time. . When each user terminal k (k = l, 2, ..., K) competes with another K-one user terminal at the same time, the user terminal located at the cell edge has a large fading and channel characteristics. It ’s bad, so it ’s a weak position for a long time. In order to solve this problem, a scheduling method based on, for example, proportional fairness is generally used. However, in this method, since it is necessary to calculate the SNR after reception in advance, there is a problem that applicable radio communication systems are limited. For example, in a wireless communication system that uses multi-user scheduling based on time division and space division at the same time, it is often impossible to calculate the SNR value after reception beforehand.

 [0015] An object of the present invention is to provide a radio communication apparatus capable of reducing the overhead of feedback information and the computational complexity of scheduling, and optimizing system throughput and fairness between user terminals. It is to provide a scheduling method.

 Means for solving the problem

[0016] The wireless communication apparatus of the present invention includes grouping means for dividing the plurality of user terminals into a plurality of groups based on the average transmission power of each of the plurality of user terminals, and one group of users for each transmission time. A configuration is provided that includes a scheduling control unit that performs scheduling only for terminals, and an extraction unit that extracts transmission data according to the scheduling result. The invention's effect

 [0017] According to the present invention, the overhead of feedback information and the computational complexity of scheduling can be reduced, and the throughput of the system and the fairness between user terminals can be optimized.

 Brief Description of Drawings

 FIG. 1 is a block diagram showing a configuration of a conventional MIMO communication system

 FIG. 2 is a block diagram showing a configuration of a MIMO communication system according to an embodiment of the present invention.

FIG. 3 is a block diagram showing a configuration of a scheduling unit according to an embodiment of the present invention.

 FIG. 4 is a flowchart showing a scheduling method according to an embodiment of the present invention.

 [Figure 5] Diagram showing simulation results of fairness performance

 [Figure 6] Diagram showing simulation results of system capacity performance

 BEST MODE FOR CARRYING OUT THE INVENTION

 [0019] The above problem is related to the number of user terminals that are scheduled simultaneously, and the above problem becomes more prominent as the number of user terminals that perform scheduling simultaneously increases. Therefore, in the present invention, group scheduling is performed in a multi-user MIMO communication system. In this group scheduling, first, all user terminals are divided into a plurality of groups, and scheduling is performed for only one group of user terminals at each scheduling time, that is, at each transmission time. That is, in the present invention, scheduling is performed for each group in order on the time axis, and scheduling is performed only for some user terminals at each transmission time. In the present invention, the correspondence between each user terminal and each group, that is, which user terminal belongs to which group differs depending on the time zone according to the channel characteristics of each user terminal. In other words, in the present invention, the correspondence between each user terminal and each group is variable, and the grouping situation can be changed according to the time zone.

[0020] Thereby, in the present invention, the number of user terminals scheduled at the same time can be reduced, so that the overhead of feedback information can be reduced and the computational complexity of scheduling can be reduced. it can. Further, in the present invention, in each time zone, based on the statistical channel characteristics of each user terminal in each time zone. Therefore, it is possible to improve the throughput of the entire system and to improve the fairness among user terminals.

 Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. In order to make the present invention easier to understand, description of unnecessary detail functions is omitted.

FIG. 2 is a block diagram showing a configuration of the MIMO communication system according to the present embodiment.

As shown in FIG. 2, the MIMO communication system according to the present embodiment includes a base station 201 and K user terminals (210-1,..., 210-K).

[0024] The base station 201 includes a data storage unit 202, a scheduling unit 203, and a MIMO transmission unit 20

4 and n transmitting antennas (205-1,..., 205-n).

 T T

 [0025] Each of the K user terminals has n (k) receive antennas (211—1,..., 211—K).

 R

 Channel estimation unit (212—1,..., 212—K) and average transmission power calculation unit (213—1,..., 2

13—K), MIMO detector (214-1, 1, 214—K), and feedback information generator (2

15—1,… 215—K).

[0026] In the base station 201, the data storage unit 202 stores each user terminal 210-k (k = 1, 2, ...,

The data to be transmitted is stored in K).

[0027] Scheduling section 203 schedules transmission data to each user terminal based on the channel estimation value to which each user terminal power is also fed back and information indicating the average transmission power, and MIMO transmission section 204 sends the scheduled transmission data. Output to. Details of the schedule using will be described later.

[0028] MIMO transmission section 204 performs transmission data input from scheduling section 203.

Perform MIMO transmission processing and output to each of n antennas. That is, MIMO

 T

 The transmission unit 204 first transmits the transmission data input from the scheduling unit 203 to n patterns.

 T

 Convert to a reall data stream. Each data stream corresponds to a respective antenna. Then, the MIMO transmission section 204 performs encoding, interleaving, and modulation processing on these parallel data streams, and outputs them to n antennas.

 T

 To help.

[0029] Each antenna (205-1,..., 205-n) transmits transmission data after MIMO transmission processing. 1 to the terminal 210—k (k = l, 2,..., K).

[0030] Each user terminal 210—k (k = l, 2,..., K) [koo! /, And antennas (211—1,... The signal to be transmitted is received and output to the channel estimation unit (212-1,..., 2 12-K).

 [0031] Channel estimation units (212-1,..., 212-K) perform channel estimation of received signals and output channel estimation values to a MIMO detection unit, an average transmission power calculation unit, and a feedback information generation unit. To do.

 [0032] The MIMO detection unit (214-1, ..., 214-K) performs MIMO detection on the received signal based on the channel estimation value.

[0033] The average transmission power calculation unit (213-1,..., 213-K) calculates the average transmission power based on the channel estimation value and outputs the average transmission power to the feedback information generation unit.

[0034] The feedback information generation unit (215-1, ..., 215-K) generates feedback information from the input channel estimation value and the average transmission power, and transmits the feedback information from the antenna to the base station via the feedback channel. Send to 201.

 Hereinafter, a specific calculation method of the average transmission power will be described.

 [0036] The average transmission power calculation unit (213—1,..., 213—K) has a time t = n XT (n = 0, 1, 2,...) Ο

 Every time, the communication time between time t-T and time t between base station 201 and own terminal k

 0 0

 Calculate the average transmission power C (t). Here, the average capacity or average F criterion

 Calculate the uniform transmission power. The calculation time interval T for calculating the average transmission power is set in the initial setting of the MIMO communication system, and this T is set much longer than the signal transmission interval at the base station.

 <0037> <Calculation method of average transmission power based on average capacity>

 In this method, the average value of transmission capacity during the period from time t—T to time t is averaged.

 0 0

 Calculated as average transmission power. The calculation formula is shown by formula (1).

[Number 1] M = ^ _ (H _k (t)) dt-(i) where is the channel characteristic H _k (t)

 Where p is the average S N R.

[0038] <Calculation method of average transmission power based on average F criterion>

 In this method, the average of the channel characteristic matrix in the time from time t—T to time t

 0 0

 Calculate the average value of F standard value (Norm) as the average transmission power. The calculation formula is shown by formula (2).

 [Equation 2]

^{C G} -T ^{F (H} … （ ² )

Since F (H _k (t)) indicates the Frobenius norm value when the channel characteristic is H _k (t), the channel capacity is)) = 22 ¹ ) I.

As described above, in the present embodiment, each user terminal 210-k (k = l, 2,..., K) calculates the average transmission power value C (t) for each time T, Fee for scheduling unit 203 of base station 201 k 0

 Doback. Therefore, according to the present embodiment, it is possible to reduce the overhead of feedback information by increasing the feedback time interval.

 Next, details of the scheduling unit 203 will be described with reference to FIG. FIG. 3 is a block diagram showing a configuration of scheduling section 203 according to the present embodiment. The scheduling unit 203 includes a grouping unit 301, a scheduling control unit 302, and a data extraction unit 303.

 [0041] The grouping unit 301 has an average transmission power C (t) (k k 0) to which each user terminal power is also fed back.

= 1, 2, ~, K) All user terminals are divided into groups. Κ indicates the number of user terminals that are active in the cell of base station 201, that is, the number of user terminals that desire communication with base station 201. In an actual communication system, an active user terminal is identified by a control signal. Also, Ν is the number of groups, and the specific number is set in the initial setting of the communication system. In addition, the grouping unit 301 generates a notification signal for notifying each user terminal of the grouping result every time grouping is performed. And output to MIMO transmission section 204. This notification signal is transmitted to each user terminal via the forward signaling channel. In the present embodiment, one of the following grouping methods is used.

 [0042] <Grouping method 1>

 Groupy method 1 is based on fairness optimization. That is, in the grouping method 1, grouping is performed so that the probability of communication with a base station is as uniform as possible between user terminals. For this reason, the grouping method 1 increases the fairness among user terminals by minimizing the difference in average transmission power between user terminals in each group.

 [0043] Specifically, first, grouping section 301 rearranges information C (t) (k = 1, 2,, K) indicating the average transmission power, to which each user terminal power is also fed back, in order. Sort order is large k 0

 Order or small !, order! /, Or deviation.

 [0044] For example, if C (t) ≥ C (t) ≥ C (t) ≥ C (t) ≥ ..., the data is rearranged in descending order. Kl 0 k2 0 k3 0 k4 0

 Then, {c (t), c (t), c (t), c (t), —}. Kl 0 k2 0 k3 0 k4 0 for sorting in ascending order

 The reverse is true.

 [0045] Next, grouping section 301 groups user terminals having a similar average transmission power into one group. Specifically, when all user terminals are to be grouped into N groups, according to the above arrangement order, KZN user terminals are grouped as the first group from the first user terminal k to KZN + 1 to K. Group all user terminals so that / N user terminals are in the second group. As a result, group 1 becomes G = {(C (t), C (t), ---, C (t)}, and group 2 becomes G = {C (t), C kl 0 k2 0 k (K / N) 0 2 k (K / N + 1) 0 k (K /

(t), ---, C (t)} and group 3 is G = {C (t), C (t),

N + 2) 0 k (2K / N) 0 3 k (2K / N + l) 0 k (2K / N + 2) 0

•••, c (t)}.

 k (3K / N) 0

 [0046] According to this grouping method 1, when scheduling is performed within one group, the number of competing user terminals is small and the average transmission power (channel characteristics) between the user terminals is close. The probability that a terminal is selected is almost equal. Therefore, according to the doubling method 1, the fairness between user terminals can be improved.

[0047] In this grouping method 1, a threshold value is set in advance, a plurality of fed back average transmission powers are compared, and a user terminal whose average transmission power difference is smaller than the threshold is set to 1 group. You can group them to make a loop.

 [0048] <Grouping method 2>

 Groupie method 2 is a method based on throughput optimization. That is, in the grouping method 2, grouping is performed so that the throughput of the entire multi-user MIMO communication system is maximized. For this reason, Group 2 method 2 minimizes the loss of communication system throughput by maximizing the difference in average transmission power between user terminals in each group.

 [0049] Specifically, first, grouping section 301, as in group method 1, information C (t) (k = l, 2, ... , K) in order k 0

 Rearranges. The arrangement order is either large order or small order.

 [0050] For example, if C (t) ≥ C (t) ≥ C (t) ≥ C (t) ≥ ..., they are arranged in ascending order kl 0 k2 0 k3 0 k4 0

 In other words, {c (t), c (t), c (t), c (t. When sorting in descending order kl 0 k2 0 k3 0 k4 0

 Is the opposite.

 [0051] Next, the grouping unit 301 performs grouping so as to divide user terminals having similar average transmission powers into different groups as much as possible. Specifically, according to the above-mentioned arrangement order, user terminals are selected one by one for every N user terminals, and K ZN user terminals whose intervals are N are grouped into the first group. That is, group 1 has G = (C (t), C

 1 kl 0

(t), C (t), ...}, and group 2 is G = (C (t), C (t), C (tk (N + 1) 0 k (2N + l) 0 2 k2 0 k ( N + 2) 0 k (2N + 2) 0

) "} And Group 3 becomes G = {C (t), C (t), C (t),"-}.

 3 k3 0 k (N +3) 0 k (2N + 3) 0

 [0052] According to this grouping method 2, since the difference in average transmission power between user terminals in each group is large, resources can be used without waste, and the throughput of the entire communication system is improved. can do.

[0053] In this grouping method 2, a threshold range is set in advance, a plurality of fed back average transmission powers are compared, and a user terminal whose difference in average transmission power is within the threshold range is set to 1. You can group them like groups!

[0054] In the above description, the method of grouping using information indicating the average transmission power to which the user terminal power is also fed back has been described. However, grouping may be performed using other information. The scheduling control unit 302 schedules a plurality of groups (G, G,..., G) input from the grouping unit 301 at different transmission times for each group.

1 2 N

 Do one ring. That is, the scheduling control unit 302 performs scheduling for only one group of user terminals at each transmission time, and performs scheduling for each group in turn. Then, the scheduling control unit 302 outputs the scheduling result to the data extraction unit 303.

 [0056] In the present embodiment, one of the following scheduling methods is used.

 [0057] High scheduling method 1>

 Scheduling method 1 is a time division scheduling method. Specifically, based on the current channel characteristics between the base station 101 and each user terminal 210—k (k = l, 2,..., K) at each transmission time, the user terminal is in accordance with a predetermined rule. Select one at a time for communication.

 [0058] <Scheduling Method 2>

 Scheduling method 2 is a scheduling method using time division and space division. Specifically, based on the current channel characteristics between the base station 101 and each user terminal 210—k (k = l, 2,..., K) at each transmission time, a plurality of users are identified according to a predetermined rule. Select a terminal at the same time to allow simultaneous communication. That is, in this scheduling method 2, user terminals in group 1 are scheduled at transmission time 1, user terminals in group 2 are scheduled at transmission time 2, and so on. Scheduling is performed for the user terminals in each group (G 1,..., G 1) in order. This scheduling

N

 , Receiving user terminals {k, k,-", k} corresponding to the respective transmitting antennas {1,2,.

 T 1 2 ηΤ is determined. Then, scheduling control unit 302 generates a notification signal for notifying the result of scheduling and outputs the notification signal to transmission unit 204. This notification signal is transmitted to the corresponding user terminal k via the forward signaling channel.

[0059] The data extraction unit 303 extracts transmission data from the data storage unit 202 according to the scheduling result, that is, extracts data of the user terminal 1 for the antenna 1 and data of the user terminal 2 for the antenna 2. Extract data up to user k in order And output to the MIMO transmitter 204.

 Thus, according to the present embodiment, all user terminals are grouped every time T based on the average transmission power of each user terminal fed back every time T. Then, scheduling is performed only for one group of user terminals at each transmission time, and scheduling is performed for each group in order of transmission time. In addition, since the calculation time T at the user terminal having an average transmission power that does not feed back information at every transmission at the base station is the feedback time interval, the frequency of feedback is reduced, and therefore The overhead of idback information can be reduced. In addition, since scheduling is performed for only one group at each transmission time, scheduling complexity can be reduced. In addition, by adopting the above grouping method, it is possible to improve the fairness between user terminals and the snooping output.

 FIG. 4 is a flowchart showing the scheduling method according to the present embodiment. Hereinafter, an example of a scheduling method for multi-users will be described with reference to FIG.

 [0062] In downlink transmission of a multi-user MIMO communication system, the base station 201 transmits data from n transmission antennas 205. Each user terminal k (k = l, 2, ..., K) is

T

 , Each having n (k) receive antennas and receiving data from the base station 201

 R

 The following flow is started.

[0063] In step 401, the average transmission power calculation unit (213-1, ..., 213-K) uses the above calculation method to calculate the base station between the previous time tT and the current time t. Plane between 201 and own terminal

 0 0

 Calculate the uniform transmission force C (t). This average transmission power C (t) is k 0 k 0 through the feedback channel.

 And fed back to the scheduling unit 203.

[0064] In step 402, grouping section 301 assigns all user terminals to N groups G 1, G 2,... According to the grouping method described above based on the fed back average transmission power of each user terminal. Group into G.

 1 2 N

 [0065] Since scheduling for G is performed first (transmission time 1), n = l is set in step 403 next.

[0066] In step 404, the scheduling control unit 302 uses the scheduling method described above to determine each user terminal of the group G based on the channel characteristics of the user terminals of the group G. To schedule.

 [0067] When G scheduling is completed, in step 405, the data extraction unit 303 extracts transmission data from the data storage unit 202 according to the scheduling result.

In step 406, n = n + 1 is set.

 [0069] In Step 407, it is determined whether or not n> N, that is, whether or not all groups have been scheduled. If it is determined that all groups have been scheduled (step 407: YES), the process returns to step 402 and the next time (t + T

 0

) Group based on new feedback information. On the other hand, if it is determined that all scheduling has not been completed (step 407: NO), the process returns to step 404 to perform scheduling for the remaining groups.

[0070] As described above, the grouping process in the grouping unit 301 is performed at each time interval T in which the average transmission power is fed back. That is, at time t = n XT (n = 0, 1, 2, ...;)

 0

 The unifying unit 301 groups the user terminals based on the average transmission power to which the user terminal power is also fed back, and obtains G 1, G 2,-, and G 1. And the next time t + from the current time t

 1 2 N 0 0

In the period up to T, the scheduling control unit 302, in order of transmission time, groups G, G, ---, G

 Scheduling one group at a time for 1 2

 N

 FIG. 5 is a diagram illustrating a simulation result of fairness “performance” between the scheduling method of the present invention and the conventional scheduling method. FIG. 6 is a diagram showing a simulation result of the system capacity performance between the scheduling method of the present invention and the conventional scheduling method.

 [0072] In these simulations, the number of transmit antennas of the base station is n = 4, and the user terminal

 T

 The number of each receiving antenna is n (1 = 2, and 1 ^ = 1, 2, ~ K. Flat fading

 R

 The average received SNR at the cell edge is OdB, the path loss factor is α = 3, and the number of groupings is Ν = 5.

[0073] In FIG. 5, the vertical axis indicates the probability that the user terminal closest to the base station is selected, and the horizontal axis indicates the number of user terminals. In FIG. 5, the curve marked with “*” shows fairness by the conventional method, the curve marked by “〇” shows fairness by the method based on the throughput optimization of the present invention, and the curve marked by “△”. Indicates fairness by a method based on fairness optimization of the present invention. Fig 5 Thus, according to the present invention, it is understood that the fairness of each user terminal in scheduling is improved, and the effect is particularly remarkable when grouping based on fairness optimization.

 In FIG. 6, the vertical axis indicates the throughput loss of the entire communication system, and the horizontal axis indicates the number of user terminals. In addition, in FIG. 6, the curve with a mark shows the throughput by the conventional method, the curve with a circle “'” shows the throughput by the method based on the throughput optimization of the present invention, and the curve with a “△” mark shows the fairness optimization of the present invention. The throughput by the method based on optimization is shown. From FIG. 6, it can be seen that when grouping is performed based on the throughput optimization of the present invention, the throughput loss of the entire system is minimized, and the effect is particularly remarkable when the number of users is large.

 [0075] As described above, according to the present invention, scheduling is performed because user terminals are grouped based on the average transmission power of each user terminal, and scheduling is performed for only one group at each transmission time. Can be reduced to the conventional lZf (N). f () is a polynomial or exponential function, and depends on the scheduling method used. In addition, information is not fed back from the user terminal at each transmission time. Since information is fed back every time the average transmission power is calculated at the user terminal, the amount of feedback information is reduced. It can be reduced to 1 / N. Furthermore, fairness or throughput can be improved depending on the grouping method employed. Furthermore, according to the present invention, it is possible to obtain a compromise between overhead, complexity, throughput, and fairness of various patterns by selecting parameters with high flexibility (grouping method, number of groups, etc.).

 [0076] Note that the user terminal in the above embodiment is a mobile terminal such as a mobile phone, a PDA, or a notebook computer.

 [0077] Further, in the above embodiment, the power explained in the case where the present invention is applied to a single-carrier non-spread MIMO system. The present invention is not limited to a single-carrier non-spread MIMO system. It can also be applied to OFDM systems and spread Ml MO systems.

[0078] Compared to single-carrier MIMO system, MIMO—Frequency in OFDM system Signals are also transmitted in the region, that is, signals are transmitted using a plurality of subcarriers simultaneously. Since there is no inter-signal interference in each subcarrier in the frequency domain in the MIMO-OFDM system, the MIMO-OFDM system can be regarded as N parallel single-carrier MIMO systems (N is the number of subcarriers used). Therefore, the present invention can be easily applied to a MIMO-OFDM system.

[0079] Also, compared with a non-spread MIMO system, a spread MIMO system transmits a signal even in the code domain, that is, a signal is transmitted using a plurality of orthogonal codes simultaneously. Since the codes are orthogonal to each other in the spread MIMO system, the spread MIMO system can be regarded as N parallel non-spread MIMO systems (N is the number of spread codes to be used). Therefore, the present invention can be easily applied to a spread MIMO system.

 As described above, various modifications, replacements, and additions can be made without departing from the spirit and scope of the present invention described in the present invention through the embodiments.

 [0081] The disclosures of the description, drawings and abstract contained in the Chinese application with the Chinese application number 200610059785.5 filed on March 7, 2006 are all incorporated herein by reference.

 Industrial applicability

The present invention can be applied to a communication system that performs scheduling of user terminals.

Claims

The scope of the claims  [1] Based on the average transmission power of each of the plurality of user terminals !, grouping means for dividing the plurality of user terminals into a plurality of groups;  Scheduling control means for scheduling only one group of user terminals at each transmission time; and  An extraction unit that extracts transmission data according to the scheduling result.  [2] The calculation means includes:  An average value of transmission capacity is calculated as the average transmission power.  The wireless communication device according to claim 1. [3] The calculation means includes:  The average F criterion value of the matrix is calculated as the average transmission power.  The wireless communication device according to claim 1. [4] The grouping means includes:  Dividing the plurality of user terminals into the plurality of groups so that a difference in average transmission power between user terminals in each group is minimized;  The wireless communication device according to claim 1. [5] The grouping means includes:  Dividing the plurality of user terminals into the plurality of groups so that a difference in average transmission power between user terminals in each group is maximized;  The wireless communication device according to claim 1. [6] Based on the average transmission power of each of the plurality of user terminals !, dividing the plurality of user terminals into a plurality of groups;  Scheduling for only one group of user terminals at each transmission time; and  Extracting transmission data according to the scheduling result; and a scheduling method comprising: