WO2014207965A1

WO2014207965A1 - Base station and central station

Info

Publication number: WO2014207965A1
Application number: PCT/JP2013/085021
Authority: WO
Inventors: Tsuguhide Aoki; Yuji Tohzaka; Noritaka Deguchi
Original assignee: Kabushiki Kaisha Toshiba
Priority date: 2013-06-24
Filing date: 2013-12-19
Publication date: 2014-12-31
Also published as: JP2015005943A

Abstract

According to an embodiment, a base station includes a communication unit which communicates with a mobile station and an interface between the base station and a central station. If first data transmitted by the mobile station in uplink in a first sub frame is not correctly received, the communication unit does not transmit, to the mobile station, a first response indicating that the first data has not been correctly received. If the first data is not correctly received, the interface receives, from the central station, a scheduling result which allows the mobile station to retransmit the first data in a second sub frame after the first sub frame. The communication unit transmits control information based on the scheduling result to the mobile station.

Description

TITLE

BASE STATION AND CENTRAL STATION

CROSS-REFERENCE TO RELATED APPLICATIONS This application is based upon and claims the benefit of priority from Japanese Patent Application No. 2013- 131456, filed June 24, 2013, the entire contents of which are incorporated herein by reference.

FIELD

Embodiments described herein relate to wireless communication.

BACKGROUND

In, for example, recent cellular systems such as LTE (Long Term Evolution) stipulated by the 3GPP (Third

Generation Partnership Project) , XGP (extended Global

Platform) complying with the next-generation PHS (Personal Handy-phone System) standard, and WiMAX (Worldwide

Interoperability for Microwave Access) , neighboring cells may use the same frequency to improve a frequency

utilization efficiency. Since these cellular systems generally adopt FDMA (Frequency Division Multiple Access) or OFDMA (Orthogonal Frequency Division Multiple Access) , inter-cell interference causes a problem when neighboring cells use the same frequency.

FFR (Fractional Frequency Reuse) is well known as a technique of mitigating inter-cell interference. According to FFR, a base station classifies frequencies can be allocated into first frequencies and second frequencies. The base station then allocates a first frequency to a user located at the center of a cell, and allocates a second frequency to a user located at the edge of the cell. With respect to the first frequencies, a neighboring cell can use (reuse) the same frequency. On the other hand, with respect to the second frequencies, a neighboring cell uses a different frequency. For the first frequencies, users located at the centers of neighboring cells are away from each other, so they are highly insusceptible to inter-cell interference. For the second frequencies, users located at the edges of neighboring cells may be close to each other but these neighboring cells use different frequencies, so these users do not experience inter-cell interference.

According to FFR, however, if a number of users are located at the edge of a given cell, the number of first frequencies reusable in neighboring cells including the given cell decreases, thereby decreasing the frequency utilization efficiency. Note that if a first frequency is also allocated to a user located at the edge of a cell, inter-cell interference increases at the first frequency, thereby decreasing the throughput .

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a view exemplifying a wireless communication system according to the first embodiment;

FIG. 2 is a timing chart exemplifying the operation of the wireless communication system according to the first embodiment ;

FIG. 3 is a block diagram exemplifying a base station according to the first embodiment;

FIG. 4 is a table exemplifying QCIs;

FIG. 5 is a block diagram exemplifying a central station according to the first embodiment;

FIG. 6 is a view showing an example of a PRB

arrangement in frequency-domain scheduling;

FIG. 7 is a view exemplifying channel information stored in a channel information storage unit shown in

FIG. 5;

FIG. 8 is a graph exemplifying the relationship between an SINR and throughput;

FIG. 9 is a view exemplifying a frequency-domain scheduling result;

FIG. 10 is also a view exemplifying a frequency-domain scheduling result;

FIG. 11 is a timing chart exemplifying the operation of the wireless communication system according to the first embodiment ;

FIG. 12 is a view for explaining a problem caused by retransmission; and

FIG. 13 is a timing chart exemplifying the operation of the wireless communication system according to the first embodiment.

DETAILED DESCRIPTION

Embodiments will be described below with reference to the accompanying drawings .

According to an embodiment, a base station includes a communication unit which communicates with a mobile station and an interface between the base station and a central station. The interface transmits information necessary for scheduling performed by the central station to the central station. If first data transmitted by the mobile station in uplink in a first sub- frame is not correctly received, the communication unit does not transmit, to the mobile station, a first response indicating that the first data has not been correctly received. If the first data is not correctly received, the interface receives, from the central station, a scheduling result which allows the mobile station to retransmit the first data in a second sub- frame after the first sub- frame. The communication unit transmits control information based on the scheduling result to the mobile station.

Note that the same or similar reference numerals denote the same or similar elements hereinafter and a repetitive description thereof will be basically omitted.

(First Embodiment)

As shown in FIG. 1, a wireless communication system according to the first embodiment includes a base station 1100, mobile stations 1200 and 1300, a base station 2100, mobile stations 2200 and 2300, and a central station 3000. The base stations 1100 and 2100 and the central station 3000 are connected to a network 4000. As will be described later, when the scheduler of the central station 3000 cooperates with the schedulers of the base stations 1100 and 2100, the base stations 1100 and 2100 can mitigate inter-cell interference while improving the frequency utilization efficiency. Therefore, the base stations 1100 and 2100 can achieve a high throughput. The central station 3000 may be called a CCC (Cell Cluster Controller) .

Note that the base stations 1100 and 2100 and the central station 3000 may be arranged within a single chassis or different chassis.

If the base stations 1100 and 2100 and the central station 3000 are arranged within different chassis, the central station 3000 may be connected to the base stations 1100 and 2100 via, for example, an optical fiber, Ethernet, or wireless LAN (Local Area Network) . If the base stations 1100 and 2100 and the central station 3000 are arranged within different chassis, a delay occurs due to

communication between the central station 3000 and the base station 1100 or 2100.

In the following description, assume that the central station 3000 is connected to the base stations 1100 and 2100 via Ethernet, and a delay of about 1 msec occurs due to communication between the central station 3000 and the base station 1100 or 2100.

The mobile stations 1200 and 1300 are wirelessly connected to the base station 1100. Similarly, the mobile stations 2200 and 2300 are wirelessly connected to the base station 2100. Note that in the following description, the mobile station and base station use the wireless interface of LTE (especially, FDD (Frequency Division Duplex) -LTE) for the sake of convenience .

For example, the base station 1100 (or base station 2100) includes a PHY (Physical) layer, a MAC (Media Access Control) layer, an RLC (Radio Link Control) layer, a PDCP (Packet Data Convergence Protocol) layer, a network

processor, a channel information storage unit, a scheduler, a buffer information storage unit, and a PDCCH (Physical Downlink Control CHannel) generation unit, as exemplified in FIG. 3.

The PHY layer serves as an interface with a physical channel. Therefore, the PHY layer can be considered as, for example, a communication unit for communicating with a mobile station. The data path between the PHY layer and the MAC layer is called a transport channel. The MAC layer executes HARQ (Hybrid Automatic Repeat reQuest) processing and the like. The data path between the MAC layer and the RLC layer is called a logical channel . The RLC layer executes ARQ (Automatic Repeat reQuest) processing and the like. The data path between the RLC layer and the PDCP layer is called a radio bearer. The PDCP layer executes encryption processing and the like. The network processor serves as an interface with the central station 3000.

The channel information storage unit stores pieces of channel information from the PHY layer and the RLC layer. The scheduler reads out the channel information stored in the channel information storage unit, as needed. The channel information can be acquired by various methods.

For example, an RSRP (Reference Signal Received Power) may be acquired as channel information. The RSRP is the reception power of a reference signal in downlink. In LTE, the mobile station needs to feed back, to a connected base station, the RSRP of another base station in addition to that of the connected base station. Note that the RSRP is an average value for the whole service band, and does not always represent a channel response at a specific frequency with high accuracy.

Channel information may be acquired using an SRS

(Sounding Reference Signal) in addition to or instead of the RSRP. The SRS is a signal to be transmitted by a mobile station in uplink. The base station generally receives only an SRS from a mobile station within its cell. The base station, however, may receive an SRS also from a mobile station in the cell of another base station and estimate an inter- cell interference component based on the received SRSs. Since an SRS can be transmitted using part of the service band, it is possible to estimate a channel response at a specific frequency with high accuracy by using the SRS.

The buffer information storage unit stores pieces of buffer information from the PHY layer and the MAC layer. The scheduler reads out the buffer information stored in the buffer information storage unit, as needed. A PDCCH generation unit generates a PDCCH as a downlink control channel according to a scheduling result from the

scheduler.

The scheduler performs radio resource control

(scheduling) based on the channel information read out from the channel information storage unit and the buffer

information read out from the buffer information storage unit .

More specifically, the scheduler executes payload selection processing of selecting data to be transmitted at a target time based on the priority level of data for which scheduling has been requested. In LTE, the priority level of data is evaluated based on a QCI (QoS (Quality of

Service) Class Identifier) . FIG. 4 exemplifies QCIs defined by the 3GPP. Note that in uplink scheduling, buffer information for each of four LCGs (Logical Channel Groups) into which QCIs are grouped is obtained from a mobile station, and thus the priority level is evaluated using the LCG.

Furthermore, the scheduler executes PRB (Physical Resource Block) allocation processing of selecting a frequency to be used to transmit the data (payload)

selected in the payload selection processing. The

scheduler also executes HARQ (that is, retransmission) control processing, MCS (Modulation and Coding Set) control processing, and the like. Note that an MCS is a

combination of a modulation scheme and error correction coding scheme to be applied to the data.

The payload selection processing will be referred to as time-domain scheduling hereinafter. On the other hand, the PRB allocation processing, MCS control processing, and retransmission control processing will be referred to as frequency-domain scheduling hereinafter.

A conventional base station autonomously performs time-domain scheduling and frequency-domain scheduling.

That is, a given base station performs scheduling

independently of that in its neighboring base station.

Therefore, unexpected inter-cell interference may occur, resulting in a decrease in throughput.

On the other hand, in the wireless communication system shown in FIG. 1, the central station 3000 performs at least frequency-domain scheduling for the base stations 1100 and 2100 at once. It is thus possible to omit

functional units for frequency-domain scheduling in the schedulers of the base stations 1100 and 2100.

Note that the base stations 1100 and 2100 may perform temporary frequency-domain scheduling, and then the central station 3000 may perform final frequency-domain scheduling. In this case, the result of the temporary frequency-domain scheduling is replaced by the result of the final

frequency-domain scheduling.

Alternatively, the central station 3000 may perform time-domain scheduling for the base stations 1100 and 2100 in addition to frequency-domain scheduling for them. It is therefore possible to omit functional units for time-domain scheduling in the schedulers of the base stations 1100 and 2100. In this case, the base stations 1100 and 2100 notify the central station 3000 of information such as buffer information necessary for time-domain scheduling.

As shown in FIG. 5, the central station 3000 includes a network processor, a time-domain scheduling information storage unit, a channel information storage unit, and a scheduler. The network processor serves as an interface with the base station 1100 or 2100.

The time-domain scheduling information storage unit stores time-domain scheduling results from the base

stations 1100 and 2100. The scheduler reads out the time-domain scheduling results stored in the time-domain scheduling information storage unit, as needed.

The channel information storage unit stores pieces of channel information from the base stations 1100 and 2100. The scheduler reads out the channel information stored in the channel information storage unit, as needed.

The scheduler performs frequency-domain scheduling based on the time-domain scheduling results stored in the time-domain scheduling information storage unit and the channel information stored in the channel information storage unit .

The wireless communication system shown in FIG. 1 can perform uplink scheduling, as shown in, for example,

FIG. 2. Note that in FIG. 2, a time unit is a subr frame (that is, 1 msec) .

To transmit a data buffer size to the base station 1100, the mobile station 1200 (or mobile station 1300) first transmits a BSR (Buffer Status Report) on a PUSCH (Physical Uplink Shared CHannel) (time to) .

The PHY layer of the base station 1100 receives the BSR transmitted at the time to. The buffer information held by the mobile station 1200 for uplink transmission is stored in the buffer information storage unit of the base station 1100.

The scheduler of the base station 1100 performs time-domain scheduling (time t2) . In the time-domain scheduling at the time t2, mobile station candidates each of which is given a transmission grant in a target

sub-frame are selected. More specifically, the scheduler reads out the buffer information from the buffer

information storage unit, and identifies, based on the buffer information, a mobile station which currently requests scheduling. Assume that at the time t2, the buffers of the mobile stations 1200 and 1300 store data and each of the mobile stations has already requested

scheduling.

Assume that the wireless communication system (LTE system) shown in FIG. 1 operates in, for example, a 10-MHz mode. Fifty PRBs (each PRB has a bandwidth of 180 kHz) in total exist in a bandwidth of 10 MHz. In this case, if two PRBs are used as a PUCCH (Physical Uplink Control CHannel) , the remaining 48 PRBs can be allocated to the mobile stations 1200 and 1300.

If the type of data for which the mobile station 1200 has requested scheduling is the same as that of data for which the mobile station 1300 has requested scheduling (that is, the LCGs of the data are the same) , the scheduler of the base station 1100 equally allocates the 48 PRBs to the mobile stations 1200 and 1300 (allocates 24 PRBs to each mobile station) at the time t2. Although not shown, it is assumed that the scheduler of the base station 2100 also equally allocates the 48 PRBs to the mobile stations 2200 and 2300 at the time t2. Note that as described above, the scheduler of the base station 1100 may perform temporary frequency-domain scheduling. For example, of the 48 PRBs, the scheduler may allocate 24 PRBs on the low frequency side to the mobile station 1200 and 24 PRBs on the high frequency side to the mobile station 1300.

The time-domain scheduling result (more specifically, information for identifying mobile stations each of which is given a transmission grant and information indicating the number of PRBs allocated to each of the mobile

stations) at the time t2 is transmitted to the central station 3000 via the network processor (time t3) . At the time t3, the central station 3000 is also notified of the channel information stored in the channel information storage unit.

Note that at the time t3, the temporary- frequency-domain scheduling result (more specifically, information indicating the frequency positions of the PRBs temporarily allocated to each mobile station) may also be transmitted to the central station 3000 via the network processor.

The network processor of the central station 3000 receives the time-domain scheduling result and the channel information transmitted at the time t3. The scheduler of the central station 3000 then performs frequency-domain scheduling using the received time-domain scheduling result and channel information (time t5) .

The scheduler of the central station 3000 performs frequency-domain scheduling based on, for example, a PRB arrangement shown in FIG. 6. According to the arrangement example shown in FIG. 6, the scheduler can allocate each of sub-band 1 and sub-band 2 used by the base station 1100 to one of the mobile stations 1200 and 1300. On the other hand, the scheduler can allocate each of sub-band 1 and sub-band 2 used by the base station 2100 to one of the mobile stations 2200 and 2300.

In this embodiment, the scheduler performs

frequency-domain scheduling using a throughput derived based on an SINR (Signal-to-Interference plus Noise Ratio) . As will be described below, it is possible to calculate an SINR based oh the pieces of channel information from the base stations 1100 and 2100. It is then possible to derive a throughput based on, for example, the relationship between the SINR and the throughput shown in FIG. 8.

The channel information stored in the channel

information storage unit of the central station 3000 may be as exemplified in FIG. 7. FIG. 7 shows a gain when each mobile station transmits data to a connected base station in uplink. Referring to FIG. 7, for example, when the mobile station 1200 transmits data to the base station 1100, a signal power received by the base station 1100 is 1.0 and an interference power observed in the base station 2100 is 0.1. Note that in the example shown in FIG. 7, common channel information is used for the whole service band. That is, it is assumed that the gain of the channel of sub-band 1 is equal to that of the channel of sub-band 2.

In an actual communication environment, however, there may be a difference between the gains of the channels of the sub-bands due to, for example, the influence of the multipath. To solve this problem, for example, the average value of the gains of the channels of sub-band 1 and sub-band 2 may be used as channel information common to the whole service band. Alternatively, channel information (that is, the gain of a channel) may be prepared for each sub-band.

As shown in FIG. 1, the mobile stations 1300 and 2200 are close to each other. When these mobile stations use the same sub-band, it is expected that inter-cell

interference readily occurs. If the mobile stations 1300 and 2200 are accommodated in sub-band 1 (or sub-band 2) , the scheduler of the central station 3000 can calculate an SINR observed in each of the base stations 1100 and 2100 as follows. In the following description, assume that the noise level of each of the base stations 1100 and 2100 is 0.01.

More specifically, the scheduler can calculate an SINR observed in the base station 1100 by:

SINR = «3dB (1)

0.5+0.01

Referring to FIG. 8, if SINR = 3 dB, a throughput of about 1 bps/Hz is achieved.

Furthermore, the scheduler can calculate an SINR observed in the base station 2100 by.

SINR = «3dB (2)

0.5+0.01

As described above, referring to FIG. 8, if SINR = 3 dB, a throughput of about 1 bps/Hz is achieved.

If, therefore, the mobile stations 1300 and 2200 are accommodated in sub-band 1, the scheduler evaluates that the total throughput of the base stations 1100 and 2100 is about 2 bps/Hz.

On the other hand, as shown in FIG. 1, the mobile stations 1200 and 2300 are away from each other. Even if these mobile stations use the same sub-band, it is expected that inter-cell interference hardly occurs. If the mobile stations 1200 and 2300 are accommodated in sub-band 1 (or sub-band 2) , the scheduler of the central station 3000 can calculate an SINR observed in each of the base stations 1100 and 2100 as follows.

More specifically, the scheduler can calculate an SINR observed in the base station 1100 and that observed in the base station 2100 by:

SINR = «10dB (3)

0.1+0.01

Referring to FIG. 8, if SINR = 10 dB, a throughput of about 2 bps/Hz is achieved. If, therefore, the mobile stations 1200 and 2300 are accommodated in sub-band 1, the scheduler evaluates that the total throughput of the base stations 1100 and 2100 is about 4 bps/Hz.

Furthermore, if only the mobile station 1300 (or the mobile station 1200, 2200, or 2300) uses sub-band 1 (or sub-band 2), inter-cell interference does not occur. If the mobile station 1300 is accommodated in sub-band 1, the scheduler of the central station 3000 can calculate an SINR observed in the base station 1100 as follows.

More specifically, the scheduler can calculate an SINR observed in the base station 1100 by: SINR = «20dB (4)

0.01

Referring to FIG. 8, if SINR = 20 dB, a throughput of about 4.7 bps/Hz is achieved. On the other hand, since the mobile stations 2200 and 2300 connected to the base station 2100 cannot transmit data, the throughput of the base station 2100 is 0 bps/Hz.

If, therefore, the mobile station 1300 is accommodated in sub-band 1, the scheduler evaluates that the total throughput of the base stations 1100 and 2100 is about 4.7 bps/Hz.

The scheduler of the central station 3000 can evaluate the total throughput of the base stations 1100 and 2100 for various PRB allocation processes. That is, the scheduler can perform scheduling so as to, for example, maximize the total throughput .

If, however, only the total throughput is used as the metric of frequency-domain scheduling, a mobile station located at the center of a cell tends to have an advantage over a mobile station located at the edge of the cell, and thus it may be impossible to fairly provide transmission opportunities to respective mobile stations.

To solve this problem, for example, total Proportional Fair (PF) may be used as a metric instead of the total throughput. Since PF considers the past throughput

variance of each mobile station, fair scheduling is

possible . The scheduler of the central station 3000 can

calculate the total PF of the mobile stations 1200, 1300, 2200, and 2300 (that is, all mobile stations in the coverage areas of the base stations 1100 and 2100) by:

∑Mi(n) (5)

i

where Mi (n) represents the PF of a mobile station i in a sub- frame n, and i represents an index for identifying a mobile station. Mi(n) can be calculated by:

( \ ^Ri (ⁿ)

Ai(n-l)

where Ri(n) represents the instantaneous throughput, which can be derived from an SINR, as described above. If, therefore, a frequency is not allocated to the mobile station i in the sub- frame n, Ri(n) = 0. Furthermore, in equation (6), Ai(n-l) represents a value based on the past throughput. Ai(n) can be calculated by:

Ai(n) = w6i(n)Ri(n)+ (l - w)Ai(n - 1) ( 7 )

According to equation (7) , if a frequency is allocated to the mobile station i in the sub-frame n, Oi(n) becomes 1; otherwise, σι (n) becomes 0. In equation (7), w

represents a weight for determining how much the past throughput is considered. For example, a value such as 1/1000 is used as the weight w.

At the time t5 shown in FIG. 2, the scheduler of the central station 3000 performs, for example, scheduling shown in FIG. 9 based on the total PF . More specifically, the scheduler determines that the mobile station 1200 is accommodated in sub-band 1 of the base station 1100, the mobile station 1300 is accommodated in sub-band 2 of the base station 1100, and the mobile station 2300 is

accommodated in sub-band 1 of the base station 2100.

Furthermore, the scheduler of the central station 3000 selects an appropriate MCS for the mobile stations 1200, 1300, and 2300 each of which is given a transmission grant.

The frequency-domain scheduling result (more

specifically, information indicating mobile stations each of which is given a transmission grant by each base

station, information indicating the positions of PRBs or a sub-band allocated to each of the mobile stations, and information indicating the value of an MCS applied by each of the mobile stations) at the time t5 is transmitted to the base stations 1100 and 2100 via the network processor of the central station 3000 (time t6) .

The network processors of the base stations 1100 and 2100 respectively receive the frequency-domain scheduling result transmitted at the time t6.

The scheduler of the base station 1100 inputs the frequency-domain scheduling result via the network

processor. The scheduler then instructs the PDCCH generation unit to generate a PDCCH. The PDCCH generation unit generates a PDCCH to notify the mobile station 1200 of an actual scheduling result. The PDCCH can carry control information indicating the identifier of a mobile station which is given a transmission grant, the locations

(frequency positions) of PRBs allocated to the mobile station, and the value of an MCS applied by the mobile station. The PDCCH generated by the PDCCH generation unit is transmitted at a time t8.

An FDD-LTE system defines that a mobile station transmits data on a PUSCH (Physical Uplink Shared CHannel) using parameters instructed by a PDCCH 4 msec after it receives the PDCCH. The mobile station 1200, therefore, transmits data on the PUSCH at a time tl2.

In the FDD-LTE system, a base station needs to

transmit a response to data (that is, information

indicating that the data has been correctly received or information indicating that the data has not been correctly received) 4 msec after it receives the data. If the data is correctly received, the base station transmits an acknowledgement (ACK) . The AC is transmitted on a PHICH (Physical Hybrid ARQ Indicator CHannel) . If, therefore, the data transmitted at the time tl2 is correctly received, the base station 1100 transmits an ACK on the PHICH at a time tl6.

At the time tl6, the base station 1100 can not only transmit the ACK on the PHICH but also transmit a new transmission grant on the PDCCH. The new transmission grant transmitted at the time tl6 is based on the result of time-domain scheduling performed by the base station 1100 at a time tlO and the result of frequency-domain scheduling performed by the central station 3000 at a time tl3.

Assume, for example, that the scheduler of the central station 3000 performs scheduling as shown in FIG. 10 based on the total PF at the time tl3 shown in FIG. 2. More specifically, the scheduler determines that the mobile station 1200 is accommodated in sub-band 1 of the base station 1100, the mobile station 2300 is accommodated in sub-band 1 of the base station 2100, and the mobile station 2200 is accommodated in sub-band 2 of the base station 2100.

Upon receiving the new transmission grant transmitted at the time tl6, the mobile station 1200 transmits data on the PUSCH at a time t20 4 msec after the time tl6.

As described above, if the scheduler of the central station 3000 performs frequency-domain scheduling for the base stations 1100 and 2100 at once, it is possible to improve the total throughput of the base stations 1100 and 2100. Furthermore, if the total PF is used as a metric in the above frequency-domain scheduling, it is possible to fairly provide transmission opportunities to the mobile stations 1200, 1300, 2200, and 2300 in the coverage areas of the base stations 1100 and 2100.

The operation example shown in FIG. 2, however, does not consider a data reception error in the base station. Even if a data reception error occurs in the base station, the central station performs frequency-domain scheduling without waiting for information of the reception error, as exemplified in FIG. 11. As a result, the mobile station performs retransmission which is not planned in the

frequency-domain scheduling. This may cause unexpected interference to decrease the throughput .

The base station 1100 cannot correctly receive

(demodulate) data transmitted by the mobile station 1300 at a time tl2 shown in FIG. 11. For example, the base station 1100 detects an error through a CRC (Cyclic Redundancy Check) result. Note that in FIG. 11, processing from a time to to a time til is the same as that shown in FIG. 2 and a description thereof will be omitted.

As described above, in the FDD-LTE system, the base station needs to transmit a response indicating whether data has successfully been received 4 msec after it

receives the data on the PUSCH. The base station 1100, therefore, transmits a NACK (Negative ACK) on the PHICH at a time tl6 4 msec after the time tl2.

Upon receiving the NACK transmitted at the time tl6, the mobile station 1300 retransmits the data transmitted at the time 12 on the PUSCH (using, for example, the same parameters as those used at the time tl2) at a time t20 4 msec after the time tl6. However, frequency-domain

scheduling for the time t20 has been done by the scheduler of the central station 3000 at a time tl3. Since the scheduler cannot know a data reception error in the base station 1100 at the time tl3, retransmission from the mobile station 1300 is not planned in the frequency-domain scheduling.

Assume, for example, that a frequency-domain

scheduling result at the time tl3 is as shown in FIG. 10. In this case, if the mobile station 1300 retransmits the data using sub-band 2, the frequency-domain scheduling result shown in FIG. 10 is interrupted, as shown in

FIG. 12. This interrupt causes inter-cell interference in sub-band 2. As a result, the SINR and throughput of sub-band 2 in the base station 2100 become lower than expected.

The main factor in this problem is that time-domain scheduling and frequency-domain scheduling have been moved up as compared with those in the general LTE system due to a delay caused by communication between the central station 3000 and the base station 1100 or 2100. As a result, it becomes difficult to immediately reflect, on the

frequency-domain scheduling, reception error information of the data transmitted on the PUSCH.

The base station according to this embodiment suspends retransmission of the data, as exemplified in FIG. 13, thereby allowing the central station to perform

frequency-domain scheduling in consideration of

retransmission. The base station 1100 does not correctly receive the data transmitted from the mobile station 1300 at a time tl2 shown in FIG. 13. Note that processing from a time to to a time til in FIG. 13 is the same as that shown in FIG. 2 or 11 and a description thereof will be omitted.

As shown in FIG. 13, the base station 1100 transmits an ACK on the PHICH instead of a NACK at a time tl6 4 msec after the time tl2. Since the mobile station 1300 receives the ACK, retransmission is not performed at a time t20 in accordance with the FDD-LTE system. Therefore, it is possible to avoid the throughput from decreasing due to a retransmission interrupt, as shown in FIG. 12.

It is, however, impossible to resolve loss of the data transmitted at the time tl2 by only transmitting an ACK instead of a NACK. To solve this problem, the scheduler of the base station 1100 considers reception error information in time-domain scheduling at a time tl8 after the time tl2. Based on the result of the time-domain scheduling, the scheduler of the central station 3000 performs

frequency-domain scheduling (time t21) . As a result, at a time t28 as the first retransmission opportunity after the time t20 (that is, 8 msec after the time t20) , the mobile station 1300 can retransmit the data already transmitted at the time tl2. Note that since retransmission at the time t28 has been planned in the frequency-domain scheduling performed by the scheduler of the central station 3000, it does not cause unexpected interference or decrease the throughput .

Note that various modifications can be made to the retransmission suspending processing by the base station explained with reference to FIG. 13. For example, although the base station 1100 transmits an ACK instead of a NACK at the time tl6 shown in FIG. 13, it may skip transmission of both a NACK and ACK (that is, a response indicating whether the data transmitted at the time tl2 has successfully been received may be omitted) . Furthermore, the retransmission suspending processing is effective for avoiding the

throughput from decreasing but prolongs the latency. The retransmission suspending processing, therefore, is not always suitable for data with a short allowable delay (for example, a packet delay budget in FIG. 4) . The

retransmission suspending processing may thus be applied to only data with a long allowable delay (for example, equal to or longer than a threshold) . In this case, the

retransmission suspending processing need not be applied to data with a short allowable delay amount (for example, shorter than the threshold) . That is, if the base station 1100 does not correctly receive data with a short allowable delay amount, it need only transmit a NACK.

As described above, in the wireless communication system according to the first embodiment, the central station performs frequency-domain scheduling for a

plurality of base stations at once. This frequency-domain scheduling maximizes a metric such as the total throughput of the plurality of base stations. According to the wireless communication system, therefore, it is possible to improve the throughput of the plurality of base stations by mitigating inter-cell interference while increasing the frequency utilization efficiency. Note that if the total PF of a plurality of mobile stations connected to a

plurality of base stations is adopted as the

above-described metric, it is possible to fairly provide transmission opportunities to the plurality of mobile stations .

Furthermore, if the central station and the plurality of base stations are arranged in different chassis, communication between the central station and each of the base stations causes a delay. In this case, therefore, data retransmission may interrupt the above

frequency-domain scheduling result, thereby causing

unexpected interference, which decreases the throughput. To solve this problem, in the wireless communication system according to the embodiment, if a data reception error occurs in a base station, the base station transmits an ACK instead of a NAC , or omits a response indicating whether the data has successfully been received. This avoids the above interrupt. On the other hand, the base station considers reception error information in subsequent

time-domain scheduling. Since, therefore, the data is retransmitted at a retransmission opportunity next to a normal retransmission opportunity (for example, 8 msec after the normal retransmission opportunity) , loss of the data in the base station is resolved.

The processing in the above-described embodiments can be implemented using a general-purpose computer as basic hardware. A program implementing the processing in each of the above-described embodiments may be stored in a computer readable storage medium for provision. The program is stored in the storage medium as a file in an installable or executable format. The storage medium is a magnetic disk, an optical disc (CD-ROM, CD-R, DVD, or the like) , a

magnetooptic disc (MO or the like) , a semiconductor memory, or the like. That is, the storage medium may be in any format provided that a program can be stored in the storage medium and that a computer can read the program from the storage medium. Furthermore, the program implementing the processing in each of the above-described embodiments may be stored on a computer (server) connected to a network such as the Internet so as to be downloaded into a computer (client) via the network.

While certain embodiments have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the inventions.

Indeed, the novel methods and systems described herein may be embodied in a variety of other forms; furthermore, various omissions, substitutions and changes in the form of the methods and systems described herein may be made without departing from the spirit of the inventions. The accompanying claims and their equivalents are intended to cover such forms or modifications as would fall within the scope and spirit of the inventions.

Claims

1. A base station comprising:

a communication unit configured to communicate with a mobile station; and

an interface between the base station and a central station,

wherein the interface transmits information necessary for scheduling performed by the central station to the central station,

if first data transmitted by the mobile station in uplink in a first sub- frame is not correctly received, the communication unit does not transmit, to the mobile

station, a first response indicating that the first data has not been correctly received,

if the first data is not correctly received, the interface receives, from the central station, a scheduling result which allows the mobile station to retransmit the first data in a second sub- frame after the first sub-frame, and

the communication unit transmits control information based on the scheduling result to the mobile station.

2. The base station according to claim 1, wherein if the first data is not correctly received, the communication unit does not transmit the first response to the mobile station and transmits, to the mobile station, a second response indicating that the first data has been correctly received.

3. The base station according to claim 1, wherein if the first data is not correctly received, the communication unit does not transmit the first response to the mobile station and does not transmit, to the mobile station, a second response indicating that the first data has been correctly received.

4. The base station according to claim 1, wherein if second data transmitted by the mobile station in uplink in a third sub-frame is not correctly received, the communication unit transmits, to the mobile station, a first response indicating that the second data has not been correctly received, and

the first data is data with an allowable delay not shorter than a threshold, and the second data is data with an allowable delay shorter than the threshold.

5. A central station comprising:

a scheduler configured to perform at least part of scheduling for a plurality of base stations at once; and an interface configured to receive information

necessary for scheduling performed by the scheduler from the plurality of base stations, and transmit a scheduling result to the plurality of base stations.

6. The central station according to claim 5, wherein the central station performs scheduling so as to maximize a total throughput of the plurality of base stations.

7. The central station according to claim 5, wherein the central station performs scheduling so as to maximize a total proportional fair of mobile stations connected to the plurality of base stations.

8. A base station which is arranged in a chassis different from that of a central station, comprising:

a communication unit configured to communicate with a mobile station; and

an interface between the base station and the central station,

the interface receives a scheduling result from the central station, and