JP2011035549A - Power control method for parallel channel - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce influence which a parallel channel with a low gain exerts to an entire FEC block in a transmission method using a Vertical Encoding method distributing one FEC block data to a plurality of parallel channels to be transmitted. <P>SOLUTION: In a power control method to a parallel channel in a transmission method using a Vertical Encoding method distributing one FEC block data to a plurality of parallel channels to be transmitted, power control is performed according to a gain difference of parallel channels, whereby influence which a parallel channel with a low gain exerts to an entire FEC block is reduced. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a power control method for a parallel channel for performing transmission power control for a plurality of parallel channels in data transmission via a plurality of parallel channels formed by, for example, a MIMO (Multiple-Input Multiple-Output) transmission method.

  The demand for high speed and large capacity in wireless broadband is increasing day by day. Under such circumstances, a MIMO transmission method has been introduced in mobile communication, wireless LAN, and the like as a method for dramatically improving the transmission capacity. In addition, OFDM that allocates data to each block by dividing the entire system band into multiple frequency blocks (referred to as subcarriers) with the feature that flexible resource control is possible and resistance to frequency selective fading is strong. In recent years, studies on MIMO-OFDM and MIMO-OFDMA schemes in which the above MIMO communication scheme is combined with / OFDMA (Orthogonal Frequency Division Multiplexing / Orthogonal Frequency Division Multiple Access) have been rapidly advanced.

  MIMO transmission is a transmission technique that utilizes the complexity of space and forms a plurality of independent parallel channels on the space axis, so that the channel capacity increases linearly according to the number of antennas.

  In such a background, Patent Document 1 describes a technique for effectively and efficiently determining the power allocation of each channel in a multi-channel communication system using an OFDM (Orthogonal Frequency Division Multiplexing) scheme. Thus, there is provided a method for allocating power (transmission antenna weight value) for each subchannel and antenna while taking into consideration the limitation of the maximum output value for each antenna.

JP 2005-527128 A

  The MIMO transmission system is a technique for improving the transmission speed by forming a plurality of independent parallel channels on the space axis. There is a gain difference between independent parallel channels formed in a MIMO transmission system, and not all of the paralleled signal streams are transmitted with the same quality. There are a horizontal encoding method and a vertical encoding method for parallelizing signals to be transmitted. In the horizontal encoding method, different FEC (Forward Error Correction) blocks are assigned to each parallel channel, and the modulation level is also different for each parallel channel.

  Therefore, the FEC block allocated to the high gain parallel channel can be demodulated without being affected by the low gain parallel channel. Furthermore, by assigning a high modulation level signal to a high gain parallel channel and assigning a low modulation level signal to a low gain parallel channel, it is possible to obtain a transmission capability asymptotic to Shannon's information theory. Horizontal encoding is the best signal parallelization method.

  On the other hand, the Vertical Encoding method is an allocation method in which a signal of one FEC block is allocated to each parallel channel, and all the parallel channels have the same modulation level. Therefore, when decoding the FEC block, the influence of the parallel channel having a low gain reaches the entire FEC block, and there is an adverse effect that the performance of the entire parallel channel depends on the parallel channel having a low gain. Also, since all the parallel channels have the same modulation level, if the modulation level is determined according to the parallel channel having a high gain, errors increase in the parallel channel having a low gain. On the other hand, when the modulation level is determined according to the parallel channel having a low gain, the parallel channel having a high gain becomes redundant and there is a problem that the transmission speed is greatly reduced.

  Therefore, the present invention has been made in view of the above points. In a transmission method using the Vertical Encoding method in which one FEC block data is distributed and transmitted to a plurality of parallel channels, a parallel channel with a low gain is used for the entire FEC block. It is an object of the present invention to provide a power control method for parallel channels that can reduce the influence on the channel.

  A power control method for parallel channels according to the present invention is a power control method for parallel channels in a transmission scheme using the Vertical Encoding method in which one FEC block data is distributed and transmitted to a plurality of parallel channels. The power control is performed according to the gain difference.

  According to the present invention, the power control according to the gain difference of each parallel channel can reduce the influence of the parallel channel having a low gain on the entire FEC block.

It is a block diagram which shows the structure of the radio base station apparatus for demonstrating the power distribution method with respect to the parallel channel which concerns on embodiment of this invention. It is a block diagram which shows the internal structure of the scheduling part shown in FIG. It is a figure for demonstrating a communication form. It is explanatory drawing of the power control method (when a coding rate is not considered) by average value and instantaneous channel quality. It is a flowchart of the power control using the channel quality level in the case of 2 parallel channels. It is a flowchart following FIG. It is explanatory drawing of the power control method (when considering a coding rate) by average value and instantaneous channel quality. It is explanatory drawing of the power control method by channel quality distribution.

  Embodiments of a power distribution method for parallel channels according to the present invention will be described below in detail with reference to the drawings. In addition, this invention is not limited by this Embodiment.

1. Apparatus Configuration FIG. 1 is a block diagram showing a configuration of a radio base station apparatus (transmitter) for explaining a power distribution method for parallel channels according to an embodiment of the present invention. In FIG. 1, a transmitter 100 performs an error correction encoding on a transmission bit sequence, which is user data, according to an error correction code type specified by the system, and generates one FEC block. A modulation unit 102 that performs primary modulation (mapping onto the constellation), a data distribution unit 103 that distributes modulation data to the number of parallel channels, and a resource that maps modulation data to radio resources according to a transmission scheme such as OFDM / OFDMA The mapping unit 104 performs radio signal processing (including OFDM processing by IDFT: Inverse Discrete Fourier Transfer) on the mapped transmission data, up-converts the signal after D / A (digital-analog) conversion, and generates an RF (Radio Frequency) band Baseband / RF processing to perform conversion to 105, parallel channel quality analysis unit 106 that acquires channel quality information of a parallel channel based on feedback information from the receiver, and CINR measurement unit that acquires CINR information of the receiving station based on feedback information from the receiving station 110, a scheduling unit 107 that determines a coding rate, a modulation degree, and power control information from CINR information from a receiving station and parallel channel quality information.

  The transmission bit sequence that is user data is subjected to error correction coding in the FEC encoding unit 101 in accordance with an instruction from the scheduling unit 107, and converted into FEC blocks for each user. This FEC block is primarily modulated by the modulation unit 102 in accordance with an instruction from the scheduling unit 107 and converted into modulated data. The modulated data is equally distributed to each parallel channel by the data distribution unit 103. The evenly distributed modulated data is subjected to power adjustment in accordance with an instruction from the scheduling unit 107 in the resource mapping unit 104 in each parallel channel, and is mapped to radio resources. The modulated data after power adjustment mapped to the radio resource is subjected to radio signal processing by the baseband / RF processing unit 105, converted to the RF band, and radiated to the space via the antenna.

  Next, the scheduling unit 107 will be described in detail with reference to FIG. Scheduling section 107 is an MCS (Modulation and Coding Scheme) level determination section 108 that determines the coding rate and modulation degree based on the CINR information of the receiving station. The coding rate information and parallel channel quality analysis section from MCS level determination section 108 The power control unit 109 determines power control information for each parallel channel based on the parallel channel quality information from 106.

2. Assumed System Hereinafter, a specific operation of the transmitter 100 will be described. The parallel channel quality analysis unit 108 acquires the quality of the parallel channel by two methods. Further, the case of the communication form as shown in FIG. 3 will be described. The cell 200 is an area covered by the transmitting station 201, and a cellular system is assumed. However, the present invention can also be applied to a spot like a wireless LAN. Here, consider a case where the transmitting station 201 transmits user data to the receiving station 202 using a parallel channel.

3. Acquisition of channel quality 3.1. Acquisition of parallel channel quality by UL-sounding When there is reversibility of the propagation path between DL (Down Link) and UL (Up Link) such as TDD (Time Division Duplex), the propagation of DL at the transmitting station by UL-sounding The path can be estimated. Further, since the transmission gain of the parallel channel depends on the MIMO decoding scheme of the receiving station, the transmitting station needs to know the MIMO decoding scheme of the receiving station. The receiving station transmits a prescribed UL-sounding packet in a short cycle or a long cycle using the UL channel. The transmitting station analyzes the received UL-sounding packet and estimates DL propagation path information (transmission path coefficient between transmission and reception).

  The parallel channel quality analysis unit 106 estimates the channel quality of the parallel channel from the estimated propagation path information and the MIMO decoding scheme of the receiving station. The channel quality is an index indicating channel reliability such as CNR (Carrier to Noise Ratio) or LLR (Log Likelihood ratio). When the period at which the receiving station transmits the UL-sounding packet is shorter than the power control period of the parallel channel defined by the system, the parallel channel quality analysis unit 106 may calculate the channel quality distribution of the parallel channel. That is, according to the UL-sounding period, the parallel channel quality analysis unit 106 calculates the following.

Short-period UL-sounding: Channel quality distribution of parallel channels Long-period UL-sounding: Instantaneous channel quality of parallel channels

3.2 Acquisition of parallel channel quality by feedback In FDD (Frequency Division Duplex), UL-sounding is impossible because there is no reversibility of the propagation path as in TDD. Therefore, it is necessary to estimate the channel quality of the parallel channel at the receiving station and feed it back to the transmitting station. The transmitting station transmits a known pilot signal, and the receiving station uses this to estimate the DL propagation path. The channel quality of the parallel channel is estimated based on the estimated propagation path and the MIMO decoding scheme of the receiving station. In addition to measuring channel quality of parallel channels, a pilot signal may always be transmitted in order to demodulate user data at a receiving station.

  Therefore, the receiving station can acquire the channel quality of the parallel channel in a very short period. Therefore, according to the feedback period required by the system, both the channel quality distribution of the parallel channels and the instantaneous channel quality can be fed back to the transmitting station. Even when the feedback period is a long period and a distribution of channel quality can be obtained, it is possible to feed back not the distribution but the average value of the channel quality. Furthermore, even when the feedback cycle is short and the transmitting station can acquire instantaneous channel quality, the transmitting station can use the channel quality distribution or the average value of the channel quality using these. Of course, this method is also effective in TDD.

4). MCS Level Determination Method The MCS level determination unit 108 determines the MCS level of user data based on the CINR information from the CINR measurement unit 110. The method of determining the MCS level is determined by the signal amplification amount P _Limit allowed by the transmitting station and the CINR information CINR _{0 of the} receiving station. A method of preparing the relational expression (1) for deriving the MCS level MCS _PC from the signal amplification amount P _Limit and the CINR information CINR ₀ can be considered in advance. As the simplest method, a discrete correspondence table is prepared in advance. A method for determining the MCS level MCS _PC in light of the correspondence table is also conceivable.

5). Parallel channel power control method 5.1. Information Source to be Used The power control unit 109 determines power control information based on the parallel channel quality information and the coding rate. There are three types of channel quality of parallel channels that can be acquired as described above.
[1]. Instantaneous channel quality,
[2]. Average channel quality,
[3]. Channel quality distribution

  Since [1] and [2] are level values of one parallel channel, both are treated similarly as channel quality levels. On the other hand, the control for [3] is different from [1] and [2] because the evaluation of the channel quality level is wide. The power control method for both [1] and [2] will be described below, and then the power control method for [3] will be described.

5.2. Power control method based on average value and instantaneous channel quality (when coding rate is not considered)
A case where power control is performed using a channel quality level will be described. The power adjustment amount of each parallel channel with respect to the channel quality level of the parallel channel is expressed by the following equation (2).

Here, N is the number of parallel channels, _FPC is a power adjustment function, C ₁ , C ₂ ,..., C _N are channel quality levels before power control, and are generally CNR (Carrier to Noise Ratio). ) Is used. In addition, λ ₁ , λ ₂ ,..., Λ _N are power control amounts from the power (P ₀ ) at the time of equal power distribution, and satisfy the following relational expression (3).

Furthermore, if g ₁ , g ₂ ,..., G _N are parallel channel gains and noise power is _PN , the quality level (CNR) of the j-th parallel channel is expressed by the following equation (4).

As an example, FIG. 4A shows the case of two parallel channels. Each parallel channel has a quality level of Ch1 quality level C ₁ 301 and Ch2 quality level C ₂ 302. The desired channel quality level 303γ _{th_req} is a channel quality level that satisfies a target BER (Bit Error Rate), and the truncation channel quality level 306γ _{th_off} is a channel quality level that is not expected to improve characteristics even by power control.

Each parallel channel is a separation channel after MIMO decoding, and is equivalent to an independent AWGN (Additive White Gaussian Noise) channel. Therefore, the desired channel quality level 303 (γ _{th_req} ) is a channel quality level that satisfies the desired BER in the AWGN channel.

In equal power distribution, the quality level difference of the parallel channels depends on the parallel channel gain g _j , and if g _j is too low, the quality improvement by the power control λ _j is very small and the channel quality level can be improved efficiently. I can't. Furthermore, in terms of BER, since poor quality channels determine the overall BER, it may be more efficient to reduce the number of parallel channels and not use poor quality channels. Therefore, the truncation channel quality level 306γ _{th_off} performs control to reduce the number of parallel channels (rank) without using the following channels.

Hereinafter, a specific example will be described in detail.
This will be described with reference to the flowcharts shown in FIGS. For simplicity, the case where there are two parallel channels is shown. First, in Stage 1, power that can be used for power control according to the gain difference of each parallel channel is calculated.

(Step 0)
Based on the number of parallel channels, the number of transmission antennas, the number of reception antennas, and the MIMO decoding method, the channel gain g _{j of} the parallel channel after MIMO decoding and the quality level C _j of the parallel channel are calculated. The calculated parallel channel quality level C _j is compared with the _cut-off channel quality level 306 (γ _{th_off} ) or the desired channel quality level 303 (γ _{th_req} ) as a threshold value to perform power control.

(Step 1)
As shown in FIG. 4E, when the Ch2 quality level 302 is equal to or lower than the truncated channel quality level 306 (γ _{th_off} ), rank reduction is performed without _performing signal transmission using the channel, and FIG. As shown in FIG. 4, signal transmission is performed only on the parallel channel Ch1. Therefore, P ₀ which is transmission power pre-assigned to the parallel channel Ch2 is added to P _{PC_max} which is the maximum value of power amplification that can be used in power control, P _{PC_max} is updated, and power control for other channels is performed. Can be used. In this example, it is shown that the power P ₀ that can be used by rank reduction is used for power control, but it is also possible to equally distribute the remaining parallel channels without using it for power control. .

(Step 2)
Next, as shown in FIG. 4 (e), when the Ch1 quality level 301 is _{equal to} or higher than the desired channel quality level 303 (γ _{th_req} ), part of the power allocated to the parallel channel Ch1.

( _{Corresponding to the} surplus channel quality 310 (C _{extra — 1} ) at the quality level) is used as a power amplification amount that can be used in power control. When there are N parallel channels, P _extra = P _{extra_1} + P _{extra_2} +... + P _{extra_N} is calculated. Here, P _extra can be treated as a power amplification amount used in power control in the same manner as P _{PC_max,} and the power amplification amount usable in the entire system is P _{PC_max} + P _extra . However, there is a condition for using the power for P _extra , and unless the condition is satisfied, the surplus power P _{extra_j} of each parallel channel cannot be subtracted from the parallel channel or used. This condition will be described later.

(Step 3)
When rank reduction is performed in Step 1, the number of streams to be separated by MIMO decoding on the receiving side is reduced, so that each parallel channel gain is changed. Therefore, when the rank is reduced, it is necessary to calculate the parallel channel gain again.

Next, in Stage 2, power calculation necessary for power control is performed in order to set all parallel channels to the desired channel quality level 303 (γ _{th_req} ).

(Step 4)
As shown in FIG. 4A, the Ch2 quality level 302 (C ₂ ) is less than the desired channel quality level 303 (γ _{th_req} ). At this time, the power control amount of each parallel channel necessary for the Ch2 quality level 302 (C ₂ ) to become the desired channel quality level 303 (γ th — _req ) is calculated by Equation (5) below.

The power P _{req_tot} required for all parallel channels having a quality level _lower than the desired channel quality level 303 (γ _{th_req} ) to have the quality of the desired channel quality level 303 (γ _{th_req} ) is _expressed by the following equation (6): ).

Further, the power control amount required by the parallel channel having a quality level _{equal to} or higher than the desired channel quality level 303 (γ _{th_req} ) is zero.

Next, the process proceeds to FIG. 6, and in Stage 3, power control is performed using information on P _{PC_max} , P _extra , and P _{req_tot} obtained in Stage 1 and Stage 2.

(Step 5)
When P _{req_tot} ≦ P _{PC_max} and all the parallel channels can satisfy the desired channel quality level 303 (γ _{th_req} ) by power control without using the power allocated to other parallel channels, only the power of P _{PC_max} is obtained. Using this, power amplification is performed so that the following equation (7) is obtained.

Here, _{CPC_j} is a quality improvement due to power control of the parallel channel Chj. FIG. 4B shows an example of the case where there are two parallel channels.

(Step 6)
If P _{req_tot} > P _{PC_max} and it is not possible to raise all parallel channels to the desired channel quality level 303 (γ _{th_req} ) only by P _{PC_max} (FIG. _4C ), P _extra is also used in addition to P _{PC_max} Perform power control. In this case, if the condition of P _extra + P _{PC — max} ≧ P _{req — tot} is satisfied, power control is also performed using P _extra . For parallel channels that do not satisfy the desired channel quality level 303 (γ th — _req ), power amplification is performed by power control so as to satisfy Equation (7).

When P _extra + P _{PC_max} > P _{req_tot} , the full power of P _extra + P _{PC_max} is not used. At this time, the power corresponding to P _extra is left, and the power of other parallel channels is used as much as possible. Therefore, when γ _{th_req} ≦ C _j , the power used for other parallel channels is reduced. FIG. 4D shows an example of the case where there are two parallel channels. The desired channel quality level 303 can be achieved by the quality improvement 304 by P _{PC_max} and the additional power 308 by P _extra .

(Step 7)
When P _extra + P _{PC_max} <P _{req_tot} , the channel quality level is not measured by power control.

5.3. Power control method based on average value and instantaneous channel quality (when coding rate is not considered)
~Abridged edition~
The method shown in 5.2 is a realistic method having a desired channel quality level 303γ _{th_req} and a truncated channel quality level 306γ _{th_off} . However, these control methods are complicated. If all the parallel channels have the same channel quality level, a simple method described below may be applied. In MIMO transmission, since the most efficient transmission is possible when the quality levels of a plurality of parallel channels are equal, the quality of the plurality of parallel channels is made equal by power control. As shown in FIG. 4 (a), when there is a quality level difference between the two parallel channels Ch1 and Ch2, the quality of each parallel channel is determined by determining the control value of both levels according to the following equation (8). It will be the same.

5.4. Power control method based on average value and instantaneous channel quality (when considering coding rate)
The above method is a power control method that does not consider error correction coding by FEC. The power control method corresponding to the coding rate of error correction coding is shown below. FIG. 7A shows the channel quality of four parallel channels. Only the Ch1 quality level 401 satisfies the desired channel quality level 405. Through these four parallel channels, the user data is FEC encoded by the FEC encoding unit 101, and a signal in which one FEC block is divided into four by the data distribution unit 103 flows. That is, one FEC block is reproduced by synthesizing user data flowing in all parallel channels. In the case of a coding rate of 1/2, theoretically, if 1/2 of all the bits constituting the FEC block is reproduced, it can be demodulated correctly. Therefore, as shown in FIG. 7B, only the parallel channel Ch2 is power-amplified to improve the desired channel quality level 405. Since only the Ch1 quality level satisfies the desired channel quality level, ½ of the total satisfies the desired channel quality level 405 and can be demodulated with the desired quality. The other parallel channel control methods are the same as in Section 5.2.

5.5. Power Control Method Based on Channel Quality Distribution Sections 5.2 to 5.4 mentioned only the channel quality level (instantaneous value or average value) of parallel channels. When the observation time of the channel matrix is long and the control interval of power control is also long, it is possible to obtain the channel quality distribution of the parallel channels.

When the average value is _{equal to} or _smaller than γ _{th_off} FIGS. 8A and 8C show an example of two parallel channels. Since the average value of the distribution of any parallel channel exceeds the truncated channel quality level 507 (γ th — _off ), rank reduction is not performed. However, if there is a parallel channel that takes an average value below the truncated channel quality level 507 (γ th — _off ), 5.2. Perform rank reduction in the same way as clauses. The above example is control using the average value of the distribution of parallel channels as a parameter, but the ratio (η _{bellow_off} ) that is equal to or less than γ _{th_off} in the distribution of parallel channels may be used as a parameter. A method of reducing the rank is also conceivable when η _{th_off} <η _{bellow_off} with _respect to the threshold η _{th_off} defined by the system.

When the average value is _{equal to} or greater than γ _{th_off,} the variance σ ₂ of the parallel channel Ch2 quality distribution 502 in FIG. 8A is large, and there are several samples that have a desired channel quality level 503 (γ _{th_req} ) or greater even before power control. As shown in FIG. 8B, when the quality of the parallel channel Ch2 is improved by the power control, the ratio of the desired channel quality level 503 or higher increases. Therefore, even when the average value of the quality distribution does not exceed γ _{th_req} , a certain quality improvement by power control is desired. Therefore, the presence / absence of power control is determined by the ratio η _{over_req} exceeding γ _{th_req} after power control. Power control is performed when η _{over_req} ≧ η _{th_req} , and power control is not performed otherwise. Here, η _{th_req} is a desired value of a ratio that becomes γ _{th_req} or more after power control. For example, if the ratio 508 exceeding the desired channel quality level in FIG. 8B is η _{over_req} ≧ η _{th_req} , power control is performed. On the other hand, when the variance σ ₂ of the parallel channel Ch2 quality distribution 504 is small as shown in FIG. 8C, even when the same power control as that shown in FIG. 7B is performed as shown in FIG. There may be no ratio exceeding γ _{th_req} . In such a case, power control is not performed.

  The above description is a case where the FEC coding rate is not taken into account, but it is also possible to combine it with power control taking into account the coding rate as shown in section 5.4.

100 transmitter, 101 FEC encoding unit, 102 modulation unit, 103 data distribution unit, 104 resource mapping unit, 105 baseband / RF processing unit, 106 parallel channel quality analysis unit, 107 scheduling unit, 108 MCS level determination unit, 109 power Control unit, 110 CINR measurement unit, 200 cells, 201 transmitting station, 202 receiving station, 301 Ch1 quality level C ₁ , 302 Ch2 quality level C ₂ , 303 desired channel quality level, 304 quality improvement, 305 desired channel quality level, 306 truncation channel quality level, 307 power reduction, 308 power addition, 401 Ch1 quality level, 402 Ch2 quality level, 403 Ch3 quality level, 404 Ch4 quality level, 405 desired channel quality level, 406 truncation channel quality Bell, 501 Parallel channel Ch1 quality distribution, 502 Parallel channel Ch2 quality distribution, 503 Desired channel quality level, 504 Parallel channel Ch2 quality distribution, 505 Parallel channel Ch2 quality distribution after power control, 506 Parallel channel Ch2 quality distribution, 507 Channel quality level, 508 The percentage that exceeds the desired channel quality level.

Claims (6)

A power control method for parallel channels in a transmission method using a vertical encoding method in which one FEC block data is distributed and transmitted to a plurality of parallel channels.
A power control method for a parallel channel, wherein power control is performed according to a gain difference between the parallel channels. The power control method for parallel channels according to claim 1,
A power control method for parallel channels, wherein power amplification is performed according to the channel quality level of each parallel channel. The power control method for parallel channels according to claim 2,
A power control method for parallel channels, wherein power control of each parallel channel is performed using a desired channel quality level and a truncated channel quality level as thresholds. The power control method for parallel channels according to claim 1,
A power control method for parallel channels, wherein power control is performed according to the channel quality distribution of each parallel channel. The power control method for parallel channels according to claim 4,
A power control method for parallel channels, wherein power control is performed in accordance with a distribution ratio exceeding a desired channel quality level. The power control method for parallel channels according to claim 4,
A power control method for a parallel channel, characterized in that power control is performed according to a distribution ratio that becomes a cutoff channel quality level or less.

Images