This invention relates to a frequency division communication system for multiplexing uplink and downlink transmission using a plurality of frequencies. In particular, this invention relates to an Orthogonal Frequency Division Multiplexing (OFDM) method in which the relation of frequencies is orthogonal so as to enable effective utilization of the frequencies used.
In the past, communication systems performing radio frequency multiplexing of uplink and downlink transmissions have adopted the Frequency Division Duplex (FDD) method or the Time Division Duplex (TDD) method.
Further, in third-generation mobile telephone systems, TDD methods adopted in TDS (Time Division Synchronous)-CDMA (Code Division Multiple Access) and other systems have enabled effective use, compared with FDD methods used in W-CDMA (Wideband Code Division Multiple Access) methods, of frequencies through the use of the same frequency band for uplink and downlink channels.
Moreover, there is the advantage that, by modifying the ratio of time allocated to uplink and downlink transmission, communication speeds can be changed flexibly, and asymmetric-rate data communication services can be efficiently provided.
Further, by using the same frequency for uplink and downlink transmission, it is expected that the uplink/downlink correlation will be high, and so the uplink channel can be used at the base station to estimate the downlink channel state. Or, the downlink channel can be used at the mobile station to estimate the uplink channel state.
For this reason, methods which in FDD systems require channel information feedback (for example, adaptive modulation, transmission diversity, and similar) can be performed without feedback in TDD systems.
However, in TDD systems it is necessary to rapidly switch between uplink and downlink transmission in order to prevent uplink/downlink interference. This results in more complex configurations for both the receiver and the transmitter. Further, because in TDD systems uplink/downlink allocation is limited to the time direction only, there is the possibility of more flexible allocation in the frequency direction.
One example of such a technique is proposed in Japanese Patent Laid-open No. 11-275036. In the CDMA/TDD method, signals having a TDMA structure are used, and by performing broadcast channel transmission and reception only in the last downlink slot of subframes, various services can be flexibly accommodated.
Hence an object of this invention is to provide an orthogonal frequency division communication system which, while maintaining advantages similar to those of TDD (Time Division Duplex) systems, also enables flexible modification of the uplink/downlink allocation ratio.
A frequency division communication system which attains the above object, in a first aspect, has a base station and a mobile station connected by an uplink and a downlink, and is characterized in that two frequencies are allocated to the uplink and downlink.
A frequency division communication system which attains the above object, in a second aspect, has a base station and a plurality of mobile stations connected by an uplink and a downlink, and is characterized in that a plurality of orthogonal frequencies are allocated, on the frequency axis and on the time axis, to the uplink and downlink, and to the plurality of mobile stations.
A frequency division communication system which attains the above object, in a third aspect, is the system of the second aspect, characterized in that the base station has a traffic monitoring portion which monitors the uplink/downlink traffic ratio, and in that allocation on the frequency axis and on the time axis of the plurality of frequencies is determined according to the traffic ratio monitored by the traffic monitoring portion.
A frequency division communication system which attains the above object, in a fourth aspect, is the system of the first or second aspect, characterized in that the frequencies allocated to the uplink and downlink are in close proximity, so that the frequency difference is such that the correlation value between uplink and downlink is high.
A frequency division communication system which attains the above object, in a fifth aspect, is the system of the fourth aspect, characterized in that the base station has a SIR measurement portion which measures the signal-to-noise ratio (SIR value) for each of the plurality of frequencies for the uplink; a modulation method decision portion which decides the modulation method according to the measurement values of the SIR measurement portion; and a modulation portion which applies the modulation method decided by the modulation method decision portion to the respective plurality of frequencies.
A frequency-division communication system which attains the above object, in a sixth aspect, is the system of the fourth aspect, characterized in that the base station has a SIR measurement portion which measures the signal-to-noise ratio (SIR value) for each of the plurality of frequencies for the uplink, and in that the SIR measurement portion determines an average value of measurement values for frequencies allocated to each of the plurality of mobile stations, and has a modulation method decision portion which decides the modulation method corresponding to the average value; a modulation method decision portion which decides the modulation method for each mobile station, according to the average value of measurement values determined by the SIR measurement portion; and a modulation portion which applies the modulation method decided by the modulation method decision portion to each of the plurality of frequencies allocated to the mobile stations.
Characteristics of this invention will become more clear through the aspects explained below, referring to the drawings.
By means of this invention, multichannel uplink and downlink transmission is performed using a plurality of frequencies. For example, the subcarriers in Orthogonal Frequency Division Multiplex (OFDM) transmission are flexibly allocated to uplink and downlink transmission. By this means, while maintaining advantages similar to those of TDD (Time Division Duplex) methods, the ratio of allocation to uplink and downlink transmission can be modified flexibly.
Below, embodiments of the invention are explained referring to the drawings. The embodiments explained below are intended to facilitate understanding of the invention, and the technical scope of the invention is not limited to these embodiments.
In
Signals converted into the time domain are converted into serial signals by the P/S converter 2, and then the guard interval (GI) insertion circuit 3 inserts guard intervals (GIs) at each symbol.
Here, as indicated in the frame structure of
A baseband signal with guard intervals (GIs) added is converted to an analog signal by the D/A converter 4, and after rolloff by a low-pass filter 5, is input to the modulator 6.
The modulator 6 modulates a carrier wave 7 at a radio frequency with the analog signal. The radio frequency signal from the modulator 6 is bandwidth-limited by the bandpass filter 8, then amplified by the power amplifier 9, and is transmitted from the antenna 11 via a circulator 10.
The signal output from the antenna 11 passes through a fading propagation path and is received by the antenna 11 of the other-side receiver. For convenience, other-side reception operation is explained using the transceiver configuration of
The received radio frequency signal is converted into a baseband signal by the bandpass filter 12, linear amplifier 13 and demodulator 14.
Then, noise is reduced by the low-pass filter 15, and the A/D converter 16 performs conversion to digital signals. In addition, synchronization is obtained, guard intervals (GIs) are removed from the baseband signal in the guard interval removal circuit 17, and data for FFT processing is extracted for each symbol. Then, the extracted data for FFT processing is converted into parallel signals by the S/P converter 18, and a Fast Fourier Transform (FFT) is performed by the FFT circuit 19 to convert signals into frequency-domain subcarrier signals.
In a transceiver with the above configuration, as indicated by the frame structure shown, when a plurality of orthogonal subcarrier frequencies arranged on the frequency axis are divided and duplex transmission is performed, flexible allocation to uplink and to downlink transmission is possible.
That is, subcarriers for transmission data to be subject to Inverse Fast Fourier Transform (IFFT) processing are flexibly allocated to the uplink and to the downlink.
By this means, high-speed uplink/downlink switching is no longer needed. Further, on the time axis also, time division allocation between the uplink and downlink is performed. Hence the uplink/downlink allocation ratio can be modified more flexibly than in conventional time division duplexing (TDD).
That is, in
On the other hand, on the receiving side only the output of subcarriers allocated on the receiving side after FFT by the Fast Fourier Transform (FFT) portion 19 is used.
By providing a circulator 10 at the antenna 11, leaking of transmission signals into the receiver can be suppressed to some extent. However, even when there is some leaking, orthogonal components are removed by FFT on the receiving side.
In
In this way, allocation for each mobile station is possible taking the asymmetry of traffic into account. Further, allocation is performed so as to make the interval between uplink and downlink allocated frequencies as small as possible. As a result, uplink signal channel estimation can be performed at the base station (using pilot symbols embedded in symbol periods at the beginning of frames; see
This means that the base station and each of the mobile stations can share uplink and downlink channel information without signaling each other. Further, through effective utilization of frequencies by means of Orthogonal Frequency Division Multiplexing (OFDM), the efficiency of frequency use is equivalent to that of Time Division Duplex (TDD) methods.
Commands are sent to each of the mobile stations to advance the transmission timing when the downlink transmission signal frame boundary (TD) lags, and to delay the transmission timing when it leads. By this means, it is possible to simultaneously maintain the downlink transmission timing a0 and the uplink reception timing b1 and c1 for the mobile stations #1 and #2.
That is, downlink reception signals and uplink transmission signals in mobile station #1 are shown in
Further, in
Further, when the effect of delay spreading due to multipath is smaller than the front guard interval length (TGI
Moreover, with respect to delay spreading of downlink reception signals, similarly to uplink signals, when the delay spreading is smaller than the front guard interval length (TGI
Here, when applying this invention, the base station decides the allocation of each subcarrier to mobile stations and uplink/downlink allocation taking into account reception quality estimate values from each mobile station and the traffic asymmetry for each mobile station.
In a transmission-side mobile station, transmission symbols are assigned to allocated subcarriers, and “0”s are allocated to other frequencies, and inverse fast Fourier transform processing is performed by the IFFT circuit 1. On the other hand, in a reception-side mobile station, after fast Fourier transform processing by the FFT circuit 19, only allocated subcarriers are used in subsequent signal processing.
Because a mobile station must be notified in advance of the allocated subcarriers, a downlink control channel or similar is prepared. When a dedicated subcarrier is used as a control channel, if for example control data is dc and other individual data is dd, then by performing orthogonal modulation of these using a subcarrier at frequency f0, the result is as expressed by equation (1).
[equation]
u
0(t)=(dc+jdd)·exp(j2πf0t)) (1)
At the mobile station, by decoding the above dc, control information from the base station can be received.
Encoding and modulation processing, handled by the encoder 30 and modulator 31, is performed on downlink transmission data to each of the plurality of mobile stations 1 to N, and the results are input to the subcarrier allocation circuit 32.
On the other hand, on the receiving side of the base station, reception signals are subjected to fast Fourier transform processing by the FFT circuit 19 and are input to the subcarrier selection circuit 34. The subcarrier allocation circuit 32 and subcarrier selection circuit 34 are controlled by the subcarrier allocation/control portion 33.
Operation of the subcarrier allocation/control portion 33 is explained referring to
The subcarrier allocation/control portion 33 takes as input the uplink/downlink traffic ratio monitored by the traffic ratio monitoring circuit 302 as traffic information, and controls the subcarrier allocation circuit 32 and subcarrier selection portion 34 such that channel allocation is always optimal.
That is, based on the traffic information, the subcarrier allocation/control portion 33 does not change the channel allocation if the computed ratio is the same as the previous value (“no” in step S2).
If the ratio is different from the previous value (“yes” in step S2), a channel allocation pattern is decided according to the example of the table shown in
Also, the subcarrier selection portion 34 is controlled such that addition and deletion of selected subcarriers is performed based on this information (step S5).
Here, methods are known for improving the data transfer rate through adaptive modulation, in which the modulation method and coding rate are changed according to the radio environment (received signal to noise ratio (SIR:Signal to Interface power Ratio)).
For example, at the radio base station, QPSK is used when the radio reception state is poor (when the SIR is low), and the 16 QAM-modulation method is used when the reception state is good. There may be cases in which the coding rate is changed as well as the modulation method. That is, selection may be performed automatically according to the reception environment so that a code with powerful error correction performance is used when the reception state is poor, and a code with weaker error correction performance is used when the reception state is good.
In this way, the combination of modulation method and coding rate is optimized for the state of the radio environment, and as a result the data transfer rate can be improved.
This invention can be combined with such adaptive modulation methods.
In
On the transmitting side, “0”s are transmitted by the transmission selection portion 400 prior to modulation using subcarriers selected by the subcarrier selection portion 34.
On the receiving side, for each of the selected subcarriers, demodulation by the demodulator 35 and channel estimation by the channel estimation portion 403 are performed, and the SIR measurement portion 404 computes SIR values.
Then, computed SIR values are compared with thresholds by the SIR value comparison portion 406, and the modulation method is decided in the modulation method decision portion 407, according to a table prepared in advance.
On the transmitting side, on the other hand, the method described in the previous embodiment is again used to allocate and select subcarrier frequencies by the subcarrier allocation portion 32 and transmission selection portion 400, and multilevel modulation is performed by the multilevel modulation portion 401 for parallel-converted bit series for each user (mobile station).
Here,
In
Modulation by digital signals d0 input from the transmission selection portion 400 is performed by each of the plurality of multilevel modulation circuits of the multilevel modulation portion 401 according to these decisions, using the modulation methods thus decided.
As shown in
In order to explain other examples of application of the invention, closed-loop transmission diversity in W-CDMA, which is a third-generation mobile communication system, is explained as one example of transmission diversity.
In W-CDMA, a method is adopted in which two transmission antennas are used.
On the receiving side of a mobile station, the pilot patterns P1 and P2 are received by the reception antenna AC, and the correlation between known pilot patterns and the received pilot signals are computed by the control quantity calculation portion 501.
Based on the computed correlations, channel impulse response vectors h1 and h2 from each of the transmission antennas AA and AB of the base station to the mobile station reception antenna AC can be estimated.
Using the channel estimation values, the amplitude and phase control vector (weight vector) for each transmission antenna of the base station which maximizes the power PW is computed:
[equation]
w=[w
1
, w
2]T (2)
and by quantizing and multiplexing the result as feedback information with channel signals in the multiplexing circuit 502, the information is transmitted from the transmission antenna AD to the base station.
However, it is not necessary to transmit both of the values w1 and W2 in the phase control vector (weight vector); when the vector is determined such that w1=1, it is sufficient to transmit only the value of w2.
Here, the power PW is expressed by equation (3).
[equation]
PW=w
H
H
H
Hw (3)
[equation]
H=[h
1
,h
2] (4)
In equation (3), h1, and h2 from the antennas AA and AB respectively, form the channel impulse response vector.
If the impulse response length is L, then h1 is expressed by the following equation (5).
[equation]
h
i
=[h
i1
,h
i2, . . . , hiL]T (5)
At the time of a soft handover, in place of equation (3), the control vector which maximizes the power as given by equation (6) is computed.
[equation]
PW=w
H (H1HH1+H2HH2+. . . )w (6)
Here Hk is the channel impulse response for signals from the kth base station.
In W-CDMA, two methods are stipulated, which are mode 1 in which weighting factors w2 are quantized to 1 bit, and mode 2 in which quantization is to 4 bits.
In mode 1, control is executed by transmitting 1 bit of feedback information for each slot, so that while control speed is fast, quantization is coarse, and so accurate control is not possible.
On the other hand, in mode 2 control employs 4 bits of information, so that more precise control is possible; on the other hand, 1 bit is transmitted for each slot, with 1 word of feedback information transmitted over 4 slots. Hence when the fading frequency is high, the fading cannot be tracked, and characteristics are degraded.
Thus when the uplink channel signal transmission rate to transmit feedback information is limited, there is a tradeoff between control precision and fading tracking response.
In the W-CDMA Release 99 specification, no consideration is made for cases in which more than two transmission antennas are used in order to avoid declines in uplink channel transmission efficiency due to feedback information transmission. However, expansion to three or more antennas is possible in order to increase the amount of feedback information and to allow a decline in the update rate.
When there are N transmission antennas, different transmission antennas are used to transmit N mutually orthogonal pilot signals P1 (t), P2 (t), . . . , PN (t) at the radio base station.
These pilot signals are related as indicated by equation (7).
[equation]
∫Pi(t)Pj(t)dt=0 (i≠j) (7)
In the above equation (6), each of the pilot signals receives the amplitude and phase changes due to fading, and the composite of these signals is input to the mobile station reception antenna AC.
In the mobile station receiver, by having the control quantity calculation portion 501 determine the correlations with P1 (t), P2 (t), . . . , PN (t) of the received pilot signals, channel impulse response vectors h1, h2, . . . , hN for each of the pilot signals can be estimated.
Using these channel impulse response vectors, the amplitude and phase control vectors (weight vectors) for each transmission antenna of the base station,
[equation]
w=[w
1
, w
2
, . . . w
N]
T
which maximize the power PW, are computed as in equation (8); these are quantized and multiplexed with uplink channel signals as feedback information, and transmitted from the antenna AD to the base station.
However, in this case also, when values are determined with w1=1, it is sufficient to transmit the values of w2, W3 , . . . , wN.
[equation]
PW=w
H
H
H
Hw (8)
[equation]
H=[h
1
,h
2
, . . . h
N] (9)
On the base station side, the feedback information is received by the reception antenna AE, and is extracted by the feedback information extraction circuit 503. The feedback information extraction circuit 503 controls the amplitude/phase control circuit 504 based on the extracted feedback information.
Thus in closed-loop transmission-diversity in a W-CDMA system, a configuration is employed in which the downlink power is sent from the mobile station to the base station as feedback information.
Through application of this invention, by estimating the propagation path state for downlink transmission from uplink transmission in a coherent band, feedback from the mobile equipment can be omitted.
The base station configuration shown in
In this base station, signals for a certain user are extracted from signals received by the two antennas 11a and 11b. In the embodiment shown in
For example, in propagation path estimation for subcarrier f0 in downlink transmission, the mobile equipment 1 uses channel estimate values for a subcarrier orthogonal to the subcarrier f0, that is, the adjacent subcarrier f1 used in uplink transmission.
That is, in
These estimation values are input to the phase/amplitude comparison portion 410, and as shown in
This embodiment is an example of a case in which frequency diversity is used, but similarly to the embodiment shown in
In
In the subcarrier selection portion 34, the subcarriers f1 and fn+1 are selected, demodulation is performed by the demodulator 35, and channel estimation is performed by the channel estimation portion 403. Then, amplitude and phase comparisons are performed by the phase/amplitude comparison portion 410 for the channel estimation values, and based on the comparison results, the weight vector is calculated by the complex weight generation portion 411 as indicated by equation (2) such that the power PW in the above equation (3) is maximum. Then, the calculated weight vector is multiplied by the multiplier 413, and the downlink power is controlled.
In the embodiment of
Aspects of this invention have been explained using OFDM, but there is no need for an orthogonal relation between the uplink and downlink subcarrier frequencies. Application to a simple FDM system is also possible.
As explained above, by performing multiplexing of uplink and downlink transmissions using orthogonal frequencies, flexible allocation to uplink and downlink transmissions is possible. As a result, a system can be provided which, while maintaining advantages similar to those of TDD (Time Division Duplex) communication, enables flexible modification of the uplink/downlink allocation ratio.
This application is a continuation of international application PCT/JP2005/000399, filed on Jan. 14, 2005.
Number | Date | Country | |
---|---|---|---|
Parent | PCT/JP05/00399 | Jan 2005 | US |
Child | 11826297 | US |