GB2364222A

GB2364222A - Receiver for OFDM signal with adaptive filtering to remove inter-carrier interference

Info

Publication number: GB2364222A
Application number: GB0100501A
Authority: GB
Inventors: Jun Ido; Hiroyuki Nakayama
Original assignee: Mitsubishi Electric Corp
Current assignee: Mitsubishi Electric Corp
Priority date: 2000-04-25
Filing date: 2001-01-09
Publication date: 2002-01-16
Anticipated expiration: 2021-01-09
Also published as: GB2364222B; GB0100501D0; JP2001308820A; CN1321014A

Abstract

To obtain an orthogonal frequency-division multiplexed signal receiver with improved receiving performance, by removing the inter-carrier interference component present in the Fourier transform 1 output, thus reducing the error probability in the recovered data. The frequency-domain signal output from the Fourier transform is filtered 3 to remove the inter-carrier interference component, and coefficients capable of removing the inter-carrier interference component are calculated 4 dynamically by an adaptive algorithm. The filtered output is then demodulated.

Description

2364222 RECEIVER FOR ORTHOGONAL FREQUENCY-DIVISION MULTIPLEXED SIGNAL

BACKGROUND OF THE INVENTION

This invention relates to a receiver for an orthogonal frequency-division multiplexed signal.

FIG 15, which is taken as an example from an article by Kageyama, Nishimura, and Ikeda entitled "Demodulation Technologies for Terrestrial Digital Broadcasting" in The Journal of the Institute of Image Information and Television, Vol 52,'No 11, pp 1571-1572, Nov 1998, is a block diagram representing a conventional demodulator for an orthogonal frequency-division multiplexed signal In FIG 15, reference numeral 10 denotes a tuner unit that receives a signal transmitted by an orthogonal frequency-division multiplexing system; reference numeral 11 denotes a synchronous recovery unit that receives the outputs rr, ri of the tuner unit and the outputs of an FFT unit; reference numeral 1 denotes the FFT unit, which receives the output of the synchronous recovery unit 11; reference numeral 2 denotes a demodulation unit that receives the outputs of the FFT unit 1; the outputs I' and Q' of the demodulation unit 2 are the recovered data, or recovered error-correction encoded data, of the transmitted digital data RRF is the signal transmitted by the orthogonal frequency-division multiplexing system; rr and ri are the real part and imaginary part, respectively, of the signal transmitted by the orthogonal frequency-division multiplexing system converted in frequency to a predetermined frequency band and expressed in complex-valued signal notation; Sr and Si are the real part and imaginary part, respectively, of the signal after frequency recovery and clock recovery by the synchronous recovery unit 11, expressed in complex-valued signal notation; and I' and Q' are the real part and imaginary part, respectively, of the output signal of the demodulation unit 2, expressed in complex-valued signal notation.

Next, the operation will be described The tuner unit receives the signal transmitted by the orthogonal frequency-division multiplexing system, converts it to a predetermined frequency band, and outputs the frequencyconverted signal The synchronous recovery unit 11 receives the outputs rr, ri of the tuner unit and the outputs I, Q of the FFT unit 1, performs synchronous recovery of the frequency of the orthogonal frequency-division multiplexed signal and clock recovery, and outputs the frequencyrecovered and clock-recovered signals Sr and Si (referred to below as time-domain signals) The FFT unit 1 receives the time-domain signals output from the synchronous recovery unit 11, and outputs the signals I and Q (referred to below as frequency-domain signals) after a Fourier-transform process with a predetermined number of points The demodulation unit 2 receives the frequency-domain signals, and demodulates each subcarrier by a demodulation method corresponding to the modulation method thereof The output of the demodulation unit 2 is the recovered data, or recovered error-correction encoded data, of the transmitted digital data The transmitted data are recovered from this output.

A problem in the conventional orthogonal frequency- division multiplexed signal receiver was that, because it was configured for demodulation of the unaltered frequency- domain signals output from the Fourier transform, there was a remaining component of interference between subcarriers (referred to below as inter-carrier interference), due to frequency error which could not be completely removed by the synchronous recovery unit, increasing the probability of error in the recovered data.

This invention is directed toward a solution to the above problem, with the object of removing the inter-carrier interference component due to frequency error that could not be completely removed by the synchronous recovery unit, thus obtaining an orthogonal frequency-division multiplexed signal receiver having improved receiving performance.

SUMMARY OF THE INVENTION

The orthogonal frequency-division multiplexed signal receiver according to the invention recovers transmitted data from an orthogonal frequency-division multiplexed signal, has Fourier transform means for transforming the received time-domain signal into a frequency-domain signal, coefficient calculation means, filter means receiving the output of the Fourier-transform means and the output of the coefficient calculation means and performing filtering with filter coefficients output from the coefficient calculation means, and demodulation means receiving the output of the filtering means, performing demodulation by a demodulation method corresponding to the modulation method of each subcarrier The coefficient calculation means receives the output of the filter means, the output of the Fourier transform means, a step-size value, and a reference value; updates the filter coefficients dynamically and adaptively, so as to minimize the inter-carrier interference in the signal output from the filter means; and outputs the updated filter coefficients.

With the above arrangement, inter-carrier interference can be reduced when the signal is demodulated, and the probability of errors in the recovered data can be reduced, because filtering is performed on the frequency-domain signal output from the Fourier transform, and filter coefficients capable of removing the inter-carrier interference component are calculated by an adaptive algorithm.

Another advantage is that changes in the degree of inter-carrier interference can be followed adaptively, because an adaptive algorithm is used as the algorithm for optimizing the filter coefficients.

The orthogonal frequency-division multiplexed signal receiver may further includes a pilot timing signal generating means that outputs a timing signal indicating whether or not the output signal of the filter means is a pilot carrier, and a step-size control means that receives the output of the pilot timing signal generating means and outputs a step-size parameter according to the timing signal.

The coefficient calculation means receives the output of the step-size control circuit as the step-size value, and updates the filter coefficients based thereupon.

With the above arrangement, the filter coefficients when inter-carrier interference is removed can be determined rapidly, because the step-size parameter used in coefficient updating is controlled according to whether the subcarrier is a pilot carrier or not.

Also, an inter-carrier interference canceler of higher accuracy can be realized, because it is configured to update the coefficients on the basis of pilot-carrier information.

The orthogonal frequency-division multiplexed signal receiver may further includes a power calculation means that calculates and outputs the signal power of each subcarrier and a step-size control means that receives the output of the power calculation means, controls a step-size parameter based thereupon, and outputs the step-size parameter The coefficient calculation means receives the output of the step-size control means as the step-size value, and updates the filter coefficients based thereupon.

With the above arrangement, inter-carrier interference can be removed adaptively even from orthogonal frequency- division multiplexed signals that have been subjected to multipath effects, because the step-size parameter used in coefficient updating can be controlled according to the power of the subcarriers.

The power calculation means may be provided to receive the output of the Fourier transform means, and calculate therefrom the signal power of each subcarrier.

With the above arrangement, power can be calculated quickly, because signal power is calculated from the output of the Fourier transform means, so the filter coefficients can be updated promptly.

It may be so arranged that, where the demodulation means calculates a quantity equivalent to the average power of each subcarrier while performing demodulation on the basis of channel characteristics estimated for each subcarrier, the demodulation means also functions as the power calculation means; the step-size control means receives information output from the demodulation means expressing the average power of each subcarrier, and controls and outputs the step-size parameter on the basis thereof.

With the above arrangement, it is not necessary to add a new circuit for the calculation of power, because the power calculation function already provided in the demodulation means can be used Also, dynamic updating of the filter coefficients can be performed in a stable manner, regardless of changes in the signal level of the transmitted data, because the average power is used.

The orthogonal frequency-division multiplexed signal receiver may further includes a power calculation means that calculates and outputs the signal power of each subcarrier, and a reference control means receiving the output of the power calculation means, and controlling a reference-level value based thereupon The coefficient calculation means receives the output of the reference control means as the reference value, and updates the filter coefficients based thereupon.

With the above arrangement, inter-carrier interference can be removed adaptively even from orthogonal frequency- division multiplexed signals that have been subjected to multipath effects, because the reference-level signal used in coefficient updating can be controlled according to the power of the subcarriers.

It may be so arranged that, where the demodulation means calculates a quantity equivalent to the average power of each carrier while performing demodulation on the basis of channel characteristics estimated for each subcarrier, the demodulation means also functions as the power calculation means; the step-size control means receives information output from the demodulation means expressing the average power of each subcarrier, and controls and outputs the reference value on the basis thereof.

BRIEF DESCRIPTION OF THE DRAWINGS

In the attached drawings:

FIG 1 is a block diagram illustrating a receiver according to a first embodiment of the invention; FIG 2 is a block diagram illustrating the configuration of the filter 3 in the receiver of the first embodiment; FIG 3 is a block diagram illustrating the configuration of the coefficient calculation unit 4 in the receiver of the first embodiment; FIG 4 is a block diagram illustrating the configuration of a tap-coefficient updating unit 400 in the receiver of the first embodiment; FIG 5 is a block diagram illustrating a receiver according to a second embodiment of the invention; FIG 6 is a block diagram illustrating a receiver according to a third embodiment of the invention; FIG 7 is a block diagram illustrating one configuration of the step-size control unit 8 in the receiver of the third embodiment; FIG 8 is a block diagram illustrating another configuration of the step-size control unit 8 in the receiver of the third embodiment; FIG 9 is a block diagram illustrating a receiver according to a fourth embodiment of the invention; FIG 10 is a block diagram illustrating the configuration of the synchronously-modulated-signal demodulation unit 20 in the receiver of the fourth embodiment; FIG 11 is a block diagram illustrating a receiver according to a fifth embodiment of the invention; FIG 12 is a block diagram illustrating one configuration of the reference control unit 9 in the receiver of the fifth embodiment; FIG 13 is a block diagram illustrating another configuration of the reference control unit 9 in the receiver of the fifth embodiment; FIG 14 is a block diagram illustrating a receiver according to a sixth embodiment of the invention; and FIG 15 is a block diagram illustrating a conventional receiver.

DETAILED DESCRIPTION OF THE INVENTION

Embodiments of the invention will now be described with reference to the attached drawings, in which like parts are indicated by like reference characters.

First Embodiment FIG 1 is a block diagram illustrating an orthogonal frequency-division multiplexed signal receiver according to a first embodiment of the invention In FIG 1, reference numeral 1 denotes an FFT unit that receives time-domain signals S and S resulting from frequency recovery and clock recovery of a signal, transmitted by an orthogonal frequency-division multiplexing system, that has been converted in frequency to a predetermined frequency band; reference numeral 2 denotes a demodulation unit that receives the output of a filter unit; the filter unit, denoted by reference numeral 3, receives frequency-domain signals I, Q output from the FFT unit 1, and receives the output of a coefficient calculation unit; the coefficient calculation unit, denoted by reference numeral 4, receives the output of the filter unit 3, the output of the FFT unit 1, a step-size value, and a reference value; the output of the demodulation unit 2 is the recovered data, or recovered error-correction encoded data, of the transmitted digital data Also in FIG 1, Cr and C 1 are the real parts and imaginary parts, respectively, of the output signals of the coefficient calculation unit 4, expressed in complex-valued signal notation; R is the step-size parameter input to the coefficient calculation unit 4; R is the reference-level signal input to the coefficient calculation unit 4; and I", Q" are the real part and imaginary part, respectively, of the output signal of the filter unit 3, expressed in complex-valued signal notation.

Next, the operation will be described The FFT unit 1 receives the time-domain signals Sr and Si, and outputs them after Fourier transform processing with a predetermined number of points The filter unit 3 performs filtering on the frequency-domain signals I, Q output from the FFT unit 1, using the filter coefficients Cr Ci output from the coefficient calculation unit 4 The outputs I", Q" of the filter unit 3 are signals from which the inter-carrier interference component remaining in the frequency-domain signals I, Q has been removed The coefficient calculation unit 4 receives the output of the filter unit 3, the output of the FFT unit 1, the step-size value, and the reference value, and dynamically updates the filter coefficients so as to minimize the inter-carrier interference component in the signals output from the filter unit 3 The demodulation unit 2 receives the outputs of the filter unit 3, and demodulates each subcarrier by a demodulation method corresponding to the modulation system thereof The output of the demodulation unit 2 is the recovered data, or recovered error-correction encoded data, of the transmitted digital data The transmitted data are recovered from this output.

Next, as an example of the structure of the filter unit 3, FIG 2 shows a block diagram of an M-tap complex filter.

In FIG 2, reference numeral 300 denotes a delay unit that imparts a certain delay to the input signal; reference numeral 301 denotes a multiplier unit that multiplies two input signals and outputs their product; reference numeral 302 denotes a first signal summing unit that receives signals output from a plurality of multiplier units 301, the inputs of which are the signal I input to the filter unit 3, or this signal as delayed by a delay unit 300, and the filter coefficients Cr input to the filter unit 3; reference numeral 303 denotes a second signal summing unit that receives signals output from a plurality of multiplier units 301, the inputs of which are the signal I input to the filter unit 3, or this signal as delayed by a delay unit 300, and the filter coefficients Ci input to the filter unit 3; reference numeral 304 denotes a third signal summing unit that receives signals output from a plurality of multiplier units 301, the inputs of which are the signal Q input to the filter unit 3, or this signal as delayed by a delay unit 300, and the filter coefficients Cr input to the filter unit 3; reference numeral 305 denotes a fourth signal summing unit that receives signals output from a plurality of multiplier units 301, the inputs of which are the signal Q input to the filter unit 3, or this signal as delayed by the delay units 300, and the filter coefficients C input to the filter unit 3; reference numeral 306 denotes a signal subtracting unit that receives the outputs of the first signal summing unit 302 and fourth signal summing unit 305; and reference numeral 307 denotes a signal summing unit that receives the outputs of the second signal summing unit 303 and third signal summing unit 304 Also in FIG 2, Cr,m and Ci m are the coefficient values at the m-th tap of the coefficients Cr and C,, respectively (m = 0, 1,, M 1).

Next, the operation of the filter unit 3 shown in FIG.

2 will be described Each delay unit 300 imparts a delay equivalent to the reciprocal of the FFT sampling rate in the FFT unit 1 to the frequency-domain signal I or Q, or a delayed I or Q signal The multiplier units 301 multiply the output signals of the delay units 300 and the filter coefficients corresponding to the taps, and output the products The first signal summing unit 302 receives signals in which the signal I input to the filter unit 3, or this signal as delayed by a delay unit 300, is multiplied by the filter coefficients Cr input to the filter unit 3, adds them, and outputs their sum The second signal summing unit 303 receives signals in which the signal I input to the filter unit 3, or this signal as delayed by a delay unit 300, is multiplied by the filter coefficients C input to the filter unit 3, adds them, and outputs their sum The third signal summing unit 304 receives signals in which the signal Q input to the filter unit 3, or this signal as delayed by a delay unit 300, is multiplied by the filter coefficients Cr input to the filter unit 3, adds them, and outputs their sum.

The fourth signal summing unit 305 receives signals in which the signal Q input to the filter unit 3, or this signal as delayed by a delay unit 300, is multiplied by the filter coefficients Cj input to the filter unit 3, adds them, and outputs their sum The signal subtracting unit 306 subtracts the output of the fourth signal summing unit 305 from the output of the first signal summing unit 302, and outputs the difference The signal summing unit 307 adds the outputs of the second signal summing unit 303 and third signal summing unit 304 and outputs their sum The output signal of the signal summing unit 307 is the output signal Q" of the filter unit 3; the output signal of the signal subtracting unit 306 is the output signal I" of the filter unit 3 If the signals output from the FFT unit 1 and coefficient calculation unit 4 are expressed in complex-valued signal notation, the filter shown in FIG 2 is an M-tap complex filter of the FIR type, and I" and Q" are the real part and imaginary part of the complex-valued signal output therefrom.

Next, the coefficient calculation unit 4 will be described On the basis of the outputs of the FFT unit 1 and filter unit 3, the coefficient calculation unit 4 calculates the optimum filter coefficients for removing the inter- carrier interference component included in the frequencydomain signals I, Q, and outputs the calculated coefficients to the filter unit 3.

The inter-carrier interference component in the orthogonal frequency-division multiplexed and transmitted signal will now be described Inter-carrier interference arises in orthogonal frequency-division multiplexed transmission systems from an offset between the transmitting and receiving carrier frequencies, and is a source of degradation of receiving performance The frequency offset arises from residual frequency error in the synchronous recovery circuit in the receiver, from phase noise in the tuner, and the like; the probability of errors in receiving increases and decreases according to the degree of frequency offset The orthogonal frequency-division multiplexed signal in an equivalent low-frequency system with an ideal channel, with channel noise ignored, is expressed by equation 1.

r(t) =Re EE Sk, exp L 2 (t-i) sj (t-i) ( 1) i-=O t 5 where ski is the complex transmit data transmitted by the k- th subcarrier in the i-th symbol in the orthogonal frequency-division multiplexed signal, Ts is the symbol period including the guard interval, ts is the symbol period excluding the guard interval, and N is the total number of subcarriers If the length of the guard interval is tg, fc is the function expressed by equation 2.

{:-t< t< t l 0: otherwise In the receiver, recovery of the transmitted data is realized by a Fourier transform of r(t) and demodulation of each subcarrier If there is an offset 8 f between the transmitting and receiving carrier frequencies at this time, the Fourier-transform output for the m-th subcarrier in the i-th symbol is expressed as in equation 3.

Xmj = 't 'r(t)expl-j 2 z(fm+ f)(t-i Tf)t ( 3) where fm is the intended frequency of the m-th subcarrier.

If equation 1 is substituted into equation 3, equation 4 is obtained.

N-I x, = S, jsmc(Sft)expl jrftl+ Es, exppl j(k -m-Sfit)lsincl(k-m f-)l ( 4) k= O km where sinc(x) is the sine function expressed by equation 5.

sin(,x) smc(x)= x ( 5) 7 AC In equation 4, the first term is the signal component transmitted by the m-th subcarrier; the second term expresses the inter-carrier interference component As can be seen in equation 4, the inter-carrier interference component is expressed as a linear sum of signals in which a gain and phase rotation determined by the frequency interval between the subcarrier of the principal component and each other subcarrier is applied to each other subcarrier.

Accordingly, the inter-carrier interference component can be removed by linear signal processing of the frequency-domain signal, using a linear filter such as a complex filter, following the Fourier transform.

In removing the inter-carrier interference component, it is necessary to optimize the coefficients of the complex filter so as to minimize the inter-carrier interference component in the complex filter output This can be accomplished by dynamically updating the filter coefficients by the method of steepest descent, for example The coefficient calculation unit 4 accordingly calculates the filter coefficients of the filter unit 3 dynamically, using an adaptive algorithm for coefficient optimization.

A coefficient updating method in which the CMA (Constant Modulus Algorithm) is employed as the adaptive algorithm used in the coefficient calculation unit 4 will be described next The CMA dynamically updates the filter coefficients so as to minimize the mean square error between the output signal of the filter and an ideal signal The formula for updating the i-th tap coefficient Ci N at time n is expressed by equation 6.

Cil = C, -u(ly^R)yxi,n ( 6) where xi N is the i-th tap input signal at time n, yn is the filter output at time n, g is the step-size parameter, R is a reference-level signal, and the asterisk () indicates the complex conjugate If a,, is the ideal signal in the filter output at time n, then the reference level R is a constant expressed by equation 7.

AR= ( 7) Next, as examples of the structure of the coefficient calculation unit 4, block diagrams in which the CMA algorithm is employed as the adaptive algorithm for coefficient optimization will be shown in FI Gs 3 and 4 In FIG 3, reference numeral 400 denotes a tap-coefficient updating unit that receives the output signal of the filter unit 3, the reference-level signal, the step-size parameter, and the frequency-domain signal input to the filter unit 3, or this signal delayed by a delay unit 401 in the coefficient calculation unit; reference numeral 401 denotes a delay unit in the coefficient calculation unit that receives the frequency-domain signal input to the filter unit 3, or a delayed version of this signal.

In FIG 3, x is the frequency-domain signal output from the FFT unit 1, expressed in complex-valued signal notation by equation 8.

x=I+ j Q ( 8) Also in FIG 3, y is the frequency-domain signal output from the filter unit 3, expressed in complex-valued signal notation by equation 9.

y= I'+j Q ( 9) Also in FIG 3, Cm is the complex-valued signal notation for the m-th tap coefficient in the filter unit 3 (m = 0, 1,, M 1).

In FIG 4, reference numeral 4000 denotes a complex conjugation unit that receives the frequency-domain signal input to the filter unit 3, or this signal delayed by a delay unit 401 in the coefficient calculation unit; reference numeral 4001 denotes a complex multiplication unit that receives the output of the complex conjugation unit 4000 and the output of the filter unit 3; reference numeral 4002 denotes a power calculation unit that receives the output of the filter unit 3; reference numeral 4003 denotes a first subtraction unit that receives the output of the power calculation unit 4002 and the reference-level signal; reference numeral 4004 denotes a first multiplication unit that receives the output of the first subtraction unit 4003 and the step-size parameter; reference numeral 4005 denotes a second multiplication unit that receives the outputs of the first multiplication unit 4004 and complex multiplication unit 4001; reference numeral 4006 denotes a second subtraction unit that receives the outputs of the second multiplication unit 4005 and a delay unit; reference numeral 4007 denotes the delay unit, which receives the output of the second subtraction unit 4006; the output of the delay unit 4007 is both the output of the tap- coefficient updating unit 400 and the output of the coefficient calculation unit 4.

Next, the operation will be described The tap- coefficient updating units 400 receive the frequency-domain signal input to the filter unit 3 or this signal delayed by a delay unit 401 in the coefficient calculation unit, the output signal of the filter unit 3, the reference-level signal, and the step-size parameter, and dynamically update the filter coefficients of the taps in the filter unit 3 so as to optimize them The delay units 401 in the coefficient calculation unit receive the frequency-domain signal input to the filter unit 3, or a delayed version of this signal, and impart a delay equivalent to the reciprocal of the FFT sampling rate in the FFT unit 1.

Next, the operation of the tap-coefficient updating units 400 will be described, using FIG 4 The complex conjugation unit 4000 outputs a complex-valued signal conjugate to the input complex-valued signal The complex multiplication unit 4001 receives the output of the complex conjugation unit 4000 and the output of the filter unit 3, multiplies these complex-valued signals, and outputs their product The power calculation unit 4002 receives the output of the filter unit 3, and calculates and outputs the square of the amplitude of that complex-valued signal The first subtraction unit 4003 subtracts the reference-level signal from the output of the power calculation unit 4002 and outputs the difference The first multiplication unit 4004 multiplies the output of the first subtraction unit 4003 by the step-size parameter and outputs the product The output of the first multiplication unit 4004 is a scalar value The second multiplication unit 4005 multiplies the complex- valued signal output from the complex multiplication unit 4001 by the scalar value output from the first multiplication unit 4004 The second subtraction unit 4006 subtracts the output of the second multiplication unit 4005 from theoutput of the delay unit 4007 and outputs the difference The delay unit 4007 receives the output of the second subtraction unit 4006, and imparts a delay equivalent to the reciprocal of the FFT sampling rate of the FFT unit 1.

A device for optimizing filter coefficients by the CMA expressed in equation 6 can be realized by configuring the coefficient calculation unit 4 as described above.

In the example of the coefficient calculation unit 4 shown in FIG 3, the signals obtained by delaying the frequency-domain signal input to the tap-coefficient updating units 400 were the output signals of the delay units 401 in the coefficient calculation unit, but the output signals of the delay units 300 in the filter 3, which are equivalent signals, can be used instead.

The CMA was shown as an example of an adaptive algorithm for coefficient optimization, but the DD (Decision Directed) algorithm, the DAMA (Decision Adjusted Modulus Algorithm), the LMS (Least Mean Squares) algorithm, or the RLS (Recursive Least Squares) algorithm can also be used.

As described above, filtering is performed on the frequency-domain signal output from the Fourier transform, and filter coefficients capable of removing the inter- carrier interference component are calculated by an adaptive algorithm, so inter-carrier interference can be reduced when the signal is demodulated, and the probability of errors in the recovered data can be reduced.

Changes in the degree of inter-carrier interference can be followed adaptively because an adaptive algorithm is used as the algorithm for optimizing the filter coefficients.

Second Embodiment In the first embodiment described above, when the filter coefficients were optimized, the coefficient calculation means operated equivalently on information for all subcarriers, but next an embodiment will be shown that performs special operations on a pilot carrier.

FIG 5 is a block diagram illustrating an orthogonal frequency-division multiplexed signal receiver according to a second embodiment of the invention In FIG 5, elements 1 to 4 are similar to the corresponding elements in the first embodiment The step-size parameter g input to the coefficient calculation unit 4, however, is the output of a step-size control unit 6 Reference numeral 5 denotes a pilot timing signal generator The step-size control unit 6 receives the output of the pilot timing signal generator 5.

Next, the operation will be described The operation of the FFT unit 1, demodulation unit 2, filter unit 3, and coefficient calculation unit 4 is similar to the operation shown in the first embodiment When signal transmission is performed by orthogonal frequency-division multiplexing, several fixed subcarriers may be used as pilot carriers for the purpose of improving the synchronizing performance of the receiver, for example In this case, since the data transmitted on the pilot carriers are known data in the receiver, when filter coefficients are calculated in the coefficient calculation unit 4, they can be optimized more rapidly and more accurately, as compared with other subcarriers In the second embodiment, therefore, the step- size parameter is made larger for the pilot subcarriers than for other subcarriers, and weighting control is enabled when the coefficients are updated The pilot timing signal generator 5 therefore outputs a signal indicating whether or not the output signal of the filter unit 3 is a pilot subcarrier The step-size control unit 6 outputs step-size parameters according to the pilot timing signal output from the pilot timing signal generator 5 For example, when the pilot timing signal indicates that the output of the filter unit 3 is a pilot carrier, a larger value is output as the step-size parameter than at other times.

When the pilot timing signal indicates that the output of the filter unit 3 is not a pilot carrier, the value of the step-size parameter in the step-size control unit 6 may be set to zero.

Since the step-size parameter is larger when coefficients for pilot carriers are updated than at other times, as described above, the filter coefficients when inter-carrier interference is removed can be determined more rapidly.

Also, an inter-carrier interference canceler of higher accuracy can be realized, because it is configured to use the pilot-carrier information in updating the coefficients.

Third Embodiment The second embodiment described above was configured to permit switching between different step-size parameters for pilot carriers and other carriers when coefficients were updated Next, an embodiment that controls the step-size parameter according to the power level of the frequency- domain signal will be described.

FIG 6 is a block diagram illustrating an orthogonal frequency-division multiplexed signal receiver according to a third embodiment of the invention In FIG 6, elements 1 to 4 are similar to the corresponding elements in the first embodiment The step-size parameter input to the coefficient calculation unit 4, however, is the output of a step-size control unit 8 Reference numeral 7 denotes a power calculation unit that receives the output of the FFT unit 1; the step-size control unit denoted by reference numeral 8 receives the output of the power calculation unit 7 Also in FIG 6, P denotes the output of the power calculation unit 7, which expresses the instantaneous power signal of each subcarrier.

Next, the operation will be described The operation of the FFT unit 1, demodulation unit 2, filter unit 3, and coefficient calculation unit 4 is similar to the operation shown in the first embodiment When a signal transmitted by orthogonal frequency-division multiplexing propagates through a multipath channel, the received subcarriers are attenuated or amplified according to the channel The signal-power-to-noise-power ratio then differs between attenuated subcarriers and amplified subcarriers; amplified subcarriers are less susceptible to the effects of noise, and are more reliable as information for coefficient updating; attenuated subcarriers are greatly affected by noise, and their reliability as information for coefficient updating is low The signal power of each subcarrier is accordingly calculated, and the value of the step-size parameter is controlled so as to be larger for subcarriers with high power and smaller for subcarriers with low power.

The power calculation unit 7 receives the output of the FFT unit 1, and calculates and outputs the power of each subcarrier The step-size control unit 8 receives the output of the power calculation unit 7, and controls and outputs the step-size parameter on the basis of the received power level.

As an example of the structure of the step-size control unit 8, FIG 7 shows a block diagram in which the step-size control unit 8 is configured to select and output the step- size parameter according to the output of the power calculation unit 7 In FIG 7, reference numeral 800 denotes a step-size selection signal generator that receives power information output from the power calculation unit 7; reference numeral 801 denotes a step-size selection unit that receives the output of the step-size selection signal generator 800 and a plurality of step-size parameter candidate values, which are real constants with different values; the output of the step-size selection unit 801 is the step-size parameter which is the output of the step-size control unit 8 The step-size parameter candidate values in FIG 7 are denoted gk (k = 1, 2,, K).

Next, the operation will be described The step-size selection signal generator 800 classifies the magnitude of the input power information into K levels, and outputs the result as a step-size selection signal The classification is done so that larger power values will lead to the selection of larger values as the step-size parameter The step-size selection unit 801 selects one of the K step-size parameter candidate values according to the signal received from the step-size selection signal generator 800, and outputs a step-size parameter corresponding to the output of the filter unit 3.

Next, as another example of the structure of the step- size control unit 8, FIG 8 shows a block diagram in which the step-size control unit 8 is configured to convert the output level of the power calculation unit 7 to a step-size parameter by a predetermined conversion function, and output the step-size parameter In FIG 8, reference numeral 802 denotes a step-size conversion table unit that receives the power information output from the power calculation unit 7; the output of the step-size conversion table unit 802 is the step-size parameter output from the step-size control unit 8.

Next, the operation will be described The step-size conversion table unit 802 converts the input power level to a step-size parameter by the predetermined conversion function, and outputs a step-size parameter corresponding to the output of the filter unit 3.

An example of a conversion function is shown in equation 10, in which P denotes the power level and a is a positive real constant.

y=a P ( 10) Another example of a conversion function is shown in equation 11, in which P denotes the power level and P and r are positive real constants.

JU =,6 fia P ( 11) If conversion is performed using equation 10 or 11, the step-size parameter increases as the power increases.

Since the third embodiment is configured to be able to control the step-size parameter used in coefficient updating according to the power of the subcarriers, as described above, inter-carrier interference can be removed adaptively even from orthogonal frequency-division multiplexed signals that have been subjected to multipath effects Also, since signal power is calculated from the output of the FFT unit 1, the power can be calculated quickly, and dynamic updating of the filter coefficients can accordingly be carried out promptly.

Fourth Embodiment The third embodiment was configured to calculate the instantaneous power of each subcarrier as power information and control the step-size parameter accordingly Next, an embodiment that controls the step-size parameter according to average channel power information for each subcarrier will be shown.

FIG 9 is a block diagram illustrating an orthogonal frequency-division multiplexed signal receiver according to a fourth embodiment of the invention In FIG 9, elements 1, 3, 4, and 8 are similar to the corresponding elements shown in the first and third embodiments The signal input to the step-size control unit 8, however, is an average power signal output from a synchronously-modulated-signal demodulation unit 20 The synchronously-modulated-signal demodulation unit 20 receives the output of the filter unit 3; the output of the synchronously-modulated-signal demodulation unit 20 is the recovered data, or recovered error-correction encoded data, of the transmitted digital data In FIG 9, P' denotes the average power signal.

Next, the operation will be described The operation of the FFT unit 1, filter unit 3, coefficient calculation unit 4, and step-size control unit 8 in FIG 9 is similar to the operation shown in the first and third embodiments When QPSK (Quadrature Phase Shift Keying) or QAM (Quadrature Amplitude Modulation) is used as the modulation system of the subcarriers transmitted by orthogonal frequency-division multiplexing, a pilot carrier for demodulation is often employed, in which case signal demodulation is realized by a device such as, for example, the synchronously-modulatedsignal demodulation unit 20 shown in FIG 10 Besides carrying out demodulation of subcarriers with modulation systems such as the above, the synchronously-modulated- signal demodulation unit 20 shown in FIG 9 outputs average power information for each subcarrier, which information is obtained in the demodulation process The step-size control unit 8 receives this average power information, namely, the average power signal P', and controls the step-size parameter.

The operation of the synchronously-modulated-signal demodulation unit 20 will be described using FIG 10 In FIG.

10, reference numeral 200 denotes a synchronous-demodulation pilot carrier demodulation unit that receives the outputs I" and Q" of the filter unit 3; reference numeral 201 denotes a channel estimation unit that receives the output of the synchronous-demodulation pilot carrier demodulation unit 200; reference numeral 202 denotes a complex conjugation unit that receives the output of the channel estimation unit 201; reference numeral 203 denotes a delay unit that receives the outputs I" and Q' of the filter unit 3; reference numeral 204 denotes a complex multiplication unit that receives the outputs of the delay unit 203 and complex conjugation unit 202; reference numeral 205 denotes a power calculation unit that receives the output of the channel estimation unit 201; reference numeral 206 denotes a division unit that receives the outputs of the power calculation unit 205 and complex multiplication unit 204; the output of the division unit 206, which is an output of the synchronously-modulated-signal demodulation unit 20, is the recovered data, or recovered error-correction encoded data, of the transmitted digital data; the output of the power calculation unit 205, which is another output of the synchronously-modulated-signal demodulation unit 20, is the average power signal P'.

Next, the operation will be described The synchronous- demodulation pilot carrier demodulation unit 200 receives the outputs I" and Q" of the filter unit 3, extracts the pilot carriers included therein for synchronous demodulation, divides them by corresponding known signals, and outputs the quotients The outputs of the synchronous-demodulation pilot carrier demodulation unit 200 are signals expressing transfer characteristics on the channel corresponding to each pilot carrier The channel estimation unit 201 receives the outputs of the synchronous-demodulation pilot carrier demodulation unit 200, estimates the channel characteristics for all subcarriers by interpolation in the time direction and frequency direction, and outputs the estimated characteristics The complex conjugation unit 202 receives the outputs of the channel estimation unit 201, and outputs complex conjugate signals thereof The delay unit 203 imparts a delay equal to the delay occurring in the complex conjugation unit 202, synchronous-demodulation pilot carrier demodulation unit 200, and channel estimation unit 201 to the I" and Q" outputs from the filter unit 3, and outputs the delayed I" and Q" The signals output from the complex conjugation unit 202 thus correspond to the subcarriers output from the delay unit 203 The complex multiplication unit 204 receives the outputs of the delay unit 203 and complex conjugation unit 202, and outputs their complex product The power calculation unit 205 receives the output of the channel estimation unit 201, and calculates and outputs the square of the amplitude of that complex- valued signal The division unit 206 divides the output of the complex multiplication unit 204 by the scalar value calculated by the power calculation unit 205, and outputs the quotient The complex signals thus obtained as the output of the division unit 206 are the recovered data, or recovered error-correction encoded data, of the transmitted digital data The output of the power calculation unit 205 is the average power information of each subcarrier.

Accordingly, when subcarriers modulated by a synchronous modulation system such as QPSK or QAM are demodulated, in the demodulation process, it is possible to obtain average power information of each subcarrier in the channel The fourth embodiment therefore controls the step- size parameter on the basis of the average power signal Pl obtained in this way, as shown in FIG 9 The step-size parameter also increases with the power in this case, as described in the third embodiment.

Since the fourth embodiment is configured to be able to control the step-size parameter used in coefficient updating according to the average power of each subcarrier, as described above, inter-carrier interference can be removed adaptively even from orthogonal frequency-division multiplexed signals that have been subjected to multipath effects, regardless of changes in the signal level of the transmitted data Also, it is not necessary to add a new circuit to calculate the power, because the power calculation function already provided in the demodulation unit 20 is used.

Fifth Embodiment The third embodiment was configured to calculate the instantaneous power of each subcarrier as power information and control the step-size parameter accordingly Next, an embodiment that controls the reference-level signal used in coefficient updating according to this instantaneous power information will be shown.

FIG 11 is a block diagram illustrating an orthogonal frequency-division multiplexed signal receiver according to a fifth embodiment of the invention In FIG 11, elements 1 to 4 and 7 are similar to the corresponding elements shown in the third embodiment In the coefficient calculation unit 4, however, a fixed step-size value is input as the stepsize parameter, and the output of a reference control unit 9 is used as the reference-level signal The reference control unit 9 receives the output of the power calculation unit 7.

Next, the operation will be described The operation of the FFT unit 1, demodulation unit 2, filter unit 3, coefficient calculation unit 4, and power calculation unit 7 is similar to the operation shown in the third embodiment.

When a signal transmitted by orthogonal frequency-division multiplexing propagates through a multipath channel, the received subcarriers are attenuated or amplified according to the channel In this case, even if inter-carrier interference does not occur, the signal level of the subcarriers in the frequency-domain signal varies with each subcarrier, so the value of the reference-level signal R used in coefficient updating can be thought to vary accordingly The reference control unit 9 therefore selects and outputs a reference value on the basis of the instantaneous power signal P output from the power calculation unit 7.

As an example of the structure of the reference control unit 9, FIG 12 shows a block diagram in which the reference control unit 9 is configured to select and output the reference-level signal according to the output of the power calculation unit 7 In FIG 12, reference numeral 900 denotes a reference-value selection signal generator that receives power information output from the power calculation unit 7; reference numeral 901 denotes a reference-value selection unit that receives the output of the referencevalue selection signal generator 900 and a plurality of reference candidate values, which are real constants with different values; the output of the reference-value selection unit 901 is the reference-level signal which is the output of the reference control unit 9 The reference candidate values in FIG 12 are denoted Rk (k = 1, 2,, K).

Next, the operation will be described The reference- value selection signal generator 900 classifies the magnitude of the input power information into K levels, and outputs the result as a reference-value selection signal.

The classification is done so that larger power values will lead to the selection of larger values as the reference- level signal The reference-value selection unit 901 selects one of the K reference candidate values according to the signal received from the reference-value selection signal generator 900, and outputs a reference-level signal corresponding to the output of the filter unit 3.

Next, as another example of the structure of the reference control unit 9, FIG 13 shows a block diagram in which the reference control unit 9 is configured to convert the output level of the power calculation unit 7 to a reference-level signal by a predetermined conversion function, and output the reference-level signal In FIG 13, reference numeral 902 denotes a reference-value conversion table unit that receives the power information output from the power calculation unit 7; the output of the reference- value conversion table unit 902 is the reference-level signal output from the reference control unit 9.

Next, the operation will be described The reference- value conversion table unit 902 converts the input power level to a reference-level signal by a predetermined conversion function, and outputs a reference-level signal corresponding to the output of the filter unit 3.

An example of a conversion function is shown in equation 12, in which P denotes the power level and N and R.

are positive real constants; the right side of equation 12 agrees with the right side of equation 7 when the signal is transmitted on an ideal channel.

R = 7 P+& ( 12) When conversion is performed using equation 12, the reference level increases with increasing power.

Since the third embodiment is configured to be able to control the reference-level signal used in coefficient updating according to the power of the subcarriers, as described above, inter-carrier interference can be removed adaptively even from orthogonal frequency-division multiplexed signals that have been subjected to multipath effects Also, since signal power is calculated from the output of the FFT unit 1, the power can be calculated quickly, and the dynamic updating of the filter coefficients can accordingly be carried out promptly.

Sixth Embodiment The fifth embodiment was configured to calculate the instantaneous power of each subcarrier as power information and control the reference-level signal accordingly Next, an embodiment that controls the reference-level signal according to average power information of the channel for each subcarrier will be shown.

FIG 14 is a block diagram illustrating an orthogonal frequency-division multiplexed signal receiver according to a sixth embodiment of the invention In FIG 14, elements 1, 3, 4, 9, and 20 are similar to the corresponding elements shown in the first, fourth, and fifth embodiments The signal input to the reference control unit 9, however, is an average power signal output from the synchronously- modulated-signal demodulation unit 20.

Next, the operation will be described The operation of the FFT unit 1, filter unit 3, coefficient calculation unit 4, reference control unit 9, and synchronously-modulated- signal demodulation unit 20 in FIG 14 is similar to the operation shown in the first, fourth, and fifth embodiments.

When QPSK or QAM is used as the modulation system of the subcarriers transmitted by orthogonal frequency-division multiplexing, since demodulation is realized by a device such as the synchronously-modulated-signal demodulation unit 20, average power information of each subcarrier can be obtained therefrom This average power information, namely, the average power signal P', is used as the input signal to the reference control unit 9, thus controlling the reference-level signal The reference level also increases with the power in this case, as described in the fifth embodiment.

Since the fourth embodiment is configured to be able to control the reference-level signal used in coefficient updating according to the average power of each subcarrier, as described above, inter-carrier interference can be removed adaptively even from orthogonal frequency-division multiplexed signals that have been subjected to multipath effects, regardless of changes in the signal level of the transmitted data Also, it is not necessary to add a new circuit to calculate the power, because the power calculation function already provided in the demodulation unit 20 is used.

Those skilled in the art will recognize that the embodiments described above can be modified in various ways within the scope claimed below.

Claims

1 An orthogonal frequency-division multiplexed signal receiver recovering transmitted data from an orthogonal frequency-division multiplexed signal, comprising:

Fourier transform means for transforming the received time-domain signal into a frequency-domain signal; coefficient calculation means; filter means receiving the output of said Fourier- transform means and the output of said coefficient calculation means, and performing filtering with filter coefficients output from said coefficient calculation means; and demodulation means receiving the output of said filtering means, and performing demodulation by a demodulation method corresponding to the modulation method of each subcarrier; wherein said coefficient calculation means receives the output of said filter means, the output of said Fourier transform means, a step-size value, and a reference value, successively and adaptively updates the filter coefficients so as to minimize inter-carrier interference in the signal output from said filter means, and outputs the updated filter coefficients.

2 The orthogonal frequency-division multiplexed signal receiver of claim 1, further comprising:

a pilot timing signal generating means that outputs a timing signal indicating whether or not the output signal of the filter means is a pilot carrier; and a step-size control means that receives the output of the pilot timing signal generating means and outputs a step- size parameter according to the timing signal; wherein said coefficient calculation means receives the output of said step-size control means as said step-size value, and updates said filter coefficients dynamically based thereupon.

3 The orthogonal frequency-division multiplexed signal receiver of claim 1, further comprising:

a power calculation means that calculates and outputs the signal power of each subcarrier; and a step-size control means that receives the output of the power calculation means, controls a step-size parameter based thereupon, and outputs the step-size parameter; wherein said coefficient calculation means receives the output of the step-size control means as said step-size value, and updates said filter coefficients dynamically based thereupon.

4 The orthogonal frequency-division multiplexed signal receiver of claim 3, wherein said power calculation means receives the output of said Fourier transform means, and calculates and outputs the signal power of each subcarrier.

The orthogonal frequency-division multiplexed signal receiver of claim 3, wherein when performing demodulation on the basis of channel characteristics estimated for each subcarrier, said demodulation means calculates a quantity equivalent to the average power of each subcarrier, said demodulation means thus also functioning as said power calculation means; and said step-size control means receives information output from said demodulation means expressing said average power of each subcarrier, and controls and outputs said step-size parameter on the basis thereof.

6 The orthogonal frequency-division multiplexed signal receiver of claim 1, further comprising:

a power calculation means that calculates and outputs the signal power of each subcarrier; and a reference control means that receives the output of the power calculation means, and controls a reference-level value based thereupon; wherein said coefficient calculation means receives the output of the reference control means as said reference value, and updates said filter coefficients based thereupon.

7 The orthogonal frequency-division multiplexed signal receiver of claim 6, wherein said power calculation means receives the output of said Fourier transform means, and calculates the signal power of each subcarrier.

8 The orthogonal frequency-division multiplexed signal receiver of claim 6, wherein when performing demodulation on the basis of channel characteristics estimated for each subcarrier, said demodulation means calculates a quantity equivalent to the average power of each carrier, said demodulation means also functioning as said power calculation means; and said step-size control means receives information output from said demodulation means expressing said average power of each subcarrier, and controls and outputs said reference value on the basis thereof.

9 An OFDM receiver comprising Fourier transform means for transforming a time-domain signal into a frequency domain signal, demodulation means and filter means provided between the Fourier transform means and the demodulation means.

An OFDM receiver as claimed in claim 9 comprising means for calculating the filter coefficients using an adaptive algorithm.

11 A method of processing an OFDM signal comprising transforming a time-domain signal into a frequency domain signal, filtering the frequency domain signal and demodulating the filtered signal.

12 An OFDM receiver substantially as hereinbefore described as an embodiment and as shown in the respective accompanying drawings.

13 A method of processing an OFDM signal substantially as hereinbefore described in an embodiment with reference to the respective accompanying drawings.