KR102088430B1 - transmission apparatus and method using multi-carrier modulation - Google Patents

Abstract

The present invention relates to an apparatus for modulating and transmitting information to be transmitted by a multi-carrier transmission method, and an IDFT unit for converting signals input to each of a plurality of points into an inverse discrete Fourier and outputting them for each point, and an IDFT unit for transmitting transmission information A signal generation processing unit that generates and outputs corresponding to the number of points, a memory unit for dividing and storing signals output from the IDFT unit for each point, a polyphase filter coefficient unit having a filter coefficient corresponding to each point of the IDFT unit, and a memory A phase shift that performs phase shifting on the signal filtered by the filtering unit and the filtering unit that performs filtering by matching the corresponding point-specific signal stored in the unit and the corresponding filter coefficient of the polyphase filter coefficient unit. Provide wealth. According to such a multi-carrier transmission method transmitting apparatus and a transmission method, FMT modulation can be efficiently implemented without including DC as a sub-channel component, and an advantage that can be applied to a multi-carrier modulation method using only different sub-channels is provided. do.

Description

Transmission apparatus and method using multi-carrier transmission method

The present invention relates to a multi-carrier transmission method transmitting apparatus and a transmission method, and in detail, a transmission method and apparatus for transmitting digital information using a sub-carrier symmetrically without a DC component among sub-carrier components while using the multi-carrier transmission method It is about.

The multi-carrier transmission (FMT) method uses multi-carrier modulation to transmit data using multiple carriers in order to overcome the frequency selective characteristics of a channel experienced when transmitting a single wideband carrier. This is how it is used.

The multi-carrier transmission method is variously posted, such as Korean Patent Publication No. 10-2013-0031189.

The basic concept of the multi-carrier transmission method can be found in a multi-channel transmission system, and because a wideband signal is analyzed and transmitted into multiple narrowband signals, a selective channel of a wideband frequency is a non-selective channel of a narrowband frequency. Since it can be approximated to, the receiving end has the advantage of compensating the channel by using a channel equalization method with low complexity for each sub-channel.

1 shows a transmission / reception structure of an FMT transmission method using a plurality of band-pass filters.

In FIG. 1, A _{M- 1} [nT] denotes a symbol transmitted from the nth symbol time to the Mth subchannel, and A _{M- 1 1} [nT] is (M + υ) times oversampled and the Mth band pass filter h _{M - 1} (t). Here, h _{M - 1} (t) is a baseband filter h (t) as a prototype filter, and varies according to the subcarrier fm and is given as h (t) e ^j2πfmt . M is the number of subcarriers, υ is 0 or an integer greater than 0, the subcarrier interval is (M + υ) / (MT), the sampling rate is (M + υ) / T, and T is the symbol transmission period.

In this method, the bandpass filters h _M (t) and h * _M (t) are actually composed of a baseband filter bank and an oscillator bank. The implementation method of this structure is called a “direct implementation” method.

In the FMT transmission method, when the reciprocal relationship between the symbol period and the subcarrier is established (υ = 0) is called a “critical sampled case”, otherwise (υ> 0) is called a “non-critical sampled case”. In the case of the critical sampled case, the non-critical sampled case is more practical in terms of complexity or interference between subchannels in the FMT transmission method because it is possible to prevent overlap between subchannels by limiting each subchannel using an ideal band limiting filter. to be. In the case of a non-critical sampled case, the subcarrier interval is (M + υ) / M times 1 / T, where M represents the number of subchannels or subcarriers used.

In the non-critical sampled case, since the spacing of subcarriers increases (M + υ) / M times, if an ideal low-pass filter with a passband width of 1 / T is used as a band-limited filter, the υ / (MT) You will have a guard band.

The FMT type multicarrier transmission system can overcome the frequency selective characteristics of a channel generated when transmitting a wideband single carrier, but has a high implementation complexity because it needs to use a filter and an oscillator as many as the number of subcarriers used. In order to prevent interference, a filter having good band-limiting characteristics must be used, and the sampling period must be shorter as a wider band is used, which increases the burden on the filter implementation when the number of subcarriers is large.

As described above, an “efficient implementation” method having an IDFT and a poly-phase filter structure has been proposed and used in order to reduce the negative feeling on the implementation of the “direct implementation” method. The FMT method using IDFT and poly-phase filter is based on the inclusion of DC (direct current) component as one of the sub-channels, and a method for signal transmission in the baseband, such as VDSL, has been proposed. However, in the digital VHF band in the maritime VHF band, DC is located in the middle of a sub-channel and does not include a DC component as a sub-channel, so a transceiver structure reflecting these characteristics is required. In addition, in order to transmit the FMT modulated signal generated in the baseband to the carrier frequency, a signal generated in the baseband must be additionally interpolated to generate a smooth analog signal, so that spurious radiation emitted outside the channel band in the carrier frequency band can be reduced. After the signal is generated, a transmitter considering the interpolation function is needed.

The present invention has been devised to solve the above requirements, and provides a multi-carrier transmission method transmission apparatus and transmission method that does not include DC as a sub-channel component and uses a sub-channel symmetrically based on DC. It has a purpose.

In addition, another object of the present invention is to use a multi-carrier modulation scheme having the same sub-channel interval and symbol rate but different in the number of sub-channels used, and multi-carrier modulation in which only two or more sub-channels differ in number of times. When only one multi-carrier modulation is used at one moment (eg, a TDMA channel connection is used, and the number of sub-channels used in each time slot is different), two or more multi-carrier modulations as described above can be performed in one. It is to provide a multi-carrier transmission method transmission apparatus and transmission method that can be implemented by using in a modulator.

In order to achieve the above object, a multi-carrier transmission method transmitting device according to the present invention is a device for modulating and transmitting information to be transmitted by a multi-carrier transmission method, and converts signals input to each of a plurality of points by inverse discrete Fourier transform. An IDFT unit that outputs not much; A signal generation processing unit which generates and outputs transmission target transmission information corresponding to the number of points of the IDFT unit; A memory unit for dividing and storing the signal output from the IDFT unit for each point; A polyphase filter coefficient unit provided with a filter coefficient corresponding to each point of the IDFT unit; A filtering unit that performs filtering by matching a signal for each point stored in the memory unit and a corresponding filter coefficient of the polyphase filter coefficient unit; And a phase shift unit configured to perform phase shift on the signal filtered by the filtering unit so that the DC component is not included in the transmission signal.

Preferably, the number of points of the IDFT unit is extended to an integer multiple of 2 to 16 times than the number of sub-channels output through the phase shift unit.

In addition, it is preferable that the signal generation processing unit is configured to support signal transmission of different frequency bandwidths.

According to an aspect of the present invention, the signal generation processing unit corresponds to a low frequency based on an intermediate value of the communication frequency in the IDFT unit having an extended point than the number of subchannels to be applied to communication, and corresponds to each of the subchannels. A real symbol signal is input to a point, and a zero value is input to a point corresponding to a high frequency based on an intermediate value of the communication frequency.

Further, the polyphase filter coefficient unit generates a discretized filter coefficient from the circular filter to correspond to the number of oversampling per symbol that increases in proportion to the number of extended points applied to transmit the transmission information, and the generated filter coefficient Is generated as a sub-filter structure corresponding to the number of over-sampling per symbol, and the coefficient of the sub-filter is determined by a value separated by the number of over-sampling per symbol from the discretized filter coefficient.

라디안(radian) 만큼 증가되도록 상기 필터링부의 출력신호를 위상회전시킨다.In addition, when the number of points in the IDFT unit is M, the phase shift unit starts from 0 and the phase is changed for each sample.

The output signal of the filtering unit is rotated in phase so as to be increased by radians.

In addition, in order to achieve the above object, a multi-carrier transmission method transmission method according to the present invention in a method for transmitting transmission information in a multi-carrier transmission method, a. An IDFT having an extended point than the number of subchannels to be applied to communication corresponds to a low frequency based on an intermediate value of a communication frequency, and inputs a real symbol signal to a point corresponding to each subchannel, and an intermediate value of the communication frequency Inputting a zero (0) value to the point corresponding to the high frequency based on the; I. An inverse discrete Fourier transform of signals input to each point in the IDFT, and outputting each point; All. Performing filtering by matching the signal output from the IDFT unit with a corresponding filter coefficient for each point; la. And performing phase shift so that the DC component is not included in the filtered signal.

Preferably, the multi-step is c-1. Generating a discretized filter coefficient from a circular filter to correspond to the number of oversampling per symbol that increases in proportion to the number of extended points applied to transmit the transmission information; C-2. Generating the filter coefficient generated in step C-1 in a sub-filter structure corresponding to the number of over-sampling per symbol, and the coefficient of the sub-filter is a value separated by the number of over-sampling per symbol from the discretized filter coefficient; It includes.

이 첫 번째 부필터의 폴리페이스 부필터에 위치하고, 첫 번째 부필터 계수 중 가장 큰 값을 갖는 h(0)가 부필터의 가운데 위치하게 하고, 그 위치를 FL이 짝수이면 FL/2 또는 FL이 홀수이면 floor(FL/2)로 결정하고, 상기 FL은 폴리페이스 필터의 부필터의 필터계수의 길이이며, 통신시스템에서 요구하는 상승시간이 N_R 심볼구간일 때 첫 번째 심볼이 생성하기 전에 (FL/2- N_R) 심볼 갯수만큼 미리 입력하여 IDFT부에서 연산을 수행하고, 연산한 데이터를 메모리부에 저장하고, 상기 다단계는 상기 메모리부의 P번째 부메모리부부터 시작하여 폴리페이스 필터의 첫 번째 부필터와 순차적으로 매칭시켜 필터링을 수행하고, 상기 라단계는

의 위상을 시작 위상으로하여 상기 필터링된 결과와 계산된 위상회전량를 승산하여 출력하고, 위상은 필터링된 결과와 곱할때마다

만큼의 위상을 증가시키며, 상기

이고, 상기

으로 계산되는 값이다.Preferably, steps a and b include a polyface filter composed of M sub-filters and a discretized sub-filter coefficient corresponding to h (0) having the largest value in a circular filter when oversampling per K symbol. Set to do

Located in the polyphase sub-filter of the first sub-filter, h (0) having the largest value among the first sub-filter coefficients is positioned in the center of the sub-filter, and FL / 2 or FL if the position is even If it is odd, it is determined as floor (FL / 2), and FL is the length of the filter coefficient of the sub-filter of the polyface filter, and before the first symbol is generated when the rising time required by the communication system is the N _R symbol interval ( FL / 2- N _R ) The number of symbols is input in advance to perform the operation in the IDFT unit, the calculated data is stored in the memory unit, and the multi-step starts at the P-th sub-memory unit of the memory unit and starts the first of the polyface filter. Filtering is performed by sequentially matching the second sub-filter.

Each phase is multiplied and output by multiplying the filtered result and the calculated phase rotation amount by using the phase of as the starting phase.

Increases the phase by

Is the above

It is calculated as.

According to the multi-carrier transmission method transmission apparatus and transmission method according to the present invention, FMT modulation can be efficiently implemented without including DC as a sub-channel component, and can also be applied to a multi-carrier modulation method using only different number of sub-channels. Offers advantages.

1 is a view for explaining a general multi-carrier transmission scheme,
2 is a diagram showing an example of an FMT signal applied to the present invention that does not include a DC component as a sub-channel,
3 is a view showing a multi-carrier transmission method transmission apparatus according to the present invention.

Hereinafter, a multi-carrier transmission method transmission apparatus and transmission method according to a preferred embodiment of the present invention will be described in more detail with reference to the accompanying drawings.

First, the transmission apparatus according to the present invention comprises a digital signal processing part for digitally processing a transmission signal in the baseband and an analog processing process part for converting the baseband analog signal into a carrier frequency band and transmitting the digital signal processing part. The signal generated in is assumed to have a structure that interfaces through a DAC (Digital-to-Analog Converter), but is not limited to a specific configuration.

Hereinafter, a process of processing a transmission signal in the case of using the multi-carrier transmission (FMT) method of the transmission apparatus according to the present invention will be described in detail.

First, as described in FIG. 1, M is the number of subcarriers, υ is an integer greater than 0, K = (M + υ), the subcarrier interval is K / (MT), and the sampling rate is K / T, T Is a transmission period of the symbol, Am [nT] represents a symbol transmitted from the nth symbol time to the mth subchannel, and Am [nT] is K times oversampled and passes through the mth band pass filter hm (t). .

Here, hm (t) is the baseband filter h (t) as a prototype filter, and varies depending on the subcarrier fm, given h (t) e ^j2πfmt , the output signal x (kK / T) of the kth modulator is As described above, it can be expressed as a sum of signals that have passed through a symbol sequence of each subchannel and a bandpass filter.

<Equation 1>

,m=0, 1, ...,N-1 에서 DC를 부채널 중간에 위치하기 위해

만큼의 주파수 편이를 추가적으로 고려하여,

를 사용하였으며,

를 적용하였다.Here, as shown in FIG. 2, the frequency of the sub-channel is reflected in the center of DC, and

To place DC in the middle of the subchannel at, m = 0, 1, ..., N-1

Considering the frequency shift additionally,

Was used,

Was applied.

In Equation 1, as shown in FIG. 2, the last term is added to reflect the arrangement around DC.

To summarize Equation 1, Equation 2 is as follows.

<Equation 2>

,,i=0,1,...,N-1 로 나타내면,here,

When represented by ,, i = 0,1, ..., N-1,

<Equation 3>

로 n번째 심볼 시간의 M개 부채널들의 입력신호를 IDFT 연산한 결과의 i번째 시간에 해당하는 결과이다.here,

As a result, it corresponds to the i-th time of the IDFT operation of the input signals of the M sub-channels of the n-th symbol time.

의 관계식을 다시 적용하고,

와 같이 표현할 수 있고 이를 적용하면,Equation 3

Reapply the relational expression of

It can be expressed as

<Equation 4>

은 k를 M으로 나눈 나머지 값으로 0부터 M-1 사이의 값을 갖는 연산이다.It becomes like this. here,

Is the remainder of k divided by M, which is an operation with a value between 0 and M-1.

은 심볼간격으로 신호가 발생하고, 이를 심볼당 K배 과샘플링하여 필터링한 후, 위상천이를 시키는 과정과 동일하므로, 도 3과 같이 구현할 수 있다.In equation (4)

The signal is generated at symbol intervals, and it is the same as the process of performing phase shift after filtering by sampling K times per symbol, so it can be implemented as shown in FIG. 3.

Hereinafter, a transmission apparatus 100 according to the present invention will be described with reference to FIG. 3.

Referring to FIG. 3, the transmission apparatus 100 according to the present invention includes a signal generation processing unit 110, an IDFT unit 120, a memory unit 130, a poly-phase filter coefficient unit 140, and filtering The unit 160 and the phase shift unit 170 are provided.

The signal generation processing unit 110 generates and outputs transmission target transmission information corresponding to the number of points of the Inveres discrete fourier transform (IDFT) 120.

The signal generation processing unit 110 may be constructed to support signal transmission of different frequency bandwidths.

The IDFT unit 120 inversely discrete Fourier transforms signals input to each of a plurality of points and outputs each point.

The memory unit 130 stores the signals output for each point from the IDFT unit 120 for each point.

The polyphase filter coefficient unit 140 is provided with a filter coefficient corresponding to each point of the IDFT unit 120.

The filtering unit 150 performs filtering by matching a signal for each point stored in the memory unit 130 and a corresponding filter coefficient of the polyphase filter coefficient unit 140.

In the illustrated example, as the switching unit 150, the first switching unit 150a and the polyphase filter coefficient unit 140 are connected to provide the point-by-point signals stored in the memory unit 130 to the filtering unit 160 sequentially for each point. ) Is schematically illustrated as a structure provided by the second switching unit 150a that is connected to provide the filter unit 150 for each point in order to provide the filter coefficients for each point.

The phase shift unit 170 multiplies and outputs the phase shift value 180 described above to the signal filtered by the filtering unit 160 so that the DC component (DC) is not included in the transmission signal.

라디안(radian) 만큼 증가되도록 필터링 결과를 위상회전시킨다.When the number of extended points is M, the phase shifting unit 170 starts with a phase of 0 and the phase is changed for each sample.

The filtering results are phase rotated so that they are increased by radians.

의 관계를 만족하는 필터계수들의 집합이고,

의 범위를 가진다. 여기서,

는 는 이산화된 원형필터

의 전체 계수 집합이다.Meanwhile, the poly-phase filter coefficient unit 140 stores filter coefficients used for calculation when filtering in the filtering unit 160, and the n-th coefficient block of the K poly-phase filter

Is a set of filter coefficients that satisfy the relationship of

Has a range of here,

Is a discretized circular filter

Is the whole set of coefficients.

When M subchannels are used, symbols to be transmitted to the M subchannels to be used are input in parallel to points of the IDFT unit 120, and the IDFT unit 120 uses the inputted M subchannel transmission symbols. The IDFT operation is performed to output M complex time domain results. The output of the IDFT unit 120 is put into M data memories of the memory unit 130 for filtering.

It is assumed that the i-th memory stores the output signal of the i-th IDFT unit 120 and has a delay line structure that moves the memory at symbol time intervals. Such a structure is for enhancing the understanding of the present invention, and the actual implementation can be configured differently, such as sequentially writing to a memory or the like and using it virtually with a delay line structure. Each memory is assumed to have P spaces, which is set to be equal to the maximum length of the sub-filter when the circular filter is discretized and the discretized circular filter is divided into K sub-filters.

After the IDFT operation of the IDFT unit 120 and the storage in the memory of the memory unit 130 are finished, the contents of the 0th memory block of the memory unit 130 and the 0th filter coefficient block of the poly phase filter coefficient unit 140 are filtered. Filtering is performed as an input of the unit 160. More specifically, two input values are regarded as respective vectors, and the dot product of the two input vectors is calculated (having the same effect as a filtering operation).

When the calculation of the first input is completed, the switch of the first switching unit 150a selecting the memory unit 130 moves to make the contents of the first memory block the input of the filtering unit 160, and the second switching unit (150b) is also moved so that the first filter coefficient of the filter coefficient unit 140 is the input of the filtering unit 160, and the filtering unit 160 calculates the second filtering sample. This process is repeated K times to generate samples oversampled K times in one symbol interval. Here, since the filter coefficient unit 140 is composed of K pieces, the same filter coefficient is used in the same sample order in every symbol period. In addition, since there are M memory blocks in the memory unit 130, the (M-1) th process proceeds sequentially as described above, and the M th moves back to the 0 th memory block and moves sequentially. When the K sample calculation for the first symbol is finished, the switch of the first switching unit 150a for the memory unit 140 is located in the KM-th memory block, and the second switching unit for the filter coefficient unit 140 The switch of (150b) is located in the 0th filter coefficient unit (40). In the second symbol symbol section, after the operation of the IDFT unit 120 for the second M subchannel symbols and the storage in the memory of the memory unit 130 are completed, the two first and second switching units 150a and 150b Starting from the memory block and the filter coefficient block indicated by), the front and the back are sequentially input to the filtering block to calculate K oversampled samples again.

을 곱하여 위상 회전시키는 연산을 수행한다. The phase shift unit 170 is input to the output value of the filtering unit 160,

Multiply by to perform the phase rotation operation.

를 테이블화하여 사용할 수도 있다. 이러한 위상변위부(170)의 연산을 통하여 위상변조된 필터링 신호에 DC성분이 없게 처리된다. Here, since the phase shift unit 170 needs only 2M phase values,

Tables can also be used. Through the operation of the phase shifter 170, the phase-modulated filtering signal is processed without a DC component.

Through this, it is possible to implement a transmitter that can be adjusted to reduce the influence of DC offset that may occur when implementing the transmitter by not including a DC component in the transmission signal.

On the other hand, in the present invention, when M and K are determined, that is, when the number of subchannels to be used (M) and the interval between subchannels (K) are already determined, in order to increase the ease of implementation of the transmitter, the signal The generation processing unit 110 processes the sample to be oversampled in excess of K times.

In addition, in the relationship between the frequency domain and the time domain of the DFT (Discrete Fourier Transform), if the IDFT operation is performed by increasing the number of points in the IDFT and filling the high frequency region with a value of 0, IDFT calculation is performed in the time domain. The effect of sampling the signal occurs.

Here, the high-frequency region refers to a region having a high frequency based on an intermediate value of a communication frequency, and the low-frequency region refers to a region having a low frequency based on an intermediate value.

That is, the signal generation processing unit 110 corresponds to the low frequency based on the median value of the communication frequency to the IDFT unit 120 having a point extended from the number of sub-channels to be applied for communication, and the point corresponding to each sub-channel is real. A symbol signal is input, and a zero value is input to a point corresponding to a high frequency based on an intermediate value of the communication frequency. Next, the signals input to each point in the IDFT unit are inverse discrete Fourier transform and output by point, and the signal output from the IDFT unit is matched with a corresponding filter coefficient for each point to perform filtering. The phase shift can be performed so that it is not included.

By applying this principle, select a multiple that is L times larger than the number of subchannels M to be used, and this is called MM1 = L * M (where L = 2,3 .... any integer is possible), and the L determined earlier Apply to calculate KK1 = L * K. Preferably, the number of points of the IDFT unit 120 is extended to an integer multiple of 2 to 16 times than the number of sub-channels output through the phase shift unit 160. That is, L is an integer of 2 to 16 is applied.

By applying the calculated MM1 and KK1 as new M and K, an implementation structure of multicarrier modulation described above is generated. Then, from the generated frequency domain of the MM1 point, the MFT of the IDFT unit 120 input corresponding to the frequency corresponding to the subchannel to be used (corresponds to the IDFT input frequency domain corresponding to the low frequency domain) is used. , Otherwise, the IDFT unit 120 is input with a value of 0. In addition, the poly-phase filter is calculated by oversampling KK1 times per symbol to calculate a circular filter, and constitutes a poly-phase filter composed of KK1 pieces. In this way, the transmitter is configured, and in the above-described multicarrier modulator implementation structure, MM1 is used as the new M, KK1 is used as the new K, and the same method is operated to implement a modulator capable of generating KK1 times and sampled sample data per symbol. You can.

In the above, when MM1 becomes an N-approved number of 2, since FFT (Fast Fourier Transform) calculation can be used instead of IDFT, a modulator can be implemented more efficiently.

를 DAC (Digital-to-Analog Converter)를 거쳐 아날로그 신호로 변환할 때 발생할 수 있는 비선형적 특성을 줄이기 위해 추가적인 보간을 사용하는 경우에 보간기 필터 설계시 통과대역과 정지대역사이의 주파수 차이를 크게 할 수 있어 보간 필터를 효율적으로 설계할 수 있다. These modulators are output signals in the digital signal processing area.

When using additional interpolation to reduce the nonlinear characteristics that can occur when converting to analog signal through DAC (Digital-to-Analog Converter), the frequency difference between passband and stopband is greatly increased when interpolator filter is designed. The interpolation filter can be designed efficiently.

<Example 1>

A system that performs digital communication in the maritime VHF band transmits digital data using a multi-carrier modulation using 32 subchannels. At this time, the interval between sub-channels is separated by 2.7 kHz, and the data is transmitted at a symbol rate of 2400 baud through the sub-channel. At this time, the existing method uses a 32-point IDFT, and a sample sequence can be generated using an IDFT result and a poly-phase filter, but when additionally interpolating the generated sample sequence, the interpolator filter is -π to π There is a problem that is impossible to implement because an ideal filter is required to pass up to, and when using a filter cut in a certain length for implementation, the quality of the interpolation filter output signal is caused by large frequency response components that can occur in the frequency band outside the pass band. There is a problem that gets worse. In accordance with the present invention for solving such a problem, IDFT is performed by applying a 64-point or 128-point IDFT unit 120, etc., which is larger than 32 points, and samples through IDFT results, poly-phase filter operation, and phase shift association Heat can be generated and interpolation can be performed using a simple interpolator filter.

In the case of applying the 64 point or 128 point IDFT unit 120, all 32 sub-channels for transmitting data are allocated to the frequency domain of 32 points corresponding to the low frequency. That is, the sub-channel corresponding to the 16 negative frequencies is input to the IDFT unit 120 point corresponding to the low frequency of the negative, and the sub-channel corresponding to the 16 positive frequencies is at a low frequency including the DC component. As an input to the corresponding IDFT unit 120 point, other input points of the IDFT unit 120 set a value of 0 to perform an IDFT operation. The result of the IDFT unit 120 generates a modulator output signal through a poly-phase filter and a phase shift block operation. When using a 64 point IDFT, it should be set to the values of M = 64 and K = 72, and when using a 128 point IDFT, it should be set to the values of M = 128 and K = 144.

When comparing the case of applying the above 64 point IDFT unit 120 and the 128 point IDFT unit 120, when applying the 128 point and applying the same filter length, when designing an interpolation filter, a filter design with a small outer curve of signal components It is possible, and when the 64 point IDFT is applied, the harmonic component is greater than 80 dB, but in the case of the 128 point IDFT, the harmonic component is less than 80 dB, and the output signal quality of the interpolation filter is also excellent.

However, when 64 point IDFT unit 120 is applied, it is the same as 72 times oversampling per symbol, and when 128 point IDFT unit 120 is applied, it is the same as 144 times oversampling per symbol, which increases implementation complexity. There are disadvantages.

The multicarrier modulator according to the present invention, which has increased the L-time oversample rate, uses the same parameters such as the same symbol rate and the interval between sub-channels, and differs only in the number of sub-channels used, and the number of sub-channels has an integer multiple relationship. At this time, it can be efficiently implemented by using a multi-carrier modulator with an increased L-time oversample rate of the present invention.

A specific example is in Example 2 below.

<Example 2>

Embodiment 2 relates to implementation of a modulator when a frequency bandwidth used for data transmission is different in a communication system using multi-carrier modulation. In this case, in the implementation of the modulator, a modulator is generated according to the first embodiment based on the multi-carrier transmission method using the largest bandwidth among the bandwidths used in data transmission. To generate a modulated signal.

As a specific example, a system that performs digital communication in the maritime VHF band transmits data using 32 subchannels, or 16 subchannels or 8 subchannels, which are separated by subchannels spaced at 2.7kHz. In all three cases, the subchannel considers the case of transmitting data having a symbol rate of 2400 baud.

At this time, since the maximum number of subchannels is 32, both the 64 point IDFT and the 128 point IDFT are used as in the previous embodiment 1, but in this description, when the 64 point IDFT is used, It is limitedly explained. When 32 subchannels are used, a 64point IDFT is used to generate an implementation structure of multicarrier modulation according to the present invention to generate a data transmission signal using 32 subchannels. When 16 subchannels are used, the multicarrier modulation implementation in case of using the 32 subchannels generated earlier is used, and only 16 points corresponding to the lower frequency among 64 IDFT input points are used as input values of the subchannels. Then, input 0 for the remaining points, perform 64 point IDFT operation, and perform the same operation afterwards. In case of using 8 subchannels, the multicarrier modulation implementation in case of using the 32 subchannels generated earlier is used, and only 8 points corresponding to the lower frequency among 64 IDFT input points are used to input the value of the subchannel. Then, input 0 for the remaining points, perform 64 point IDFT operation, and perform the same operation afterwards.

When the frequency bandwidth used for data transmission is different in a communication system using multi-carrier modulation, and two or more data transmissions are not simultaneously performed, the above method is applied to apply one multi-carrier. Data transmission is possible using only the carrier modulator, thereby reducing the complexity of the implementation.

<Example 3>

In the third embodiment, when the transmission data is input to the modulator, the transmission is performed with a delay proportional to about 1/2 of the filter length, but in burst packet communication, the first signal must be transmitted within a specific time in a predetermined time slot. When the first symbol of a packet starts, when all subchannels must have the same phase at 0 degrees, a method of enabling this is possible.

이 첫 번째 부필터의 폴리페이스 부필터에 위치하고, 첫 번째 부필터 계수중 가장 큰 값을 갖는 h(0)가 부필터의 가운데 위치하고, 그 위치를 FL/2라 한다. 여기서, FL은 폴리페이스 필터의 부필터의 필터계수의 길이로 짝수로 가정한다. 또한, 통신시스템에서 요구하는 상승시간이 N_R 심볼구간이라고 가정한다.A set containing a polyface filter composed of M sub-filters and a discretized sub-filter coefficient corresponding to h (0) having the largest value in a circular filter when oversampling per K symbol

It is located in the polyface sub-filter of the first sub-filter, h (0) having the largest value among the first sub-filter coefficients is located in the middle of the sub-filter, and its position is called FL / 2. Here, FL is assumed to be an even number of the filter coefficients of the sub-filter of the polyface filter. In addition, it is assumed that the rise time required by the communication system is the N _R symbol period.

In order to remove the delay caused by filtering under the above conditions, the initial parameter values of the multiple modulation method are set and operated as follows.

이고,

으로 계산되는 값이다.At this time, before the first symbol is generated, the number of symbols (FL / 2-N _R ) is input to the modulator in advance, and the IDFT unit 120 performs calculation, and the calculated data is stored in the memory unit 130. To save. Thereafter, filtering is performed by sequentially matching the first sub-filter of the polyface filter, starting from the P-th sub-memory of the memory 130. here,

ego,

It is calculated as.

의 위상을 시작 위상으로하여 상기 필터링된 결과와 계산된 위상회전량를 승산하여 출력하고, 위상은 앞에서 설명된 바와 같이 필터링된 결과와 곱할때마다

만큼의 위상을 증가시킨다. In the phase shift unit 170

Each phase is multiplied and output by multiplying the filtered result and the calculated phase rotation amount by using the phase of as the starting phase, and the phase is multiplied by the filtered result as described above.

Increase the phase by

By setting the initial value as described above, an output signal corresponding to the first symbol length is generated, and then, according to the operation procedure of the multicarrier transmission system described above, its truncation is repeated until the packet symbol transmission is completed.

In Example 3, the case where the sub-filter length of the polyface filter is even is described, but it can be applied even when the number is odd. For odd numbers, use floor (FL / 2) instead. Here, floor (x) means the largest integer not greater than x.

110: signal generation processing unit 120: IDFT unit
130: memory unit
140: poly-phase filter coefficient unit
160: filtering unit 170: phase shift unit

Claims (11)

delete delete delete delete delete delete delete delete delete delete In the method of transmitting the transmission information in a multi-carrier transmission method,
end. The IDFT unit having an extended point than the number of sub-channels to be applied to communication corresponds to a low frequency based on an intermediate value of the communication frequency, and inputs a real symbol signal to each point corresponding to the sub-channel, and the middle of the communication frequency. Inputting a zero (0) value to a point corresponding to the high frequency based on the value;
I. An inverse discrete Fourier transform of signals input to each point in the IDFT unit, and outputting each point;
All. Performing filtering by matching the signal output from the IDFT unit with a corresponding filter coefficient for each point;
la. Including the step of performing a phase shift so that the DC component is not included in the filtered signal;
Steps a and b above
A set containing a polyface filter composed of M sub-filters and a discretized sub-filter coefficient corresponding to h (0) having the largest value in a circular filter when oversampling per K symbol

Located in the polyface sub-filter of the first sub-filter, h (0) having the largest value among the first sub-filter coefficients is positioned in the center of the sub-filter, and the position of the filter coefficient of the sub-filter of the polyface filter If the length FL is even, FL / 2 or FL is odd, then floor (FL / 2) is determined. When the rising time required by the communication system is the N _R symbol interval, before the first symbol is generated (FL / 2- N _R ) The number of symbols is input in advance to perform the operation in the IDFT unit, and the calculated data is stored in the memory unit.
The multi-step performs filtering by sequentially matching the first sub-filter of the polyface filter starting from the P-th sub-memory of the memory unit,
The above step is

Each phase is multiplied and output by multiplying the filtered result and the calculated phase rotation amount by using the phase of as the starting phase.

Increases the phase by
remind

Is the above

Multi-carrier transmission method transmission method characterized in that the calculated value.

Images