JP2685433B2 - Demodulation method - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a demodulation system, more specifically to a system for performing demodulation by digital signal processing, and more particularly to a digital phase modulation or quadrature amplitude modulation carrier band transmission. Regarding demodulation. [Prior Art] As a method for digitally demodulating a received signal modulated by the PSK or QAM method, IEE, transaction-on-communication, CMO 25,2 (1977 Year) Page 238 to 250 (IEEE, Transaction on Communication, COM
-25,2, (1977) pp238-250), the input band signal is passed through a Hilbert filter to obtain π
There is a method of obtaining two signal sequences that are / 2 shifted and are orthogonal to each other. In this method, the output of this filter is a carrier band signal because the Hilbert filter is extended to two orthogonal passband filters. Therefore, it is necessary to prepare a cos component and a sin component of the carrier frequency and make them into a base band by multiplication. In order to perform this operation digitally, the carrier wave cos is conventionally used.
The main method is to draw the ROM of the discrete values of each of the components and sin components and perform multiplication, and the amount of memory for points corresponding to the sampling frequency is required. In addition, an operation of applying waveform shaping filtering to the baseband signal is added, which increases the amount of arithmetic processing and tends to complicate the circuit configuration. [Problems to be Solved by the Invention] The above-mentioned conventional techniques are sufficient in terms of complexity of operation, circuit scale, phase characteristics of Hilbert filter, etc. No consideration was given. The object of the present invention is to introduce digital signal processing technology,
The digital demodulation method is provided in combination with digital technology such as carrier wave demodulation, separation of quadrature / synchronous components of the modulated signal in the base band, and further carrier wave position rotation correction, clock reproduction, waveform shaping filtering, etc. The goal is to achieve circuit miniaturization, LSI integration, and low power consumption by eliminating adjustment, improving performance, halving memory, and reducing the amount of computation. [Means for Solving the Problems] In order to achieve the above object, the operation of converting a carrier band signal into two orthogonal components in the base band is performed by directly sampling the carrier band signal and repeating it. This is achieved by focusing on the spectrum of the frequency band and negatively shifting the center frequency of this frequency band. The quadrature / in-phase component signals separated by the above processing are subjected to interpolation processing so as to be a signal synchronized with the bit rate of the modulation signal, whereby waveform shaping filtering, carrier phase rotation correction, clock extraction, etc. Can be processed at sample points. That is, the feature of the present invention is that the carrier desired band signal that is digital phase modulated or digital quadrature amplitude modulated is sampled at a predetermined frequency fs, and the discrete signal (fs, Nfs ...) At the time of sampling is 2 This is a demodulation method for obtaining I and Q components of a baseband signal by separating each of the separated discrete signals into a series and demodulating each of the separated discrete signals, wherein the carrier frequency fc of the carrier band signal is set to the modulation bandwidth (BW ), The sampling frequency fs is set to a frequency equal to or more than twice, and a frequency that is an integer multiple (N × fs) of the sampling frequency plus a frequency equal to or more than 1/2 of the modulation bandwidth, And the carrier frequency fc is Set so that the condition of
This is a demodulation method configured to enable digital processing from the discrete signal sampled by fs. According to a preferred embodiment of the present invention, the carrier frequency fc of the carrier band signal is set to the sampling frequency fs of the band signal.
The discrete signal obtained by the sampling is set to be alternately separated into two series one by one in time. Further, according to a preferred embodiment of the present invention, the carrier frequency fc is To obtain the I and Q components as baseband signals by alternately inverting the positive and negative signs of the two sequences one by one. [Operation] The principle of the present invention will be described below with reference to FIGS. 1 and 2. FIG. 2 is a block configuration diagram for performing the signal processing shown in FIG. It is assumed that the received signal r (t) has a spectrum as shown in FIG. 1 (a) in the frequency domain. The band indicated by the shaded area is considered to be the desired received signal. In Figure (a), the carrier frequency is an integer multiple of the sampling frequency fs (fs
× N) and a frequency fa equal to or more than ½ of the reception signal bandwidth BW are set. However, fc + BW / 2 <fs × N +
The condition of fs / 2 shall be satisfied. This is to prevent the spectrum folding from affecting the band BW when sampling with fs. A desired band signal is obtained by the band limiting filter 12 on the carrier wave described above. This output is shown in (b). Next, let this band signal be frequency fs
When the sampling is carried out at, the spectrum of the received signal appears repeatedly as shown in (c), and a desired band signal having the frequency fc (= fa) as the center frequency is obtained. This sampling is performed by the A / D converter 13. Consider the case where fc = fa = fs / 4. First, the discrete signals output from the A / D converter 13 are alternately separated into two series one by one. By this operation, it is considered that the two series of discrete signals are sampled at the frequency fs / 2 and the phase is shifted by π / 2 of the period of fs / 4. However, since these two series of discrete signals are still band signals having the center frequency fc, it is necessary to shift the frequency by (-fc) in order to convert them into base band signals. Considering the z region now, z = e ^jwT = e ^j2πf / fs ′ where fs ′ = fs / 2, and shifting (−fc) = fs / 4 is e ^{j2π (f + fs / 4)} / fs / 2 = e ^{j2π (fs / 4) / (fa / 2)} · e ^{j2πf / fs / 2)} = e ^jπ · e ^jwT = −z, which is equivalent to the conversion of z → −z. Therefore, when considering the time response, z ^{+ n} → (-z) ^-n = (-1) ^-n (z) ^-n = (-1)
ⁿ (z) ^-n, which is equivalent to alternating the positive and negative signs of the discrete signal. Using the above theory, two orthogonal components in the base band can be obtained. [Embodiment] An embodiment of the present invention will be described below with reference to FIG. FIG. 3 is a block diagram of a demodulator of a receiver in a carrier band transmission in which a quadrature component and an in-phase component of a carrier are phase-modulated by signals, that is, digital phase modulation or quadrature amplitude modulation. It is a figure. It is considered that the received signal is surrounded by unnecessary wave components in the frequency domain as shown in FIG. The carrier wave of the input received signal is fs / at an integral multiple of the sampling frequency fs after BPF12 in Fig. 3.
Set to the frequency with only 4 added. (Fc = Nfs + fs /
Four). It is considered to obtain a received signal from band signals arranged in this way. BPF12 must have sufficient passband to meet the received signal bandwidth, and must have stopband attenuation characteristics to the extent that aliasing due to sampling at fs does not adversely affect the received signal. This BPF12 can be realized with an analog filter. Next, output the BPF12 to the frequency fs by the A / D converter 13.
Sampling with. By sampling, the received signal is obtained as a band signal whose center frequency is fc. The received discrete signals are alternately separated into two series along the time series,
Positive and negative sign conversions are alternately performed by the sign converters 14 and 15 in each series. By the above operation, the quadrature component and the in-phase component of the modulated wave as the baseband signal are separated into two series. Next, if the discrete time period (2 / fs) of the signal, which is a component that is orthogonal to each other, is a sampling point that is not synchronized with the bit rate of the signal, it is necessary to obtain the sampled value at the synchronized time by interpolation. There is. Therefore, the interpolation 16 and 17 can be realized by the following waveform shaping filtering, carrier phase rotation correction, clock extraction, AGC, etc., or only by the sampling points for the identification points. [Effect of the Invention] According to the present invention, a quadrature component and an in-phase component in a base band can be obtained from a carrier band signal that is digitally phase-modulated or quadrature-amplitude-modulated by performing sampling and positive / negative sign conversion processing. It can be done, and all can be done by digital signal processing. In addition, waveform shaping filtering, phase correction, clock extraction, AGC, etc. can be realized with sample values only at the discrimination points by performing interpolation operations to obtain signals synchronized with the bit rate of modulated signals from discrete signals of baseband signals. . Therefore, all the troubles of phase adjustment such as in the case of realizing with an analog circuit, the level fluctuation, and the defect of the characteristic deterioration due to the element sensitivity are all solved, and the memory and the multiplier of the discrete value of the carrier wave like the conventional digital system are solved. It becomes useless. Therefore, it is economically and advantageous in terms of the number of parts.
It can contribute to the reduction of the chip area when performing I conversion.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a spectrum of a signal processing process of the basic method of the present invention, FIG. 2 is a block diagram of the signal processing of FIG. 1, and FIG. 3 is a block diagram of the present invention. FIG. 1 shows an embodiment and shows a demodulation method in a block configuration. 12 …… Bandwidth limiting filter (BPF), 13 …… A / D converter, 14,15 …… Sign converter, 16,17 …… Interpolator, 18,19 …… Wave shaping roll-off filter, 20 ……
Carrier wave phase rotation correction circuit, clock extraction circuit and AG
Block diagram of C circuit.

Continuation of front page    (72) Inventor Yasufumi Takahashi               1-280 Higashi Koigabo, Kokubunji-shi               Central Research Laboratory, Hitachi, Ltd.                (56) References IEEE Transactions                 on Communication               s, vol COM-25, No. 2, P               238-250 (1977)

Claims (1)

(57) [Claims] A carrier desired band signal that has been digitally phase-modulated or digital quadrature amplitude modulated is sampled at a predetermined frequency fs, the discrete signals (fs, Nfs ...) At the time of sampling are separated into two series, and the separated discrete signals are respectively separated. A demodulation method for demodulating to obtain I and Q components of a baseband signal, in which the carrier frequency fc of the carrier band signal is set to a frequency twice or more the modulation bandwidth (BW) of the carrier band signal. The carrier frequency fc is a condition that a frequency obtained by adding 1/2 or more of the modulation bandwidth to a frequency that is an integer multiple (N × fs) of the converted frequency fs, and that the carrier frequency fc is Set so that the condition of
A demodulation method characterized by being configured so that digital processing can be performed from a discrete signal sampled by fs. 2. The carrier frequency fc of the carrier band signal is set such that discrete signals obtained by sampling the band signal at the sampling frequency fs are alternately separated into two series one by one in time. The demodulation method according to claim 1. 3. The carrier frequency fc is 3. The demodulation system according to claim 2, wherein the I and Q components are obtained as baseband signals by alternately inverting the positive and negative signs of the two sequences one by one.