JPH07114342B2 - Digital frequency shift method - Google Patents

Description

FIELD OF THE INVENTION The present invention relates to a digital frequency shift method.

2. Description of the Related Art Such a digital frequency shift method is used in a circuit for converting a high frequency signal such as a television signal into a low frequency group and processing it.

FIG. 4 is a configuration example of a generally known digital frequency shift circuit, and FIG. 5 is a spectrum distribution diagram thereof. Here, (1) by multiplying the signal of the frequency (mu ₀₎ with respect to the input signal Ein, a shift circuit which performs the frequency (mu ₀₎ only the frequency shift,
(2) is a filter that extracts only a desired output from the multiplication outputs. In FIG. 5, (A) is an input signal group,
(B) (B ') indicates a multiplication output (sidebands), (F _s) represents the sampling rate.

In a digital frequency shift system in which a certain frequency group (A) is amplitude-modulated with a carrier (μ ₀ ) of a constant frequency and a desired sideband is obtained by a digital filter,
When the sampling rate is (F _s ), as shown in Fig. 6, F _s
When trying to shift the frequency by / 2 or more, handling a signal of F _s / 2 or more exceeds the limit of Nyquist, which causes problems such as noise. So in this case,
As shown in the spectrum distribution chart of FIG. 8, frequency shifting is performed by dividing the carrier (μ ₀ ) and (μ ₁ ) for F _s / 2 or less into two times. An example of a circuit in that case is shown in FIGS. This will be described with reference to FIG. 8. First, in the example of FIG. 7A, the input signal Ein
The frequency shift circuit (3) shifts the frequency with the shift carrier (μ ₀ ) and the unnecessary image spurious (B ′) is removed by the first filter (4). Next, frequency-shifted (B) is similarly frequency-shifted by the second frequency shift circuit (5) by the shift carrier (μ ₁ ). Then, the unnecessary image spurious (C ') is removed by the second filter (6), and the desired frequency-shifted (C) is taken out as the output signal Eout.

On the other hand, in the example of FIG. 7 (b), after the input signal Ein is continuously frequency-shifted by the first and second frequency shift circuits (3) and (5), unnecessary image spurious (C This is an example of a circuit in which the desired frequency shift signal (C) is taken out as the output signal Eout by removing '), (D) and (D').

Problems to be Solved by the Invention Comparing the above two circuit examples, the former has a drawback that the number of filters is large. The latter has one filter and seems to be advantageous, but referring to FIG. 8 (c), the (C) signal and the (D) signal are very close to each other, and the filter for separating them is realized. It can be said that it is impossible. If an attempt is made to realize if the multiplier to the input (in) as shown in Figure 9 in order to improve the characteristics _{(M 0) (M 1)} ...... (M n)
Is multiplied by a constant (k ₀ ) (k ₁ ) …… (k _n ), and its output is added to the output of the circuit (Z ⁻¹ ) and the adder (Y ₀ ) (Y ₁ ) …… (Y
_The digital filter configured to perform addition in (n) becomes a high-order (n is large), and the circuit scale becomes large and the signal delay also becomes large.

The present invention has been made in view of the above circumstances, and an object thereof is to provide a frequency shift method in which signal processing is simple and frequency shift can be stably performed.

Means for Solving the Problems In the present invention for achieving the above object, in a digital frequency shift method for frequency shifting a signal, when the frequency shift amount exceeds the limit of Nyquist, the frequency shift amount is subtracted from the sampling frequency. The frequency carrier is multiplied and the multiplication output is filtered to obtain a desired frequency-shifted signal.

As a result of multiplying the carrier of the frequency obtained by subtracting the frequency shift amount from the working sampling frequency, the upper sideband that is frequency-shifted and exceeds the Nyquist limit is folded back for sampling and appears as a relatively low frequency band. And
Filtering out the lower sideband provides the desired frequency shifted signal.

EXAMPLES Examples of the present invention will be described below with reference to the drawings. In this embodiment, as shown in FIG. 1, the frequency shift carrier is set to the frequency (F _s −μ) obtained by subtracting the desired frequency shift amount (μ) from the sampling frequency (F _s ) and the frequency of this carrier is shifted. The input signal (Ein) to be multiplied by is multiplied by the multiplier (11), and the multiplication output is selected by the filter (12). Both the input signal (Ein) and the carrier are discrete values sampled by the sampling clock.

Next, the principle of frequency shift in FIG. 1 will be described with reference to the spectrum distribution diagram in FIG. To apply a shift amount of a frequency higher than half the sampling frequency (μ) in a digital signal (A) as shown in FIG. 2 (a), the signal (F _s −
The multiplication output (B ') (B ") of the distribution shown in Fig. 2 (b) should be obtained by multiplying the carrier of (μ), but the upper sideband (B') exceeds the Nyquist limit. Second because
It is folded back to the position of (B) of FIG.

FIG. 3 illustrates how a frequency exceeding the Nyquist limit is folded back to a low frequency. FIG. 3A shows a waveform before sampling, the horizontal axis represents time, and (T)
Is the sampling period. The points (SP) on the waveform indicate sampling points. In Fig. 3 (a), if the sampling frequency is (F _s ), the frequency of the original waveform (W ₁ ) is 7 / 8F _s , but the points where the original waveform (W ₁ ) was sampled were connected. The frequency of the waveform (W ₂ ) in Fig. 3 (b) is 1 /
It is 8F _s . In this way, the frequency f that exceeds the Nyquist limit is folded back to the frequency (F _s −f). The present invention utilizes such folding, and is very effective when the frequency shift amount exceeds the Nyquist limit.

EFFECTS OF THE INVENTION As described above, according to the present invention, even in the frequency shift of a digital signal in which the amount of frequency shift exceeds the limit of Nyquist, the frequency shift can be satisfactorily performed with a very simple configuration. Further, cost, power consumption, circuit scale, data delay, etc. can be reduced.

[Brief description of drawings]

FIG. 1 is a block diagram of a frequency shift circuit embodying the present invention, FIG. 2 is a spectrum distribution diagram associated with its operation, and FIG. 3 is a principle explanatory diagram. FIG. 4 is a block diagram of a general frequency shift circuit, and FIG. 5 is its spectrum distribution diagram. FIG. 6 is a spectrum distribution diagram when the frequency shift amount exceeds the Nyquist limit in general. FIG. 7 is a block diagram showing a concrete configuration example of a conventional example, FIG. 8 is a spectrum distribution diagram when the frequency shift amount exceeds the limit of Nyquist, and FIG. 9 is a digital filter used in the conventional example. 3 is a block circuit diagram of FIG. (Μ) …… Frequency shift amount, (F _s ) …… Sampling frequency, (11) …… Multiplier, (12) …… Filter.

Claims (1)

[Claims] 1. A digital frequency shift method for frequency-shifting a signal, wherein when a frequency shift amount exceeds a limit of Nyquist, a sampling frequency is subtracted by a carrier of a frequency, and the multiplication output is filtered. A digital frequency shift method, characterized in that a desired frequency-shifted signal is obtained.