JPH0542855B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to an FM stereo demodulation method in which stereo demodulation is performed using sample and hold in an FM stereo receiver.

Conventional configuration and its problems When receiving stereo broadcasts with an FM stereo receiver, the ideal frequency discriminator output, ie, the stereo composite signal Si(t), is as shown in equation (1).

Si(t)=(L+R)+ _Psinωs /2t+(L-R) _sinωst ...(1) However, L+R: Main signal (L-R) _sinωst : Sub-signal _Psinωs /2t: Pilot signal L :Left channel audio signal R:Right channel audio signal P:Pilot signal amplitude _ωs :Subcarrier angular frequency (=2π×38KHz) Currently, this method is used to separate left and right audio signals from the above composite signal (stereo demodulation). The switching method is the mainstream. This extracts the pilot signal from the signal shown in equation (1), inputs it to the phase comparator that forms part of the PLL, causes the PLL to generate a switching signal synchronized with the pilot signal, and then converts it into equation (1). In this method, the pilot signal component is removed from the signal by a trap circuit or the like, and the remaining signal is separated into left and right channel audio signals by switching the signal using the switching signal (rectangular wave).

However, in the above conventional example, when a stereo composite signal that is not in an ideal state is input,
The disadvantage is that sufficient stereo separation cannot be obtained.
Another drawback was that it was difficult to integrate.

OBJECT OF THE INVENTION The present invention eliminates the above-mentioned conventional drawbacks, and provides a system that can obtain a practically sufficient degree of stereo separation even when a stereo composite signal that is not in an ideal state is input, and that is easy to integrate. This provides an FM stereo demodulation method.

Composition of the Invention In order to achieve the above-mentioned object, the present invention removes a pilot signal from a stereo composite signal without using a trap circuit, and adds a signal obtained by subjecting the output of a sample and hold circuit to a certain processing to the composite signal. The present invention provides a stereo demodulation method using sample and hold, which performs the following and provides a sufficient degree of separation even when the stereo composite signal is not in an ideal state.

DESCRIPTION OF EMBODIMENTS First, the principle of the present invention will be explained. The stereo composite signal of equation (1) is expressed as: t ₁ =2π/ω _s・(2n+1/4) ……(2) t ₂ =2π/ω _s・(2n+5/4) ……(3) t ₃ = 2π/ω _s・(2n+3/4) ...(4) t ₄ =2π/ω _s・(2n+7/4) ...(5) However, at the timing when n=0, 1, 2, 3,... If you hold each sample,
The outputs Si ₁ , Si ₂ , Si ₃ , and Si ₄ are (6), (7), (8), respectively.
(9) becomes as follows. (However, the signal component with an angular frequency of ω _s /2 and its harmonic components are ignored.) Si ₁ = 2L + P / √2 ... (6) Si ₂ = 2L - P / √2 ... (7) Si ₃ = 2R + P / √2 ... (8) Si ₄ = 2R - P / √2 ... (9) Pilot signal component in the signal shown in equations (6) to (9)
Perform the following calculation to find Sip.

Sip=(Si ₁ −Si ₂ +Si ₃ −Si ₄ )/4=P/√2……(10
) (Sip=(Si ₁ − Si ₂ )/2 or Sip=(Si ₃ −
Si ₄ )/2 is also possible) In order to remove the pilot signal component from the signals Si ₁ to _{Si 4} , the following calculation is performed.

Si ₅ = Si ₁ - Sip = (2L + P / √2) - P / √2 = 2L ... (11) Si ₆ = Si ₂ + Sip = (2L - P / √2) + P / √2 = 2L ... ( 12) Si ₇ = Si ₃ −Sip = (2R + P / √2) − P / √2 = 2R ... (13) Si ₈ = Si ₄ + Sip = (2R - P / √2) + P / √2 = 2R ... ...(14) By alternately sampling and holding the signals Si ₅ and Si ₆ shown in equations (11) and (12) at a 1/ _s period, the left channel audio signal can be expressed as shown in equations (13) and (14) again. Signals Si ₇ , Si ₈
The right channel audio signals are obtained by alternately sampling and holding the signals at a 1/ _s period.

However, stereo composite signals are generally
It is not ideal like equation (1), but roughly looks like equation (15).

S(t)=(L+R)+Psinω _s /2t+ (L′−R′)sin(ω _s t+) ……(15) However, L′: Left channel audio signal in the sub signal R′: Left channel audio signal in the sub signal Right channel audio signal: Amount of subcarrier phase deviation Other conditions are the same as in equation (1) In this case, the stereo composite signal S is processed using the sample hold timing controlled as described above.
Signal S ₁ ~ obtained by sampling and holding (t)
S ₄ is each S ₁ = (L + L'cos) + (R - R'cos) + P/√2 ... (16) S ₂ = (L + L'cos) + (R - R'cos) - P/√ 2 ...(17) S ₃ = (L-L'cos) + (R+R'cos) + P/√2 ...(18) S ₄ = (L-L'cos) + (R+R'cos) - P/ √2...(19)

A pilot signal component _SP is determined from the signals S ₁ to _{S 4} .

S _P = (S ₁ - _{S 2} + S ₃ - S ₄ )/4 = P/√2 ... (20) (S _P = (S ₁ - S ₂ )/2 or S _P = (S ₃ - S ₄ )/2 is also possible) In order to remove the pilot signal component from the signals _S1 to _S4 , the following calculation is performed.

S ₅ = S ₁ −S _P = {(L+L′cos)+(R−R′cos)
+P/√2}-P/√2=(L+L'cos)+(R-R
′cos)
...(21) S ₆ = S ₂ + S _P = {(L+L′cos)+(R−R′cos)
−P/√2}+P/√2=(L+L′cos)+(R−R
′cos)
...(22) S ₇ = S ₃ −S _P = {(L−L′cos)+(R+R′cos)
+P/√2}-P/√2=(L-L′cos)+(R+R
′cos)
……(23) S ₈ = S ₄ + S _P = {(L−L′cos)+(R+R′cos)
−P/√2}+P/√2=(L−L′cos)+(R+R
′cos)
...(24) This result shows that the right channel audio signal is present in the signals S ₅ and S ₆ that should be only the left channel audio signal, and the left channel audio signal is present in the signals S ₇ and S ₈ that should be only the right channel audio signal. However, the stereo separation becomes insufficient. As a countermeasure for this, the following process is performed.

That is, _the signal K・S ₈ or KS ₇ obtained by multiplying the signal S 8 or S ₇ by a constant number K ₁ is obtained from the signals S ₅ and S ₆ ,
Also, the signal K ₂ obtained by multiplying the signal S ₅ or S ₆ by a constant number K ₂
S ₅ or K ₂ S ₆ is subtracted from the signals S ₇ and S ₈ , respectively.

S ₉ =S ₅ −K ₁ S ₈ , or S ₉ =S ₅ −K ₁ S ₇ …(25) S ₁₀ =S ₆ −K ₁ S ₇ , or S ₁₀ =S ₆ −K ₁ S ₈ … …(26) S ₁₁ = S ₇ −K ₂ S ₅ , or S ₁₁ = S ₇ −K ₂ S ₆ …(27) S ₁₂ = S ₈ −K ₂ S ₆ , or S ₁₂ = S ₈ −K ₂ S ₅ ……(28) Here, K ₁ and K ₂ are respectively K ₁ = R−R′cos/R+R′cos……(29) K ₂ =L−L′cos/L+L′cos ……( 30) Then, S ₉ to S ₁₂ in equations (25) to (28) above are each S ₉ = (L + L'cos) (1-K ₁ ) ... (25)' S ₁₀ = (L + L'cos) (1-K ₁ ) ...(26)' S ₁₁ = (R+R'cos) (1-K ₂ ) ......(27)' S ₁₂ = (R+R'cos) (1-K ₂ ) ...(28 )′, and by sampling and holding S ₉ and S ₁₀ alternately at a 1/ _s period as before, only the left channel audio signal is sampled and holding S ₁₁ and S ₁₂ alternately at a 1/ _s period. As a result, only the right channel audio signal can be separated and obtained. In other words, stereo demodulation can be performed. Also, the above (29), (3
K ₁ and K ₂ in equation 0) can also be simplified as K ₁ =K ₂ =K.

Next, a specific configuration of the present invention will be explained with reference to FIG.

In FIG. 1, 1 is a low-pass filter for extracting only the stereo composite signal shown in equation (15) from the frequency discriminator output, and 2 to 5 are stereo composite signals that are the outputs of the low-pass filter 1, respectively. The pilot signal (see Figure 2a, however, the stereo composite signal is shown divided into a pilot signal and a main signal + sub signal) is shown in Figures 2b to e.
Sample hold (sampling complete, hold start) is performed at the timing shown in equations (2) to (5) above using sample pulses SP1 to SP4 shown in
Signals S ₁ to _{S 4} shown in equations (16) to (19) (see Figure 2 f to i)
A sample and hold circuit 6 inputs the outputs of the sample and hold circuits 2 to 5, and outputs the sample and hold circuit (20).
A cancel signal generating circuit 7 outputs the pilot signal level S _P to be canceled by the calculation (including integral operation) shown in the formula, and 7 inputs the outputs of the sample and hold circuits 2 to 5,
SP ₁ , SP ₂ , SP ₃ , SP ₄ and SP ₁ ′, SP ₂ ′, SP ₃ ′,
Sample pulse generation circuit that outputs SP ₄ ′ (For the configuration and operation of this sample pulse generation circuit, see
For example, it is described in Japanese Patent Publication No. 58-23983. However, sample pulses SP ₁ ′ to SP ₄ ′ are each sample pulse SP ₁ to _{SP 4} for a certain period of time (≒
1/2 _s ) can be generated by delaying. ) 8 and 10 are the outputs S ₁ and S ₃ of the sample and hold circuits 2 and 4, respectively, and the cancel signal generation circuit output S _P
The signal S ₅ shown in equations (21) and (23) above is obtained by subtracting
The difference circuits 9 and 11 that output S ₇ are shown in equations (22) and (24) by adding the outputs S ₂ and S _{4 of the sample and hold circuits 3 and 5} and the cancel signal generation circuit _SP , respectively. A summation circuit 12 outputs the signals S ₆ and S ₈ , which outputs the signal S ₅ with the sample pulse SP ₃ , the signal S ₆ with the sample pulse SP ₄ , the signal S ₇ with the sample pulse SP ₂ , Said sample pulse SP ₁
This is a separability correction signal generation circuit which sequentially and alternately samples and holds the signal _S1 , and multiplies this output (see FIG. 2 j) by a constant number K to obtain a separability correction signal ΔS. is set so that the degree of separation between the left channel audio signal and the right channel audio signal, which will be described later, is maximized. 13 to 16 are each of the above signals
A difference circuit 17 that subtracts the separation degree correction signal ΔS from S ₅ to _{S 8} alternately samples the output of the difference circuit 13 with the sample pulse SP ₁ and the output of the difference circuit 14 with the sample pulse SP ₂ ′. A composite signal S _L (second
Sample and hold circuit that outputs (see figure o), 1
8 converts the output of the difference circuit 15 into the sample pulse
At SP ₃ ′, the output of the difference circuit 16 is sampled and held alternately using the sample pulse SP ₄ ′, and the output is obtained from the above (27)′.
A sample-hold circuit that outputs a composite signal S _R (see p in Figure 2) of the signals shown in equation (28)', 19, 2
0 is a low-pass filter that removes 38 KHz and its harmonic components from the signals S _L and S _R, respectively, and passes only the audio signal component.

In the above embodiment, in order to obtain the separability correction signal, the constant K by which the signals S ₅ , S ₆ and S ₇ , S ₈ are multiplied is set as shown in equations (29)' and (30)', Set the constant K ₁ and the correction constant K ₂ for the right channel separately to absorb variations in the gain of circuits 2 to 16 or variations in the pulse width and timing of sample pulses SP ₁ to SP ₄ and SP ₁ ′ to SP ₄ ′. It is also possible to do so. If the deterioration of the frequency characteristics of the left and right channel audio signal outputs caused by the stereo separation method described above is a problem, the signals S _L and S _R can be switched with pulses as shown in q and r in Figure 2, respectively, and the second Figure s,
PAM like t (PULSE AMPLITUDE
MODULATION) signal, each of the above LPF1
9 and 20, deterioration of frequency characteristics can be reduced.

Note that although the above embodiment is a case in which processing is performed in an analog manner using a sample and hold circuit, it goes without saying that this can be processed digitally using an entirely similar concept.

Effects of the Invention As described above, the stereo demodulation method according to the present invention has the advantage that a practically sufficient degree of stereo separation can be obtained even when a stereo composite signal in a non-ideal state is input, and also has the advantage of being easy to integrate. It is.

[Brief explanation of drawings]

FIG. 1 is a block diagram of a circuit implementing an FM stereo demodulation method according to an embodiment of the present invention, and FIG. 2 is an explanatory diagram of the operation of the circuit. 1...Low-pass filter, 2-5...Sample hold circuit, 6...Cancel signal generation circuit, 7...
... Sample pulse generation circuit, 8, 10 ... Difference circuit, 9, 11 ... Sum circuit, 12 ... Separation degree correction signal generation circuit, 13-16 ... Difference circuit, 17-18
...Sample hold circuit, 19,20...Low pass filter.

Claims (1)

[Claims] 1 Stereo composite signal S(t) S(t)=(L+R)+Psinω _s /2t+ (L'-R') sin(ω _s t+) However, (L+R): Main signal Psinω _s /2t: Pilot signal (L'-R') sinω _s t: Subsignal L, L': Left channel audio signal R, R': Right channel audio signal P: Pilot signal amplitude ω _s : Subcarrier angular frequency (= 2π×38KHz): The amount of phase deviation from the normal phase of the subcarrier, t ₁ = 2/ω _s (2nπ×π/4) t ₂ = 2/ω _s (2nπ×5/4π) t ₃ = 2 /ω _s (2nπ×3/4π) t ₄ =2/ω _s (2nπ×7/4π) However, the signal S obtained by sampling and holding each at the timing of n:, 1, 2, 3,... ₁ , _S2 , _S3 , and _S4 , the signal S _P S _P = (S ₁ − _{S 2} + S ₃ − S ₄ )/4 (or S _P = (S ₁ − S ₂ )/2 or S _P =( _S3 − _S4 )/
2) Find the signal from the signals S ₁ to S ₄ and the signal S _P.
S ₅ ~S ₈ S ₅ = S ₁ −S _P S ₆ = S ₂ +S _P S ₇ = S ₃ −S _P S ₈ = S ₄ +S _P is obtained, and signals S ₉ ~S are obtained from signals S ₅ ~S _{8 12} S ₉ = S ₅ −K ₁ S ₇ or S ₉ = S ₅ −K ₁ S ₈ S ₁₀ = S ₆ −K ₁ S ₈ or S ₁₀ = S ₅ −K ₁ S ₇ S ₁₁ = S ₇ −K ₂ S ₆ or S ₁₁ = S ₇ −K ₂ S ₅ S ₁₂ = S ₈ −K ₂ S ₅ or S ₁₂ = S ₈ −K ₂ S ₆ However, K ₁ K ₂ is a constant that can be adjusted externally. , by alternately sampling and holding the signals S ₉ and S ₁₀ , the left channel audio signal is
Further, an FM stereo demodulation method characterized in that by alternately sampling and holding the signals S ₁₁ and S ₁₂ , right channel audio signals are obtained respectively. 2. The FM stereo demodulation method according to claim 1, wherein K ₁ =K ₂ .