STEREO-PHONIC FREQUENCY MODULATION RECEIVER AND METHOD FOR SEPARATING DUAL SOUND CHANNELS

Abstract

A stereo-phonic frequency modulation receiver includes: a frequency modulation demodulation circuit, receiving a reception signal, and generating a demodulated signal according to the reception signal; a frequency-division demultiplexer, generating a sum signal, a difference signal and a pilot amplitude signal according to the demodulated signal; a dual sound channel separation circuit, generating a left-channel output signal and a right-channel output signal according to the sum signal and a weakened difference signal; and a weakening circuit, weakening the difference signal according to the pilot amplitude signal or a signal-to-noise ratio (SNR) to generate the weakened difference signal.

Claims

1. A stereo-phonic frequency modulation receiver, comprising: a frequency modulation demodulation circuit, receiving a reception signal, and generating a demodulated signal according to the reception signal; a frequency-division demultiplexer, generating a sum signal, a difference signal and a pilot amplitude signal according to the demodulated signal, wherein the sum signal is associated with a sum of a left sound channel and a right sound channel, and the difference signal is associated with a difference between the left sound channel and the right sound channel; a dual sound channel separation circuit, generating a left-channel output signal and a right-channel output signal according to the sum signal and a weakened difference signal; and a weakening circuit, coupled between the frequency-division demultiplexer and the dual sound channel separation circuit, weakening the difference signal according to the pilot amplitude signal or a signal-to-noise ratio (SNR) to generate the weakened difference signal.

2. The stereo-phonic frequency modulation receiver according to claim 1, wherein the weakening circuit comprises a switch coupled between the frequency-division demultiplexer and the dual sound channel separation circuit, and the weakening circuit controls the switch to be cutoff when the pilot amplitude signal is smaller than a threshold.

3. The stereo-phonic frequency modulation receiver according to claim 1, wherein the weakening circuit comprises: a decision circuit, generating a weakening coefficient according to the pilot amplitude signal and or the SNR; and a multiplier, multiplying the difference signal by the weakening coefficient to generate the weakened difference signal.

4. The stereo-phonic frequency modulation receiver according to claim 3, wherein the weakening circuit generates the weakening coefficient in a value 0 when the pilot amplitude signal is smaller than a threshold.

5. The stereo-phonic frequency modulation receiver according to claim 3, wherein the weakening coefficient decreases as the SNR lowers.

6. The stereo-phonic frequency modulation receiver according to claim 3, wherein the decision circuit stores a mapping table that stores a mapping relationship between the SNR and the weakening coefficient.

7. The stereo-phonic frequency modulation receiver according to claim 1, further comprising: an SNR calculation circuit, coupled to the frequency modulation demodulation circuit, calculating the SNR according to an amplitude signal generated by the frequency modulation demodulation circuit, comprising: a filter, receiving the amplitude signal and generating an estimated amplitude value; a subtraction circuit, subtracting the estimated amplitude value from the amplitude signal to generate an estimated SNR; and an SNR circuit, calculating the SNR according to the estimated amplitude value and the estimated SNR.

8. The stereo-phonic frequency modulation receiver according to claim 1, wherein the frequency modulation demodulation circuit comprises a superheterodyne receiver.

9. The stereo-phonic frequency modulation receiver according to claim 1, wherein the frequency-division demultiplexer comprises: a first low-pass filter (LPF), receiving the demodulated signal and generating the sum signal; a frequency multiplier, generating a frequency multiplied signal according to an oscillation signal; a band-pass filter (BPF), receiving the demodulated signal and generating a band-pass filtered result; a mixer, generating a mixed result according to the frequency multiplied signal and the band-pass filtered result; a second LPF, receiving the mixed result and generating the difference signal; and an amplitude circuit, generating the pilot amplitude signal according to the pilot signal.

10. A method for separating dual sound channels, applied to a stereo-phonic frequency modulation receiver, the method comprising: receiving a sum signal, a difference signal and a pilot amplitude signal from a frequency-division demultiplexer of the stereo-phonic frequency modulation receiver, wherein the sum signal is associated with a sum of a left sound channel and a right sound channel, and the difference signal is associated with a difference between the left sound channel and the right sound channel; weakening the difference signal according to the pilot amplitude signal or a signal-to-noise ratio (SNR) to generate a weakened difference signal; and generating a left-channel output signal and a right-channel output signal according to the sum signal and the weakened difference signal.

11. The method according to claim 10, wherein the step of weakening the difference signal according to the pilot amplitude signal or the SNR to generate the weakened difference signal comprises: generating a weakening coefficient according to the pilot amplitude signal or the SNR; and multiplying the difference signal by the weakening coefficient to generate the weakened difference signal.

12. The method according to claim 11, further comprising: generating the weakening coefficient in a value 0 when the pilot amplitude signal is smaller than a threshold.

13. The method according to claim 11, wherein the weakening coefficient decreases as the SNR lowers.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0009] FIG. 1 is a block diagram of a stereo-phonic frequency modulation receiver according to an embodiment of the present invention;

[0010] FIG. 2 is a block diagram of a weakening circuit according to an embodiment of the present invention;

[0011] FIG. 3 is a block diagram of a weakening circuit according to an embodiment of the present invention;

[0012] FIG. 4 is a schematic diagram of a spectrum of a demodulated signal;

[0013] FIG. 5 is a block diagram of a frequency-division demultiplexer according to an embodiment of the present invention;

[0014] FIG. 6 is a block diagram of an SNR calculation circuit according to an embodiment of the present invention;

[0015] FIG. 7 is a block diagram of a frequency modulation demodulation circuit according to an embodiment of the present invention; and

[0016] FIG. 8 is a flowchart of a method for separating dual sound channels according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

[0017] FIG. 1 shows a block diagram of a stereo-phonic frequency modulation receiver 10 according to an embodiment of the present invention. The stereo-phonic frequency modulation receiver 10 includes a frequency modulation demodulation circuit 100, a frequency-division demultiplexer (demux) 102, a dual sound channel separation circuit 104, a weakening circuit 106 and a signal-to-noise ratio (SNR) calculation circuit 108. The frequency modulation demodulation circuit 100 receives a reception signal r, and generates a demodulated signal m and an amplitude signal A according to the reception signal r. The frequency-division demultiplexer 102 receives the demodulated signal m, and generates a sum signal (L+R), a difference signal (LR) and a pilot amplitude signal A.sub.pilot according to the demodulated signal m. The SNR calculation circuit 108 receives the amplitude signal A, and generates an SNR according to the amplitude signal A. The sum signal (R+L) is received by the stereo-phonic frequency modulation receiver 10 and is associated with a sum of a left-channel signal L and a right-channel signal R. The difference signal (LR) is received by the stereo-phonic frequency modulation receiver 10 and is associated with a difference between the left-channel signal L and the right-channel signal R. Further, the amplitude signal A corresponds to an amplitude of the reception signal r, and the pilot amplitude signal A.sub.pilot corresponds to an amplitude of a pilot signal x.sub.p in the demodulated signal m.

[0018] The weakening circuit 106, coupled between the frequency-division demultiplexer 102 and the dual sound channel separation circuit 104, receives the difference signal (LR) and weakens the energy/amplitude of the difference signal (LR) to generate a weakened difference signal (LR). The dual sound channel separation circuit 104 generates a left-channel output signal L_out and a right-channel output signal R_out according to the sum signal (L+R) and the weakened difference signal (LR). The left-channel output signal L_out may be represented as L_out=(L+R)+(LR), and the right-channel output signal R_out may be represented as R_out=(L+R)(LR).

[0019] More specifically, the weakening circuit 106 generates a weakening coefficient c_wkn according to the pilot amplitude signal A.sub.pilot or the signal-to-noise ratio SNR, and multiplies the difference signal (LR) by the weakening coefficient c_wkn to generate the weakened difference signal (LR). FIG. 2 shows a block diagram of a weakening circuit 206 according to an embodiment of the present invention. The weakening circuit 206 realizes the weakening circuit 106, and includes a decision circuit 260 and a multiplier MP. The decision circuit 260 receives the pilot amplitude signal A.sub.pilot and the signal-to-noise ratio SNR, and generates the weakening coefficient c_wkn according to the pilot amplitude signal A.sub.pilot or the signal-to-noise ratio SNR. The multiplier MP multiples the difference signal (LR) by the weakening coefficient c_wkn to generate the weakened signal (LR), which may be represented by (LR)=c_wkn(LR), wherein the value of the weakening coefficient c_wkn may be between 0 and 1. It should be noted that, how the decision circuit 260 generates the weakening coefficient c_wkn according to the pilot amplitude signal A.sub.pilot or the signal-to-noise ratio SNR is not limited. In one embodiment, when the pilot amplitude signal A.sub.pilot is smaller than a threshold TH, the decision circuit 260 may generate the weakening coefficient c_wkn in a value 0, and the left-channel output signal L_out and the right-channel output signal R_out are both the sum signal (L+R) at this point. In another embodiment, the decision circuit 260 may generate the weakening coefficient c_wkn according to the signal-to-noise ratio SNR. More specifically, the decision circuit 260 may be stored with a look-up table LUT, in which a mapping relationship between the signal-to-noise ratio SNR and the weakening coefficient c_wkn is stored. When the decision circuit 260 receives the signal-to-noise ratio SNR, the decision circuit 260 may look up the weakening coefficient c_wkn corresponding to the signal-to-noise ratio SNR from the look-up table LUT, and output the weakening coefficient c_wkn to the multiplier MP. The mapping relationship between the signal-to-noise ratio SNR and weakening coefficient c_wkn may be adjusted based on system requirements or actual conditions, instead of being limited to a predetermined mapping relationship. For example, referring to Table I, when the signal-to-noise ratio SNR is greater than 30 dB, the decision circuit 260 may output the weakening coefficient c_wkn in a value 0.707; when the signal-to-noise ratio SNR is smaller than 18 dB, the decision circuit 260 may output the weakening coefficient c_wkn in a value 0.054; when the signal-to-noise ratio SNR is between 30 dB and 18 dB, the decision circuit 260 may output the weakening coefficient c_wkn that reduces as the signal-to-noise ratio SNR decreases.

TABLE-US-00001 TABLE I LUT SNR (dB) c_wkn >30 0.707 29 0.69966 28 0.55962 27 0.44766 26 0.3582 25 0.28656 24 0.22932 23 0.18342 22 0.1467 21 0.11736 20 0.09396 19 0.07506 18 0.06012 <18 0.054

[0020] It should be noted that, the weakening circuit is not limited to being implemented in the form shown in FIG. 2. FIG. 3 shows a block diagram of a weakening circuit 306 according to another embodiment of the present invention. The weakening circuit 306 is similar to the weakening circuit 206, and so the same elements use the same denotations. One difference from the weakening circuit 206 is that, the weakening circuit 306 further includes a switch SW. When the pilot amplitude signal A.sub.pilot is smaller than the threshold TH, a decision circuit 360 included in the weakening circuit 306 generates a switch control signal c_sw to cause the switch SW to be cutoff. At this point, the difference signal (LR) cannot be transmitted to the dual sound channel separation circuit 104, and so both of the left-channel output signal L_out and the right-channel output signal R_out are the sum signal (L+R). When the pilot amplitude signal A.sub.pilot is greater than the threshold TH, the decision circuit 360 generates the switch control signal c_sw to cause the switch SW to be conducted, and the decision circuit 360 may then generate the weakening coefficient c_wkn according to the signal-to-noise ratio SNR. Associated operation details of the decision circuit 306 are similar to those of the decision circuit 260, and shall be omitted herein.

[0021] Signal components and spectrum of the demodulated signal m are described in detail below. For a stereo-phonic frequency modulation system (given that channel response and noise are omitted), the time function of the demodulated signal m may be represented as m(t)=[L(t)+R(t)]+x.sub.p(t)+[L(t)R(t)] cos(22f.sub.pt+2.sub.p), where L(t) and R(t) respectively represent the time functions of the left-channel signal L and the right-channel signal R, x.sub.p(t) is the time function of the pilot signal x.sub.p, the pilot signal x.sub.p(t) may be represented as x.sub.p(t)=A.sub.p cos(2f.sub.pt+.sub.p), A.sub.p, f.sub.p and .sub.p respectively represent the amplitude, frequency and phase of the pilot signal x.sub.p(t), L(t)+R(t) represents the time function of the sum signal (L+R), and L(t)R(t) represents the time function of the difference signal (LR). Further, the spectrum of the demodulated signal m(t) may be as shown in FIG. 4. In general, the spectrum of L(t)+R(t) is substantially located between 0 KHz and 15 KHz, the spectrum of the pilot signal x.sub.p(t) is substantially located at 19 KHz, and the spectrum of [L(t)R(t)] cos(22f.sub.pt+2.sub.p) is substantially located between 23 KHz and 53 KHz.

[0022] Details of the frequency-division demultiplexer 102 are given with reference to FIG. 5. FIG. 5 shows a block diagram of the frequency-division demultiplexer 102 according to an embodiment of the present invention. The frequency-division demultiplexer 102 includes a low-pass filter (LPF) 500, a band-pass filter (BPF) 502, a mixer 504, an LPF 506, a pilot filter 508, a frequency multiplier 510 and an amplitude circuit 512. The LPF 500 filters out the signal L(t)+R(t) located at 0 KHz to 15 KHz from the signal m(t) to obtain the sum signal (L+R). The pilot filter 508 filters out the pilot signal x.sub.p(t) from the signal m(t) to obtain the pilot signal x.sub.p. The amplitude circuit 512 receives the pilot signal x.sub.p, and generates the pilot amplitude signal A.sub.pilot corresponding to the pilot signal x.sub.p. For example but not limited to, the amplitude circuit 512 may obtain the pilot amplitude signal A.sub.pilot by a coordinate rotation digital computer (CORDIC) algorithm according to the pilot signal x.sub.p. The BPF 502 filters out the signal [L(t)R(t)] cos(22f.sub.pt+2.sub.p) located at 23 KHz to 53 KHz from the signal m(t) to generate a band-pass filtered result R_BPF. The frequency-division demultiplexer 102 generates a frequency multiplied signal (substantially located at 38 KHz) of the pilot signal x.sub.p by the frequency multiplier 510, generates a mixed result R_MX from multiplying the band-pass filtered result R_BPF by the frequency multiplied signal by the mixer 504, filters out high-frequency components from the mixed result R_MX by the LPF 506 to obtain the difference signal (LR). In other words, the mixer 504 and the LPF 506 relocate [L(t)R(t)] cos(22f.sub.pt+2.sub.p) to a frequency band of baseband to obtain the difference signal (LR).

[0023] Details of the SNR calculation circuit 108 are given with reference to FIG. 6 below. FIG. 6 shows a block diagram of the SNR calculation circuit 108 according to an embodiment of the present invention. The SNR calculation circuit 108 includes a filter 600, a subtraction circuit 602 and an SNR circuit 604. The filter 600 is an LFP, which receives the amplitude signal A and performs a low-pass filter/average calculation on the amplitude signal A to filter out high-frequency components from the amplitude signal A to generate an estimated amplitude value A.sub.est. The subtraction circuit 602 subtracts the estimated amplitude value A.sub.est from the amplitude signal A to generate an estimated SNR n.sub.est. The SNR circuit 604 may include a magnitude unit 606, a filter 608 and a ratio unit 610. The magnitude unit 606 performs a magnitude calculation on the estimated SNR n.sub.est to obtain an SNR magnitude value |n.sub.est|. The filter 608 is also an LPF, which performs a low-pass filter/average calculation on the SNR magnitude value |n.sub.est| to obtain a noise average |n.sub.ave|. The ratio unit 610 receives the estimated amplitude value A.sub.est and the noise average |n.sub.ave| to calculate the signal-to-noise ratio SNR, which may be represented as SNR=|A.sub.est|.sup.2/|n.sub.ave|.sup.2. Thus, the SNR calculation circuit 108 outputs the calculated signal-to-noise ratio SNR to the weakening circuit 106 that then generates the weakening coefficient c_wkn according to the signal-to-noise ratio SNR.

[0024] Details of the frequency modulation demodulation circuit 100 are given with reference to FIG. 7 below. FIG. 7 shows a block diagram of the frequency modulation demodulation circuit 100 according to an embodiment of the present invention. The frequency modulation demodulation circuit 100 may be a superheterodyne receiver, and includes a radio-frequency (RF) filtering and amplification circuit 700, an RF mixer 702, an intermediate-frequency (IF) filtering and amplification circuit 704, an IF mixer 706 and a demodulation circuit 708. The RF filtering and amplification circuit 700 filters out signal components at an RF band from the reception signal, and amplifies the signal components of the RF band to generate an RF signal y.sub.RF. The RF mixer 702 and the IF filtering and amplification circuit 704 relocates the RF signal y.sub.RF of the RF band to an IF band using a first oscillation signal LO_1 to generate an IF signal y.sub.IF. Further, the IF mixer 706 relocates the IF signal y.sub.IF to a band of baseband using a second oscillation signal LO_2. The demodulation circuit 708 then accordingly demodulates the signal m and the amplitude signal A.

[0025] It should be noted that, when the signal quality of the pilot signal x.sub.p is unsatisfactory (i.e., the pilot amplitude signal A.sub.pilot is smaller than the threshold TH) or the signal-to-noise ratio SNR is poor, the signal [L(t)R(t)] cos(22f.sub.pt+2.sub.p) cannot be effectively relocated to the band of baseband and a frequency offset is thus incurred. In this situation, when the signal quality of the pilot signal x.sub.p or the signal-to-noise ratio SNR of the pilot signal x.sub.p is unsatisfactory, the present invention uses the weakening circuit 106 to adaptively weaken (or even block) the difference signal (LR), so as to reduce the effect of the difference signal (LR) (having a frequency offset) upon the sum signal (L+R) to further enhance the separation between the left and right sound channels. Proven experimentally, the stereo-phonic frequency modulation receiver 10 achieves 40.6 dB separation between the left and right sound channels, whereas only 20 dB separation between left and right sound channels is provided by a conventional stereo-phonic frequency modulation receiver. In other words, the stereo-phonic frequency modulation receiver 10 achieves better separation between left and right sound channels.

[0026] One person skilled in the art can understand that, the function units/circuits in FIG. 1 to FIG. 3 and FIG. 5 to FIG. 7 may be realized or implemented by digital circuits (e.g., RTL circuits) or a digital signal processor (DSP), and associated details are omitted herein.

[0027] Operation details of the stereo-phonic frequency modulation receiver 10 may be concluded into a dual sound channel separation process. FIG. 8 shows a flowchart of a dual sound channel separation method 80 according to an embodiment of the present invention. The dual sound channel separation method 80 may be performed by the stereo-phonic frequency modulation receiver 10, and includes following steps.

[0028] In step 800, the frequency modulation demodulation circuit 100 generates the demodulated signal m according to the reception signal.

[0029] In step 802, the frequency-division demultiplexer 102 generates the sum signal (L+R), the difference signal (LR) and the pilot amplitude signal A.sub.pilot according to the demodulated signal m.

[0030] In step 804, the weakening circuit 106 generates the weakening coefficient c_wkn according to the pilot amplitude signal A.sub.pilot or the signal-to-noise ratio SNR.

[0031] In step 806, the weakening circuit 106 multiples the difference signal (LR) by the weakening coefficient c_wkn to generate the weakened difference signal (LR).

[0032] In step 808, the dual sound channel separation circuit 104 generates the left-channel output signal L_out and the right-channel output signal R_out according to the sum signal (L+R) and the weakened difference signal (LR).

[0033] Operation details may be referred from the description in the above paragraphs, and shall be omitted herein.

[0034] Therefore, as demonstrated, the present invention adaptively weakens the difference signal when the signal quality of the signal-to-noise ratio of a pilot signal is unsatisfactory to reduce the effect of the difference signal having a frequency offset, and to further enhance the separation between the left and right sound channels.

[0035] While the invention has been described by way of example and in terms of the preferred embodiments, it is to be understood that the invention is not limited thereto. On the contrary, it is intended to cover various modifications and similar arrangements and procedures, and the scope of the appended claims therefore should be accorded the broadest interpretation so as to encompass all such modifications and similar arrangements and procedures.