KR20000068083A - Receiver for independent sideband signals - Google Patents

Abstract

The receiver for the independent sideband signal alternately contains a complex FIR filter containing each real lowpass filter whose coefficients (C ₁ to C _N-1 , C ₀ to C _N ) are nonzero Means 30-54 for making digitized orthogonal zero IF forms of the received signal applied to the structure 56. The FIR filter structure is described in more detail in FIG. 3 (not shown). Each of the upper and lower bands (USB, LSB) is recovered by finding the sum and difference of the respective filter outputs. Each sideband signal USB, LSB is stored in RAM 58 and is expanded, equalized, and converted into an analog signal supplied to audio converter 64 when the stored signal needs to be reproduced.

Description

Receiver for independent sideband signals

An independent sideband signal (ISB) consists of two independent single sideband signals. Thus, in its baseband representation, the ISB signal is a complex signal. Only complex representations distinguish negative frequencies from positive frequencies. Sometimes only one of the sidebands is involved at the same time. Traditionally, three different methods exist in the analog domain to obtain one of the sidebands as a real signal. Three methods are the filter method, the paging method, and the Weaver method. In the digital domain, the equivalence of the phase method usually applies. The required broadband 90 ° shift can be achieved by the Hilbert transform.

The Hilbert transform shifts the positive frequencies by -90 ° (inducing backwards) and the negative frequencies by 90 ° (inducing leading edges). The imaginary part of the complex signal lags behind the real part by 90 °. If the imaginary part of the complex signal is Hilbert transformed, the negative frequencies show 0 ° phase shift while its positive frequencies show -180 ° phase shift. Adding the original real part of the signal to the Hilbert transformed part cancels the positive frequency represented by the upper side band (USB) in the baseband representation. The negative frequencies represented by the lower side band (LSB) are structurally summed. Similarly, if the Hilbert transformed imaginary part is subtracted from the real part, the LSB cancels out and USB is obtained. In practice, the delay induced in the actual implementation of the Hilbert transform must be compensated for in the real signal path. The symmetric N ^th order FIR approach has a constant group delay of N / 2, which can be easily derived from the real line path.

1 of the accompanying drawings illustrates this approach based on the Hilbert transform FIR filter 10. In Figure 1, the orthogonal real and imaginary sideband signals I and Q are applied to each in-phase and quadrature phase path, respectively.

The Hilbert transform FIR filter 10 is provided in the quadrature phase path, each of which induces a delay of Z ⁻¹ , with N in series connected delay steps D1, D2, D3, D4 ... D (N-2), D (N-1), D (N)). The multipliers M0, M1, M2, M3 ... M (N-2), M (N-1), M (N) are input of the first delay step D1 and all delay steps D1 to D ( N)) is connected to each output. The coefficients c ₀ , c ₁ , c ₂ , c ₃ ... c _(N-2) , c _(N-1) , c _N are applied to the multipliers M0 to M (N), respectively.

The products made by the multipliers M0 to M (N) are added at summing means 12 and appear at the output 14.

The in-phase path includes a delay device 16 for delaying the real I signal by N / 2.

The signal at output 14 is added in signal combining step 18 with the signal from delay device 16 to provide the LSB signal filtered at band pass filter 22. The signal at output 14 is subtracted from the signal at delay device 16 in signal subtraction step 20 to provide a USB signal filtered by band pass filter 24.

Alternatively, the Hilbert transform may be made at the real part of the signal instead of the imaginary part of the signal. The outputs for the LSB and USB are then swapped. The order N of the Hilbert transform FIR filter is not desired but depends on the required sideband suppression. The ideal Hilbert transform is somewhat similar to an ideal brickwall filter that can only be accurately approximated by selecting N large enough. The approximation is particularly near 0 Hz, with the phase response jumping from 90 ° to -90 °. This means that the suppression of unwanted sidebands for low frequencies is not very high compared to the rest of the band.

Band pass filters 22 and 24 are needed to suppress out-of-band noise and any residual carrier noise.

The present invention relates to a receiver for receiving and demodulating independent sideband signals, which may include compressed analog speech samples. Such a receiver may be used in a voice paging system.

1 is a schematic block diagram of a conventional method for demodulating an ISB signal.

2 is a schematic block diagram of one embodiment of a receiver made in accordance with the present invention;

3 is a schematic block diagram of an FIR filter arrangement for demodulating each sideband.

4 is a spectrum of an ISB speech signal.

5A, 5B, and 5C are graphs illustrating characteristics of examples of a low pass filter derived from a low pass filter coefficient, a USB filter, and an LSB filter, respectively.

Like reference numerals in the drawings are used to indicate like parts.

It is an object of the present invention to reduce the complexity of restoring the sidebands of an ISB signal.

In accordance with one aspect of the present invention, a receiver for an independent sideband (ISB) signal is provided, the receiver comprising: an input for a modulated ISB signal, and means for providing an orthogonal zero IF ISB signal; Demodulation means for demodulating the zero IF ISB signal to generate an upper sideband signal and a lower sideband signal, respectively, wherein the demodulation means alternately comprise first and second real digital filters having non-zero coefficients; Coupling means for obtaining a sum of outputs from the first and second filters as a lower band signal, and subtraction means for obtaining a difference between outputs from the first and second filters as an upper band signal.

The present invention provides a receiver for an independent sideband (ISB) signal, the receiver comprising an input for a modulated ISB signal, means for providing an orthogonal zero IF ISB signal, and an upper sideband and a lower sideband signal. Demodulation means for demodulating the zero IF ISB signal to generate, respectively, wherein the demodulation means comprises an in-phase and quadrature phase component of the zero IF ISB signal, respectively, as inputs of the first and second real filters; Summing means having first and second real filters, first and second inputs coupled to outputs of the first and second real filters, and outputs for providing one of the independent sideband signals, and the first An output for providing another of the independent sidebands with the first and second inputs connected to the output of the first and second real filters, respectively, to subtract the output of the second filter from the output of the filter. Compared comprises a subtracting means.

Providing a real filter on both sides of the orthogonal signal path reduces the complexity of the receiver, in particular by not requiring a band pass filter to remove any out-of-band noise and any remaining carrier signal. This is made possible by the desired frequency response, which consists of an ISB filtering arrangement.

The present invention is based on the fact that by concentrating on the band of interest, no effort is wasted in trying to produce an ideal phase jump around 0 Hz, which indicates that the Hilbert transform is difficult.

In an embodiment of the present invention, the first and second real filters include a digital filter including N (N is an integer) delay stages in which each filter is connected in series, and outputs odd-numbered steps of the first filter. These coefficients are each connected to a respective multiplier to which they are applied, and the output of the multiplier is connected to a summation means for providing a sum signal as the output of the summation means, and the first step input and the even-numbered step output of the second filter are And each output of the multiplier is connected to a summation means for providing a sum signal as an output. The coefficients applied to the multipliers of the first and second filters are real, although they are derived in the complex filter.

First and second real digital filters having alternating non-zero coefficients in accordance with a second aspect of the invention, combining means for obtaining a sum of outputs from the first and second filters, and the first and second filters An ISB filter is provided that includes a subtraction means for finding the difference between the outputs from.

The invention is now described by way of example with reference to the accompanying drawings.

Referring to FIG. 2, the receiver includes an antenna 30 coupled to the band pass filter 32 and an rf amplifier 34 on the first input side of the mixer 36. The first local oscillator 38 is coupled to the second input of the mixer 36 and used to frequency downconvert the input signal to the first IF signal. The first IF signal is filtered in bandpass filter 40 and frequency downconverted to a lower, second IF signal in mixer 42 using the signal derived in second local oscillator 44. A second IF signal with a frequency around 10 kHz is applied to an anti-alias filter 46 which reduces the bandwidth of the signal and removes all dc components and its output quadruples the second IF. Is digitalized in the A / D converter 48. The digitized signal is frequency downconverted into quadrature-related real and imaginary baseband signals I and Q using multipliers 50 and 52 and a frequency source 54 generating a 90 ° phase shifted output. The I and Q outputs are applied to the FIR filter arrangement 56, which will be described in more detail later with reference to FIG. The output from array 56 includes an upper band signal and a lower band signal USB and LSB, respectively. This is stored in RAM 58 in preparation for being read by the user. When reading the USB and LSB signals, this applies to an extension / equalization step 60 that reassembles the voice message, adjusts the pitch and corrects amplitude fluctuations. The reassembled message is converted into an analog signal at the D / A converter 62 and the output is applied to the audio converter 64.

The FIR filter arrangement 56 shown in FIG. 3 includes two real low pass filters that process the I and Q signals, respectively.

For the I signal path, the filter includes a series of delay stages (DI1, DI2 ... DI (N)). Multipliers M1, M3, M (N-3) and M (N-1) are coupled to the outputs of odd-delay steps DI1, DI3 ... DI (N-3), DI (N-1) do. The real coefficients c ₁ , c ₃ ... c _N-3 and c _N-1 are applied to each of the multipliers M1, M3 ... M (N-3), M (N-1). The results of the calculated product are combined at summing step 12A with output 14A.

In the Q signal path, the filter is connected in series with delay stages (DQ1, DQ2 ... DQ (N-2), DQ (N-1), DQ (N)), multipliers (M0, M2, M4 ... M (N-2), M (N)) and similarities in that it includes a summing step 12B for combining the outputs from the multiplier and providing the output signal to the output 14B. However, the difference is that the multiplier M0 ... M (N) is the input of the first delay step DQ1 and the output of the even-numbered delay steps DQ2, DQ4 ... DQ (N-2), DQ (N). Is connected to. Also, the real coefficients (c ₀ , c ₂ , c ₄ ... c _N-2 , c _N ) are different from the coefficients applied to the multipliers of other filters.

Compared with FIG. 1, FIG. 3 has two real low pass filters, each with a series of delay steps, but the number of multipliers per filter is half the order used in the Hilbert filter shown in FIG. By having two real lowpass filters, the total number of multipliers is no more than that used in the Hilbert filter and the output from summing steps 14A, 14B uses the arrangement shown in Figure 1 without any additional bandpass filtering required. Can be used.

ISB signals are often audio signals, the frequency range starting at approximately 100 or 200 Hz Extends to the kHz region, incorporating the total bandwidth. If a ^{th th} order FIR filter with coefficients {c ₀ , c ₁ , ..., c _N } If you have a low pass frequency response with a cutoff frequency of in At the bandwidth of interest up to, the filter inducing LSB and USB can be easily configured with the same coefficients as shown in FIG. Is the sampling frequency of the filter. Therefore, the filter order of the filter shown in FIG. 3 is smaller than the filter order of the Hilbert transform for the same unwanted sideband attenuation. Even if two filters for I and Q are used, only N + 1 multiplication should still be performed. Having a lower order N means less addition and multiplication is needed and the group delay of the circuit is smaller. Moreover, additional filtering, for example for noise shaping, can be avoided using this approach.

The real low pass filter may be used to derive a complex band pass filter by mixing the exponential function and the coefficients of the FIR filter as in Equation 1 below.

here, Is the coefficient of the complex bandpass FIR filter, Is the k _th coefficient of the original lowpass FIR filter, Is the mixing or moving frequency, Is the sampling frequency and p is an arbitrary phase. By choosing p as a multiple of Equation 2, the amplitude response of the filter remains unchanged.

Therefore, Equation 1 may be expressed as Equation 3.

A special case of the following Equation 4 is the coefficient of the USB filter as shown in Equation 5.

Similarly, moving the filter response in the negative direction as shown in Equation 6 yields the coefficients of the LSB filter as shown in Equation 7.

Therefore, the set of two coefficients that determine USB and LSB are Equations 8 and 9, respectively.

and

to be. Two special characteristics can be observed. First, each coefficient is real or imaginary but not mixed. Second, the set of coefficients for the USB and LSB filters differ only in their sign at every second position. Only imaginary coefficients have inverted signs. Using these characteristics results in a structure as shown in FIG.

As an example, in the special case of making N = 8, it has been found that 40 dB of unwanted sideband suppression can be obtained by applying a 16 th order FIR filter as a low pass filter. The filter has been designed using an equiripple approach.

For illustrative purposes, FIG. 4 illustrates an example of the ISB signal and it will be appreciated that the USB and LSB are different from each other by investigation.

5A illustrates the frequency response of a standard low pass filter when N = 16, f _s = 6.4 kHz and f = 1.6 kHz.

5B and 5C illustrate the frequency response of the USB filter and LSB filter derived from the low pass filter coefficients, respectively.

The ISB filter can be applied wherever the AM-SSB signal is present. ISB filters can be implemented in hardware as components programmed and commercialized in FPGAs, or as software running on custom ASICs or DSPs.

By referring to this specification, various changes are apparent to those skilled in the art. Such modifications may include other features that are already known in the design, manufacture, and use of ISB receivers and filters and their components that may be used instead of or in addition to the features already described herein.

In accordance with the present invention there is provided an independent sideband communication system such as a voice paging system.

Claims (8)

Means for providing a receiver for an independent sideband (ISB) signal, an input for a modulated sideband (ISB) signal, a quadrature related zero IF ISB signal, Demodulation means for demodulating the zero IF ISB signal to produce a waveband and a lowerband signal, respectively, The demodulation means includes first and second real digital filters having alternating non-zero coefficients, combining means for obtaining a sum of outputs from the first and second filters as a lower band signal, and the first A subtraction means for subtracting the difference between the outputs from the first and second filters into a high sideband signal. A receiver for an independent sideband (ISB) signal, means for providing an input for the modulated ISB signal, means for providing the orthogonal zero IF ISB signal, and for generating the upper sideband and the lower sideband signals, respectively; Demodulation means for demodulating an IF ISB signal, The demodulation means are first and second real filters, wherein the inputs of the first and second real filters comprise in-phase components and quadrature components of the zero IF ISB signal, respectively; Wow, Summing means having an output for providing one of the independent sideband signals and the first and second inputs coupled to the outputs of the first and second real filters; An output for providing another of the independent sidebands and the first and second inputs respectively connected to the outputs of the first and second real filters to subtract the output of the second filter from the output of the first filter; A receiver for an independent sideband signal comprising a subtraction means. 3. The receiver of claim 2 wherein the first and second real filters are alternating digital filters whose coefficients are not zero. 3. The digital filter of claim 2, wherein the first and second real filters comprise a digital filter comprising a delay step in which each filter is connected in series with N (N is an integer), The odd-numbered step output of the first filter is connected to each multiplier to which coefficients are applied respectively, and the output of the multiplier is connected to a summation means for providing a sum signal as an output An independent side, characterized in that the first stage input and the even-numbered stage output of the second filter are connected to respective multipliers to which coefficients are respectively applied, and the output of the multiplier is connected to a summation means providing a sum signal as an output. Receiver for the band signal. 5. The receiver of claim 4 wherein said coefficients applied to the multipliers of said first and second filters are real. An ISB filter comprising: first and second real digital filters having alternating non-zero coefficients, combining means for obtaining a sum of outputs from the first and second filters, and the first and An ISB filter comprising subtraction means for finding the difference between the outputs from the second filter. 7. The method of claim 6, wherein the first and second real filters comprise a digital filter comprising a delay step in which each filter is connected in series with N (N is an integer), The odd-numbered step output of the first filter is connected to each multiplier to which respective coefficients are respectively applied, and the output of the multiplier is connected to a summation means providing a sum signal as an output A first step input and an even-numbered step output of the second digital filter are connected to respective multipliers to which coefficients are respectively applied, and the output of the multiplier is connected to the adding means for providing a sum signal as an output. ISB filter. 8. The ISB filter of claim 7, wherein the coefficients applied to the multipliers of the first and second filters are real.