CN101316252A - A Demodulation Method for Uniform Binary Modulation Signal - Google Patents

Abstract

A demodulation method for unifying binary modulated signals includes a narrow-band digital band-pass filter and a threshold detector: the narrow-band digital band-pass filter is composed of one or more IIR digital filter basic links cascaded, each The first basic link uses a pair of zeros and poles with very close resonant frequencies to generate a zero-group delay point on the phase-frequency characteristic curve, which can highlight the modulation characteristics of the signal while removing noise; the threshold detector is directly adjusted by setting or automatically The level value of the narrow-band digital band-pass filter is used to judge the amplitude jump corresponding to the signal modulation characteristic, so as to demodulate the binary modulation information of the unified binary modulation signal.

Description

A Demodulation Method for Uniform Binary Modulation Signal

technical field

The invention relates to a digital demodulation method for unifying binary modulation signals, which belongs to the technical field of digital information transmission.

Background technique

Research on efficient communication technology at home and abroad is mostly limited by specific application backgrounds or domain goals, focusing on a certain actual system (such as third-generation mobile communication) or specific indicators (such as code rate), which has a recognized theoretical basis and has been obtained. New technologies such as orthogonal frequency division multiplexing (OFDM) modulation, ultra-wideband (UWB) transmission, multi-antenna (MIMO) space-time/space-time-frequency coding and other new technologies that have been tested in practice and therefore have very low risk are very hot and fruitful. . But for new concepts or potential technologies with higher risks that have not yet reached a consensus, little or no one cares about them. For example, for a series of ultra-narrowband (UNB: Ultra Narrow Bandwidth) modulation/demodulation patented technologies proposed by Dr. Harold R. Walker in the United States, they are easily denied based on "common sense" or "experience". , few people can explain clearly (including Walker himself) the theoretical basis of whether it is feasible. According to our painstaking research over the years, we found that the key to realizing practical UNB communication lies in how to realize special filters suitable for various small-angle modulation waveforms.

Based on the idea of zero group delay filter, Walker has achieved a spectral efficiency much higher than that of binary phase shift keying (BPSK) modulation and a power efficiency not inferior to that of BPSK. The analog band with high quality factor (Q value) invented by Walker Although the effect of the pass filter (Figure 1, see H.R.Walker: Ultra Narrow Band Modulation Textbook, 2006, http://www.vmsk.org/) has been recently confirmed by Bell Laboratories in the United States, it is realized by using quartz crystal , and the center frequency of the crystal filter cannot be adjusted once it is set, not only has poor flexibility, is easily affected by temperature and vibration, but also is difficult to digitally integrate, especially his so-called "zero group delay" theory is not perfect; at the same time, It has long been believed (including Walker) that traditional digital filters, whether finite impulse response (FIR) or infinite impulse response (IIR) filter structures, cannot be used for ultra-narrowband communications.

Contents of the invention

Technical problem: the purpose of this invention is to design a unified demodulation method for binary modulated signals, which includes a digital filter capable of approaching analog quartz crystal filters in functional performance, and a unified binary Threshold detector for demodulation of modulated information. This filter is compatible with the small-angle modulation commonly used in UNB, especially the extended binary phase modulation (EBPSK, see "Unified binary phase modulation and demodulation method", invention patent application number: 200610040767.2, publication number: CN1889550A) proposed by us Cooperate with the modulated signal waveform of unified binary orthogonal offset keying (UBSK, see "Unified binary orthogonal offset keying modulation and demodulation method", invention patent application number: 200710025203.6, publication number: 101094209A) , in order to keep the signal characteristics as much as possible and filter the noise to the greatest extent, and simplify the receiver structure.

Technical solution: No matter which specific method is used for ultra-narrowband modulation, a common problem is to allow phase mutation signals to pass through narrowband filters. Observing the three types of Walker zero-group delay filters shown in Figure 1, it is not difficult to find that quartz crystals are crucial among them. However, when Walker uses the so-called "Zero Group Delay" (Zero Group Delay) filter based on quartz crystal at the sending and receiving ends of its UNB system, it is limited by the accuracy, stability and drift of the quartz crystal's own parameters. , the transceiver frequency difference is unavoidable, which is often catastrophic for the performance of the UNB system with "ultra-narrow" bandwidth, and the offset of the carrier frequency or the center frequency. In addition, once the quartz crystal is determined, the adjustable range of the center frequency of the filter is extremely small and it is difficult or unsuitable to be controlled by electrical tuning, so the reliability, stability and flexibility are very poor. However, for a long time, it has been believed that some special properties of quartz crystals are necessary for ultra-narrowband communication, so it is impossible for traditional FIR or IIR structure digital filters to complete ultra-narrowband filtering.

However, we noticed that when the series and parallel resonance points of the quartz crystal are very close, its amplitude-frequency and phase-frequency synthesis curves coincide exactly with the filter amplitude-frequency and phase-frequency curves given by Walker. If the digital realization of the crystal filter is considered, the series and parallel resonance points correspond to the zero and pole points of the filter respectively. By designing an IIR filter with N pairs of very close conjugate zero and pole points, the Walker crystal can be completed. Digital implementation of the filter.

A demodulation method for a unified binary modulation signal of the present invention is based on a narrow-band digital bandpass filter for removing noise and simultaneously highlighting signal modulation characteristics, and a threshold for realizing demodulation of unified binary modulation information Detector; described narrow-band digital bandpass filter is formed by cascading of one or more IIR digital filter basic links, and each basic link utilizes a pair of resonant frequency very close zero and pole to produce zero on the phase-frequency characteristic curve Group delay point, but the center frequency of the basic link is not consistent with the signal carrier frequency, and its offset is determined by matching the modulation angle of the unified binary modulation signal with the phase-frequency characteristics of the basic link of the filter, so that in While filtering out noise, the modulation angle change of the input signal can be highlighted as the amplitude jump of the output signal; the threshold detector for demodulation of the unified binary modulation information directly adjusts the amplitude through the set or automatically adjusted threshold The high and low levels are judged by jumping to demodulate the binary modulation information of the unified binary modulation signal.

Beneficial effect: the advantages of the present invention are as follows:

① It can remove noise well while maintaining signal characteristics.

The digital filtering mechanism proposed by the present invention not only has strong ability to filter out noise (because of narrow bandwidth), but also can keep signal characteristics well (because of special phase-frequency characteristics), which is very useful for radar, sonar, image processing, etc. are particularly important, but are the Achilles' heel of conventional filters. At the same time, this understanding from the frequency domain means that the bandwidth presented by the filter to the modulated signal is larger (or even much larger) than the bandwidth presented to noise and interference, thus providing practical support for expanding the Shannon channel capacity. illustration.

② Applicable surface width. The number of zeros and poles of the digital filter invented by this patent application is not limited, and the deviation between the carrier frequency and the center frequency of the filter can be set according to different system requirements, and the filtering performance can be controlled with a large degree of freedom. To adapt to different channel environments.

③ Simplifies the structure of the receiver.

The special filter proposed by the present invention cooperates with small-angle modulation signals such as EBPSK and UBSK, and overshoots at the beginning of the modulation waveform corresponding to "1", and its amplitude is obviously higher than that of the original phase non-jump waveform. Therefore, the decision can be made directly through amplitude detection, which greatly simplifies the receiver demodulation structure with the phase-locked loop as the core proposed in the patent application for "unified binary phase modulation and demodulation method" (publication number: CN1889550A). Since the phase-locked loop does not participate in the demodulation of the EBPSK signal, it can directly select (or integrate) the widely used phase-locked loop chip (that is, the content of the dotted line box in Figure 7), so the entire EBPSK transmission system can realize full digital processing, which is especially beneficial IC chip integration and simplification of the receiver.

Description of drawings

Fig. 1 is the circuit diagram of three kinds of zero-group delay filters invented by Walker.

Figure 2 is a typical amplitude-frequency characteristic and phase-frequency characteristic curve of the zero group delay filter invented by Walker.

Figure 3 and Figure 4 are schematic curves of the amplitude-frequency and phase-frequency characteristics of the parallel and series resonance of the quartz crystal, respectively.

Fig. 5 is a structural block diagram of the filter proposed by the present invention.

Fig. 6 is the amplitude-frequency characteristic and phase-frequency characteristic curves of the filter proposed by the present invention.

Figure 7 is the time-domain waveform of the EBPSK signal before and after passing through the filter: (a) is the original EBPSK signal, (b), (c) and (d) are the carrier frequencies located on the left, right and right sides of the center frequency of the filter respectively Strictly consistent (i.e., "exact tuning" as commonly understood and adopted by people).

Fig. 8 is a simplified structural diagram of an ultra-narrowband receiver.

Fig. 9 is the experimental result of the performance improvement of the filter proposed by the present invention to the EBPSK high-efficiency modulation system, which is verified by the bit error rate curve.

Detailed ways

Figure 2 is the amplitude-frequency and phase-frequency curves of the zero-group delay filter given by Walker, and Figure 3 and Figure 4 are the schematic diagrams of the amplitude-frequency and phase-frequency characteristic curves when the parallel and series resonances of the quartz crystal occur, respectively. Obviously, when the series, When the parallel resonance point is very close, the amplitude-frequency and phase-frequency synthesis curve of the crystal coincides with the amplitude-frequency and phase-frequency curve of the filter given by Walker. If the digital realization of the crystal filter is considered, the series and parallel resonance points correspond to the zero and pole points of the digital filter respectively. We can complete the Walker by designing an IIR filter with very close zero points and conjugate poles. Digital implementation of crystal filters.

The present invention takes a pair of conjugate zero, the IIR filter of pole as embodiment, and its transfer function is:

$\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="0"\; label="b"\; filenames="A20081012447500082.png,A20081012447500091.png"\; state="\{\{state\}\}">b</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="1"\; label="b"\; filenames="A20081012447500082.png,A20081012447500102.png"\; state="\{\{state\}\}">b</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="1"\; label="a"\; filenames="A20081012447500082.png,A20081012447500102.png"\; state="\{\{state\}\}">a</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

Where b ₀ =1, b ₁ =-1.630, b ₂ =1, a ₁ =1.608 and a ₂ =-0.996, the corresponding block diagram structure is shown in Figure 5, and the amplitude-frequency and phase-frequency characteristic curves are shown in Figure 6 . Figure 7 shows the filtering results in various cases when the filter center frequency deviates from the carrier frequency of the EBPSK signal. Among them, Figure 7(a) is the original EBPSK signal, and Figure 7(b) is the The filtering result when the center frequency is on the left side, Fig. 7(c) is the filtering result when the carrier frequency is on the right side of the filter center frequency, Fig. 7(d) is the result when the carrier frequency coincides with the filter center frequency, obviously , the phase jump at this time is completely erased, and the 0 and 1 signals cannot be distinguished at all.

This EBPSK transmission system uses the result shown in Figure 7(b), that is, the overshoot phenomenon occurs at the phase jump point after filtering, and adopts the receiver structure shown in Figure 8, by analyzing the initial time of each symbol Waveform amplitude detection is used to judge "0" and "1" signals, which greatly simplifies the realization of EBPSK receiver. Figure 9 shows the comparison of bit error rate with and without filter at the receiving end. Obviously, after using the receiving filter, the bit error rate performance of the system is significantly improved.

The present invention proposes that the center frequency of the digital filter should have a slight deviation from the carrier frequency of the EBPSK signal. When the carrier frequency is on the left side (low end) of the filter center frequency, the signal phase jump point will generate increased parasitic amplitude modulation, namely The amplitude of the phase hopping part increases obviously; when the carrier frequency is on the right side (high end) of the filter center frequency, the waveform amplitude of the phase hopping part decreases obviously, and the amplitude of the non-hopping part remains unchanged; and if the When the signal carrier frequency is exactly the same as the center frequency of the digital filter, the phase jump of the EBPSK signal will be completely erased, and the "0" and "1" modulation signals cannot be distinguished at all. This is the fundamental reason why it is generally believed that Walker's quartz crystal filter cannot be realized digitally.

The offset between the carrier frequency and the center frequency of the filter cannot be set arbitrarily, but to make the modulation angle of the EBPSK signal match the phase-frequency characteristics of the digital filter. The carrier frequency should be set exactly at the falling edge of the phase-frequency curve, using its transient characteristics to make the output of the EBPSK signal through the filter produce overshoot at the phase jump.

After the EBPSK signal passes through the special filter proposed by the present invention, the waveform produces an overshoot at the period corresponding to the initial jump of the "1" modulation waveform, and the amplitude is obviously higher than the sine wave part whose initial phase is 0, so In the case of precise synchronization, we can directly judge "0" and "1" signals through amplitude detection, which greatly simplifies the structure of the receiver. At the same time, due to the use of the transient characteristics of the filter, the research shows that it can maintain the signal characteristics as much as possible and filter the noise to the greatest extent, showing a larger signal bandwidth and a smaller noise bandwidth, which makes the expansion of the channel capacity It becomes possible and reveals a new way for the development of filtering theory and the improvement of communication performance.

Claims (1)

1. A demodulation method for unifying binary modulated signals, characterized in that the demodulation method is based on a narrow-band digital band-pass filter for removing noise and simultaneously highlighting signal modulation characteristics, and a method for realizing unified binary modulation A threshold detector for element modulation information demodulation; the narrowband digital bandpass filter is formed by cascading one or more IIR digital filter basic links, and each basic link utilizes a pair of resonant frequency very close zeros and poles to generate The zero group delay point on the phase-frequency characteristic curve, but the center frequency of the basic link is inconsistent with the signal carrier frequency, and its offset is determined by the modulation angle of the unified binary modulation signal and the phase-frequency characteristic of the basic link of the filter It is determined in conjunction with each other, so that the modulation angle change of the input signal can be highlighted as the amplitude jump of the output signal while filtering out the noise; the threshold detector for the demodulation of the unified binary modulation information directly passes the set or The automatically adjusted threshold judges the high and low levels of the amplitude jump to demodulate the binary modulation information of the unified binary modulation signal.

Images