CN101093624B - Method of Realizing Multiple Analog-to-Digital/Digital-to-Analog Conversion Based on Experimental Platform - Google Patents

Abstract

Based on the experimental platform to achieve a variety of analog-to-digital/digital-to-analog conversion methods, a high-precision AD chip is used to ideally sample the input analog signal, and then an FPGA chip or CPLD chip is used to quantify and encode the signal to complete the analog-to-digital conversion. The digital-to-analog conversion is the inverse process corresponding to the analog-to-digital conversion, which sequentially decodes the input digital signal, dequantizes it, and finally restores it to an analog signal. The present invention can realize all known analog-to-digital/digital-to-analog conversion methods through the new combination of high-precision AD chips and large-scale programmable devices, and can independently program the three processes of sampling, quantization and encoding, through The oscilloscope and spectrum analyzer have excellent programmability and observability for observing the experimental effect. In addition, the complete openness of the present invention makes it also applicable to the development and research of new analog-to-digital/digital-to-analog conversion methods.

Description

Method of Realizing Multiple Analog-to-Digital/Digital-to-Analog Conversion Based on Experimental Platform

technical field

The invention belongs to the technical field of communication electronics and relates to the conversion of analog signals and digital signals, in particular to a method for realizing multiple analog-to-digital/digital-to-analog conversions based on an experimental platform.

Background technique

Digital communication system has gradually become the development direction and mainstream of communication due to its outstanding advantages such as strong anti-interference ability, error controllability, and easy signal processing. However, many information in nature are analog signals, such as voice, image, etc., so analog Digital/digital-to-analog conversion is particularly important. The so-called analog-to-digital conversion is to convert an analog signal into a digital signal. The typical method is divided into three steps: sampling, the operation of the analog signal in the time domain, and completing the time discretization; quantization, the sampling value of the analog signal Quantization, so that the amplitude becomes a limited number of values; encoding, using a set of binary codes to replace the limited number of values. The digital-to-analog conversion is the inverse process of the analog-to-digital conversion. It performs decoding and low-pass filtering on the received digital signal to restore the original analog signal.

With the rapid development of communication technology, the overall requirements for the modern communication principle experiment system are: 1. Multi-functionality, that is, a variety of communication principles at different levels and in different ways or methods can be carried out on the same experimental system, platform or device Experiment; 2. Reproducibility and innovation, that is, during the whole process of the communication principle experiment, the experimenter can not only choose to set the existing communication methods or methods on the experimental system at the system level and module level, but also reproduce The results of predecessors can verify relevant principles, and can also be designed on the experimental system to create new communication functions; 3. Demonstration and observability functions that completely correspond to communication theories and principles, such as analog signal digitization One of the typical ways is pulse code modulation. The three steps in the digitization process—sampling, quantization, and encoding—should be observable through experimental systems.

Figure 1 shows the functional block diagram of the communication principle experiment system. Its component module subsystems, except for the analog-to-digital/digital-to-analog conversion part (the source encoder/decoder in the figure), are relatively easy to develop to meet the above requirements The various experimental subsystems, but the most basic part - the analog-to-digital/digital-to-analog conversion module has not been able to meet the above requirements so far. Take the existing communication principle teaching experiment system put into use as an example, such as the SEED-DTK_FPD experiment box of Hezhongda Company, a high-performance stereo audio Codec chip TLV320AIC23B launched by TI is used for experiments, but the analog-to-digital conversion of TLV320AIC23B ( ADCs) and digital-to-analog conversion (DACs) components are highly concentrated inside the chip, the user can only make some limited choices on the sampling rate, sampling length and number of sampling bits, and cannot observe the waveforms of each link in the digitization process of the analog signal, such as due to sampling Different results from rate changes and quantization noises with different accuracies from different quantization methods, etc., and users cannot fully participate in the self-design of each step in the process of digitizing analog signals. Another example is the JH5001A communication principle experiment box of Jiehui Company, which is only limited to voice signals. Taking PCM as an example, this experimental system can only verify the sampling theorem and the principle of PCM coding and decoding. How to choose the optimal quantization according to the different statistical characteristics of the signal method, as well as comparing different quantization signal-to-noise ratios produced by different quantization methods and observing the waveforms of each stage of sampling, quantization, and encoding of PCM, etc. cannot be tested, and a new analog-to-digital/digital-to-analog conversion method cannot be designed for experimental verification. In addition, the openness of the open communication principle experiment platform with application number 200520022910.6 is limited to modulation and demodulation, error correction, coding and decoding, while the analog-to-digital and digital-to-analog conversion subsystems still simply use common analog-to-digital and digital-to-analog integrated chips Therefore, the modulus and digital-to-analog conversion methods are relatively fixed, and the platform structure is complex and huge, and each step in the conversion process is not easy to separate, and the corresponding parameters cannot be freely controlled and changed for experiments, so it is impossible to conduct experiments on the modulus and digital Each step of the analog conversion system is tracked and observed or designed, which reduces the experimental effect of the experimental system.

The reason why the analog-to-digital/digital-to-analog conversion subsystem of the existing communication principle experiment system cannot meet the requirements of the above-mentioned communication experiment teaching for the experimental system is that the analog-to-digital/digital-to-analog conversion part is simple by the analog-to-digital/digital-to-analog integrated chip However, existing analog-to-digital/digital-to-analog conversion chips, such as MAX132 and MAX1400, can only achieve a fixed analog-to-digital/ Digital-to-analog conversion methods, such as PCM or DM, cannot implement two or more analog-to-digital conversion methods on one chip, and cannot choose different quantization methods according to the information characteristics of the signal, or choose different encoding methods flexibly , the experimental function demonstration of the analog-to-digital/digital-to-analog conversion subsystem of the communication principle experimental system constituted by it is incomplete, the degree of freedom in design is small, and the experimental effect is not good.

Contents of the invention

The technical problem to be solved by the present invention is: the openness of the existing modulus/digital-analog experiment system is insufficient, the degree of freedom of use is small, and the demonstration is incomplete. Provide a method based on the experimental platform to realize multiple analog-to-digital/digital-to-analog conversions. It uses fewer devices and can realize multiple analog-to-digital/digital-to-analog conversion methods in the same system. The sampling rate, quantization algorithm and encoding method can be freely selected. And the conversion process can be observed through an oscilloscope and a spectrum analyzer.

The technical scheme of the present invention is: realize the method for multiple analog-to-digital/digital-to-analog conversions based on the experimental platform, including an analog-to-digital conversion module and a digital-to-analog conversion module, and the analog-to-digital conversion module first uses a high-precision AD chip to realize the input analog signal Ideal sampling, and then use an FPGA chip or CPLD chip to realize the quantization and coding of the signal, and complete the analog-to-digital conversion. Among them, changing the clock signal of the high-precision AD chip to obtain different sampling rates, and programming the FPGA chip or CPLD chip to achieve different quantization And encoding method, thus can achieve a variety of modes of analog-to-digital conversion, digital-to-analog conversion to the inverse process corresponding to analog-to-digital conversion, sequentially decode the input digital signal, inverse quantize, and finally restore it to an analog signal, the analog-to-digital conversion module All chips of the digital-to-analog conversion module use a unified clock signal.

The analog signal of the present invention filters out the noise with a filter before the analog-to-digital conversion, after the digital-to-analog conversion, the low-pass filter filters out the original analog signal, and downloads it to the FPGA chip or CPLD chip by the single-chip microcomputer or computer programming, and the single-chip microcomputer or computer also provides and controls all Chip clock signal. The signals after sampling, quantization, decoding, and inverse quantization are respectively input into a digital-to-analog conversion chip, such as test point 1, test point 2, test point 3, and test point 4 in Figure 2 and Figure 3, with an oscilloscope or a spectrum analyzer Observe the waveform or frequency spectrum output by the chip.

The invention realizes ideal sampling through a high-precision AD converter, and provides a basis for realizing various quantization and encoding methods by FPGA/CPLD chips. The principle of the present invention is shown in Figure 4. Analog signals, commonly including single-frequency sine wave signals, voice signals, and image signals, first pass through analog filters to filter out redundant noise, and then enter the analog-to-digital conversion module. Single-chip microcomputer or computer controls the clock signal input terminal to change the sampling rate; use single-chip microcomputer or computer to download the program written or the sample program provided by the system to the FPGA/CPLD chip to realize different digitization methods or quantization methods and different encoding methods; Use a trace oscilloscope and/or a spectrum analyzer to observe the time-domain and frequency-domain characteristics of the sampled or quantized signal. Digital-to-analog conversion is the inverse process of corresponding analog-to-digital conversion. According to the quantization and decoding methods used in analog-to-digital conversion, the FPGA/CPLD chip is used for corresponding decoding and inverse quantization, and then the high-precision DA chip restores the analog signal, and uses filtering The filter filters out the noise received during signal transmission, and the time-domain and frequency-domain characteristics of the decoded and inverse-quantized signal can also be observed with a dual-trace oscilloscope and/or a spectrum analyzer.

In order to make the whole system easy to use, a friendly man-machine interface is also designed. The whole experimental system is controlled by a single-chip microcomputer, and the demonstration, design, programming or participatory experiments are performed through the computer in the form of a man-machine interface.

The beneficial effects of the present invention are: through the new combination of high-precision AD chip and large-scale programmable device function, all known analog-to-digital/digital-to-analog conversion modes can be realized; The two processes are independently programmed, and the user can freely change the sampling rate, quantization accuracy and encoding method according to the needs, and can observe the experimental effect through the dual-trace oscilloscope and spectrum analyzer, providing a teaching experiment system that is convenient for in-depth observation and learning, and flexible use. It has excellent programmability and observability. In addition, the complete openness of the present invention makes it also applicable to the development and research of new analog-to-digital/digital-to-analog conversion methods.

Description of drawings

Figure 1 is a functional block diagram of the communication principle experiment system.

Fig. 2 is a schematic diagram of the analog-to-digital conversion module of the present invention.

Fig. 3 is a schematic diagram of the digital-to-analog conversion module of the present invention.

Fig. 4 is a schematic diagram of the principle of the present invention.

FIG. 5 is a schematic diagram of hardware for a specific implementation of the present invention.

FIG. 6 shows the sampled signal and two different quantization results in the PCM analog-to-digital conversion according to the embodiment of the present invention.

Figure 7 is a block diagram of the DPCM system.

FIG. 8 is a waveform diagram of signals after sampling in DPCM analog-to-digital conversion according to an embodiment of the present invention.

FIG. 9 is a waveform diagram of signals after quantization in DPCM analog-to-digital conversion according to an embodiment of the present invention.

Fig. 10 is the original signal recovered after DPCM digital-to-analog conversion according to the embodiment of the present invention.

Figure 11 is a functional block diagram of the DM system.

FIG. 12 is a waveform diagram of a signal sampled in DM analog-to-digital conversion according to an embodiment of the present invention.

FIG. 13 is a waveform diagram of signals after quantization in DM analog-to-digital conversion according to an embodiment of the present invention.

Detailed ways

The invention can realize various analog-to-digital/digital-analog conversion methods, among which, it can be an existing mature analog-to-digital/digital-to-analog conversion method, and can also be used as an experimental platform for realizing or testing a new analog-to-digital/digital-to-analog conversion method. FIG. 5 is a schematic diagram of hardware for a specific implementation of the present invention. Among them, the high-precision AD conversion chip chooses AD7678 of ADI Company, and its conversion precision is 18 bits; the high-precision DA chip chooses AD760 of the same series, and its conversion precision is 18 bits; the FPGA chip chooses EP2C8 of Altera Company; the common DA chip chooses a typical DAC 0832; 80C51 series single-chip microcomputer is selected as the general controller, and the observation instruments are TDS2012 dual-trace oscilloscope and E4438C spectrum analyzer. The following combines the experimental system shown in Figure 5, taking pulse code modulation (PCM), differential pulse code modulation (DPCM) and delta modulation (DM) conversion methods as examples to illustrate the functions and effects of the system.

PCM conversion mode, input an analog voice signal m(t), its maximum value |m(t)|<0.2V, frequency (300, 3400Hz):

1. Sampling: Use an 18-bit high-precision AD converter to achieve ideal sampling, that is, the pulse width τ of the sampled pulse tends to zero, and the sampling rate is 16kHz. The sampled analog signal is represented by a high-precision multi-bit binary digital signal. Due to the sampling rate The size can be controlled by a switch or a computer program, and the time-domain waveform or frequency-domain waveform of the sampled signal at various sampling rates can be observed through an oscilloscope or a spectrum analyzer, and the sampling theorem can also be intuitively understood;

2. Quantization: Convert the signal that takes an infinite variety of possible values after sampling into a limited number of possible values through a quantizer. FIG. 6 is a comparison of the sampled signal and two different quantization results, wherein the quantization method 1 is an 8-bit uniform quantizer; the quantization method 2 is a 4-bit uniform quantizer. Users can design their own quantizers on the computer, select or design the quantization series, quantization steps and quantization intervals of the quantizer according to the computer's analysis of the statistical characteristics or amplitude characteristics of the input signal, program and download the program to the FPGA in the system , to achieve quantizers with different quantization accuracy or performance, and observe the quantized time-domain and frequency-domain waveforms through dual-trace oscilloscopes and spectrum analyzers;

3. Coding: Map the quantized signal level value into a binary code group to complete the conversion from analog signal to digital signal. The user programs and designs the encoder on the computer, and downloads the written program to the FPGA chip of the experimental platform to realize the corresponding encoding function. Binary codes are generally used in PCM. There are three commonly used binary code types: natural binary code, folded binary code and Gray binary code. As shown in Figure 6, the sampled signal is quantized by method 1, taking the first ten bits as an example, the binary code group using natural binary coding is:

1st place: 0 1 1 1 1 0 1 0

2nd place: 0 1 1 1 1 0 0 0

3rd place: 0 1 1 1 0 1 1 0

4th place: 0 1 1 1 0 1 1 1

5th place: 0 1 1 1 1 0 0 0

6th place: 0 1 1 1 1 0 1 0

7th place: 0 1 1 1 1 0 0 0

8th place: 0 1 1 1 0 1 1 0

9th place: 0 1 1 1 0 1 1 1

10th digit: 0 1 1 1 1 0 0 0

The encoding method can use equal-length binary codes to encode the quantized signal samples, or can use variable-length binary codes, such as Huffman (Huffman), Lempel-Ziv and other encoding methods to complete the encoding.

The digital-to-analog conversion part is the inverse process corresponding to the analog-to-digital conversion, and its decoding and inverse quantization processes are also programmable and observable. Users can conduct step-by-step observation, learning and experiential operations at the principle level, such as learning and deepening understanding of sampling theorem; on the other hand, they can also conduct research and development operations, such as writing new quantization algorithm programs and downloading them to FPGA/ In the CPLD chip, the quality of the quantization method designed by the user is verified through the distortion degree display, the analog waveform display and the effect of the speaker.

代表重建信号。预测器的输出为

差值

作为量化器输入，e_qn为量化器输出，量化后的每个预测误差e_qn被编码成二进制数字序列，通过信道传送到目的地。该误差e_qn同时被加到本地预测值而得到

在接收端装有与发送端相同的预测器，它的输出

与e_qn相加产生

信号

既是所要求的预测器的激励信号，也是所要求的解码器输出的重建信号。本实施例以一段语音信号为模拟信号源，经过16KHz抽样后，其波形如图8所示，根据DPCM的原理框图编写量化程序和编码程序，采用4电平量化，量化后波形如图9所示，最后由解码器输出的语音重建信号如图10所示。DPCM is converted into an analog-to-digital conversion method using difference coding. Figure 7 is a schematic block diagram of DPCM, x _n represents the current source sample value, and the input of the predictor

Represents the reconstructed signal. The output of the predictor is

difference

As the input of the quantizer, e _qn is the output of the quantizer, and each quantized prediction error e _qn is encoded into a sequence of binary numbers and transmitted to the destination through the channel. The error e _qn is simultaneously added to the local predictor and get

The receiving end is equipped with the same predictor as the sending end, and its output

Adding to e _qn yields

Signal

Both the required excitation signal for the predictor and the required reconstruction signal for the output of the decoder. In this embodiment, a section of speech signal is used as the analog signal source. After sampling at 16KHz, its waveform is shown in Figure 8. The quantization program and encoding program are written according to the principle block diagram of DPCM, and 4-level quantization is adopted. The waveform after quantization is shown in Figure 9. Shown, the final speech reconstruction signal output by the decoder is shown in Figure 10.

Delta modulation (DM) is an analog-to-digital conversion method that uses difference coding, and only uses one bit coding to represent the relative size of adjacent samples, thus reflecting the changing trend of the waveform at the sampling moment. Figure 11 is a schematic block diagram of DM. The specific conversion process is as follows: the sampling is realized by a high-precision AD conversion device, the sampling rate is controlled by the user using a switch or programming, and the sampling rate can be increased or decreased at will. The time-domain waveform or frequency-domain waveform of the sampled signal at the sampling rate, the difference between adjacent sampled values reflects the change law of the analog signal, and the difference between adjacent sampled values is either greater than zero or less than zero, using "1" code and " The 0" code represents that the difference between adjacent samples is greater than zero and less than zero, thus completing quantization and encoding. Correspondingly, the decoding process is to increase and decrease a quantization step according to the received "1" and "0", respectively, and then restore the original speech signal through the filter. The sampling rate of this embodiment is 16kHz. Figure 12 is the waveform diagram of the analog signal after sampling, and Figure 13 is the waveform diagram of the signal after quantization. The user can observe and analyze the output waveform of the quantizer and compare it with the theoretical waveform to verify the experimental effect.

The experimental system of the invention can not only realize the existing analog-to-digital/digital-to-analog conversion method, but also serve as an experimental platform for developing, innovating and researching a new analog-to-digital/digital-to-analog conversion method.

Claims (9)

1. The method for realizing multiple analog-to-digital/digital-to-analog conversions based on the experimental platform is characterized in that the analog-to-digital conversion first uses a high-precision AD chip to realize ideal sampling of the input analog signal, and then uses an FPGA chip or CPLD chip to realize the signal. Quantization and coding, complete the analog-to-digital conversion, change the clock signal of the high-precision AD chip to obtain different sampling rates, and program the FPGA chip or CPLD chip to achieve different quantization and coding methods, thereby achieving various modes of analog-to-digital conversion. Digital-to-analog conversion is the inverse process corresponding to analog-to-digital conversion. The FPGA/CPLD chip is used to decode and inversely quantize the input digital signal in sequence, and then the high-precision DA chip restores the analog signal. All chips used for analog-to-digital conversion and digital-to-analog conversion Unified clock signal. 2. The method for realizing multiple analog-to-digital/digital-to-analog conversions according to claim 1 is characterized in that the analog signal is filtered out with a filter before the analog-to-digital conversion, and the original analog signal is filtered out by low-pass filtering after the digital-to-analog conversion . 3. the method for realizing multiple analog-to-digital/digital-to-analog conversions according to claim 1 or 2 is characterized in that it is downloaded to FPGA chip or CPLD chip by single-chip microcomputer or computer programming, and single-chip microcomputer or computer also provides and controls the clock of all chips Signal. 4. The method for realizing multiple analog-to-digital/digital-to-analog conversions according to claim 1 or 2, is characterized in that the signals after sampling, quantization, decoding, and inverse quantization are input into a digital-to-analog conversion chip respectively, and are analyzed with an oscilloscope or frequency spectrum The analyzer observes the waveform or spectrum of the chip's output. 5. The method for realizing multiple analog-to-digital/digital-to-analog conversions according to claim 3 is characterized in that the signals after sampling, quantization, decoding, and inverse quantization are respectively input into a digital-to-analog conversion chip, and are analyzed with an oscilloscope or a spectrum analyzer. Observe the waveform or frequency spectrum output by the chip. 6. the method for realizing multiple analog-to-digital/digital-to-analog conversions according to claim 1 or 2 is characterized in that the high-precision AD chip is an 18-bit analog-to-digital converter, and the high-precision DA chip is an 18-bit analog-to-analog converter. 7. the method for realizing multiple analog-to-digital/digital-to-analog conversions according to claim 3 is characterized in that the high-precision AD chip is an 18-bit analog-to-digital converter, and the high-precision DA chip is an 18-bit analog-to-analog converter. 8. the method for realizing multiple analog-to-digital/digital-to-analog conversions according to claim 4 is characterized in that the high-precision AD chip is an 18-bit analog-to-digital converter, and the high-precision DA chip is an 18-bit analog-to-analog converter. 9. the method for realizing multiple analog-to-digital/digital-to-analog conversions according to claim 5 is characterized in that the high-precision AD chip is an 18-bit analog-to-digital converter, and the high-precision DA chip is an 18-bit analog-to-analog converter.

Images