Sonar signal simulator based on DDS technology

Abstract: This paper proposes an implementation plan for a digital universal sonar signal simulator based on DDS technology. By controlling the amplitude and phase of the output signal of the DDS device to simulate the output of the imaging sonar array, the target echo at any distance and orientation can be accurately simulated, and the echo signal of a moving target can be simulated. Specific issues in the design and implementation of multi-channel signal simulators are discussed. Keywords: sonar beam form DDS CPLD In recent years, with the development of ocean and naval technology, the research of sonar equipment has attracted more and more attention. However, since hydroacoustic equipment testing usually requires a suitable hydroacoustic environment, such as an anechoic pool, a lake or the ocean, the complexity and cost of the test are high. In order to simulate the target echo signal in a normal laboratory environment, a dedicated sonar signal simulator needs to be designed according to the requirements of various sonar equipment. 1 Basic principles of sonar signal simulator 1.1 Introduction to beam forming principle This article aims to develop a certain high-resolution imaging acoustic signal simulator. The imaging sonar receiving sound array adopts a 48-element equally spaced linear array, with an operating frequency of 800kHz, an operating distance of 0.5 to 25 meters, and an angular resolution of 0.35%. The imaging sonar performs beamforming on the receiving array signal to achieve the acquisition of acoustic images. The basic principle of sonar beamforming is shown in Figure 1. Figure 1 is a schematic diagram of an equally spaced linear array receiving echo signals in the far field. The incident sound wave is incident parallel to the normal direction of the matrix at an angle θ. The primitives are numbered 1, 2,...i, t+1,...N from left to right, and the spacing between the primitives is d. If primitive No. 1 is selected as the time reference point, and the signal received by it is Acos2πft, then there is a sound path difference Δ=dsinθ between two adjacent primitives, so the signal received by the i-th primitive is: si(t )=Acos{2πf[t+(i-1)dsinθ/c]} (1) where c is the speed of sound. Since imaging sonar is a narrow-band active sonar, the phase difference between the received signal of element I and element 1 is φi=2π(i-1)d/λsinθ, where λ is the wavelength. Therefore, if you want to orient the linear array in the θ0 direction, you only need to delay the signal of the i-th element by τi(θ0)=2π(i-1)d/λsinθ0. The above is the basic principle of linear array beamforming, but this is only an approximation in the far field case. For near-field conditions, the error produced by such an approximation can be large. For the high-frequency imaging sonar in this article, since the entire operating range belongs to near-field conditions, a focusing method must be used during beam forming. The basic principle is the same as above, except that the delay (or phase shift) of each primitive signal is not linear, and this article will not go into detail. 1.2 Principle of sonar signal simulator The signal simulation quantity used for imaging sonar is generally the same as the number of primitives, and the output of each channel simulates the signal of one primitive in the sonar array. Because imaging sonar operates at a short distance and the noise level in the high-frequency band in the underwater acoustic environment is very low, the received signal-to-noise ratio is usually high. For this reason, no additional noise is added to the output of the signal simulator. Imaging sonar works in a strong reverberation environment. Since the simulation of reverberation is difficult and the impact on imaging is not serious, the simulation of reverberation is not considered in the design and only focuses on simulating near-field targets. echo. According to the orientation and distance of the point target to be simulated input by the user, the signal simulator calculates the phase difference of the corresponding target echo reaching each element of the receiving array, and then generates corresponding multi-channel sinusoidal signals based on these phase differences. These signals are added to the input end of the imaging sonar to replace the real array output, so that the imaging sonar can be easily debugged and measured under onshore laboratory conditions. 1.3 Defects of traditional sonar signal simulators Traditional sonar signal simulators usually use a fixed oscillator to generate a sinusoidal signal with the same operating frequency as the sonar system. The local oscillator signal is passed through a set of multi-tap simulated delay lines, and then signals are extracted from different taps of the delay line as the output of the simulator. This signal simulator structure has several flaws and shortcomings. First, due to the use of analog devices to form a tapped delay line structure, the minimum variable delay length is limited. Especially considering the scale and cost of the system hardware, the number of taps of the general delay line is not large, which causes a large error between the delay time and the theoretical value, thus reducing the accuracy of the simulator. Secondly, in order to simulate the echo signals of targets in different directions, the outputs of different tap delay lines must be switched or combined, and then output to the sonar equipment as a primitive signal. Therefore, the entire simulator is large in scale and can only simulate several targets in discrete azimuths and distances. It cannot simulate the echo of point targets at any azimuth distance, otherwise it will be difficult to achieve without increasing the complexity. In addition, using tapped delay lines composed of analog devices makes it difficult to ensure channel consistency and makes debugging difficult. Moreover, the frequency range of the delay line is narrow. If the frequency parameters change, it will not be used normally. Therefore, the applicable range is narrow and the cost performance is very low. In order to overcome these shortcomings of traditional sonar signal simulators, this paper uses DDS technology to design and implement a new signal simulator. This DDS-based simulator structure can achieve accurate simulation of point target echo signals at any azimuth distance, is suitable for different frequency parameters and has certain expansion capabilities, making it highly cost-effective. 2 Signal simulator composed of DDS 2.1 Introduction to DDS technology DDS technology appeared in the 1970s and is an all-digital frequency synthesis technology. It introduces advanced digital signal processing theories and methods into the field of signal synthesis to achieve the unification between the frequency conversion speed and frequency accuracy of the synthesized signal. It has many advantages such as continuous phase transformation, fast frequency conversion speed, extremely high frequency resolution, low phase noise, easy to control with various methods such as microcomputer, small size, and high integration. Therefore, DDS has gained great popularity in theory and application in recent years. Rapid development. The basic structure of DDS is shown in Figure 2. Since DDS has the characteristics of numerical control with frequency and phase determination, the DDS device is used as a key component of the imaging sonar signal simulator, and supplemented with corresponding control and interface logic, etc., it is possible to achieve target echo processing at any orientation and distance. Accurate simulation. 2.2 Simulator structure composed of DDS The structure of the imaging sonar signal simulator based on DDS technology is shown in Figure 3. The simulator has a total of 48 signal channels, and each channel simulates the output of a primitive in the sonar receiving array. The channel circuit consists of the single-chip DDS device AD9830 and its interface logic circuit, output IV converter, filtering and following circuit. The interface between the DDS device of each channel and the CPLD adopts a 16-bit wide parallel bus. The user inputs the target orientation, distance, signal amplitude and other information to be simulated into the host application interface. Based on this information, the application calculates the phase difference and other parameters of each channel signal relative to the reference channel according to the near-field focusing algorithm. , and then download these parameters to the signal simulator through the RS-232 serial bus. The microcontroller in the signal simulator receives and decodes these parameters, and writes the frequency and phase parameters of each channel signal into the control register of the DDS device of the corresponding channel through CPLD. The flow chart of AD9830 initialization and parameter setting is shown in Figure 4. 2.3 Analog signal of target distance The parameters received by the simulator from the host computer include, in addition to the frequency and phase difference of each channel, the amplitude (gain) control curve parameters of the output signal. The amplitude control parameters are also decoded by the microcontroller and sent to the DAC according to the timing parameters of the curve. The amplitude control signal after digital-to-analog conversion is sent to the Rset end of the AD9830, thereby controlling the amplitude of the output signal. Figure 4 AD9830 initialization and parameter setting flow chart. This amplitude control circuit is an open-loop system with good dynamic performance and a bandwidth of up to 100kHz. By using an appropriate amplitude (gain) control curve and cooperating with an external trigger source, the simulation of the target echo signal at a predetermined distance can be achieved. Its working principle is shown in Figure 5. To simulate two point targets that are L1 and L2 apart from the receiving matrix, the corresponding amplitude control curve is generated according to the conversion rate of the DAC in the simulator. This amplitude control curve is synchronized in the simulator with an external trigger signal that is synchronized with the imaging sonar transmitter, that is, its rise is aligned with the leading edge of the sonar transmit pulse. In this way, the point target echo at a predetermined distance can be accurately simulated under the control of the amplitude curve. There are a total of 4 phase registers in the DDS device AD9830 used in this article. If the calculated phase parameters are written in advance and the amplitude curve is used to control the corresponding phase conversion, up to 4 different phase registers can be simulated in one transmit echo. Point targets in bearing and range. 3 Several notes 3.1 Reference clock fan-out (Fan-out) In order to improve the reliability and scalability of the system, the entire simulator adopts the structure of 3U EuroCard chassis + backplane + plug-in board. Each 8-channel circuit is in It is implemented on a plug-in board, and signals such as bus interface and reference clock are located on the backplane. This structure makes the wiring topology of the reference clock more complicated. And due to the large number of channels, all channel DDS devices share a reference clock, so clock fanout, wiring and impedance matching are very important. If the fanout is unreasonable, the reference clock at the DDS input end of each channel will be delayed, which will affect the accuracy of the system. In addition, since the reference clock frequency is as high as 50MHz, signal integrity problems will also affect the normal operation of the system. This article uses Cypress's high-speed clock distribution device CY2308 to fan out the reference clock generated by the quartz crystal oscillator into 6 independent clocks, which are sent to 6 channel boards respectively, strictly ensuring the path of each clock signal on the PCB, etc. long and provide precise impedance matching. At the same time, the same fanout and routing methods are used in each channel board. In this way, the inter-clock delay of each channel is less than 200ps, which can ensure the accuracy of the simulator. 3.2 Wiring of the hybrid circuit Since there are a large number of high-frequency digital logic control signals in the signal simulator, and the output signal is a multi-channel weak analog signal (mV level), special attention must be paid to the wiring avoidance, decoupling, and power supply of the digital-analog hybrid circuit. and ground plane division, etc. There are many monographs discussing this aspect and will not be discussed in detail in this article. It is worth noting that DDSLayout should strictly follow the reference design to ensure system performance. 3.3 Safety of DDS devices Monolithic integrated DDS devices are mostly produced using CMOS technology and are relatively fragile and easy to damage. Special attention should be paid to design and debugging. Since the output compensation may be accidentally short-circuited during use of the signal simulator, a follower is used in the output stage to avoid accidental damage to the DDS. In addition, a certain margin should be left when designing the amplitude control circuit to avoid DDS failure due to excessive output current. This paper proposes and implements a new sonar signal simulator using DDS technology. The completed simulator prototype overcomes the shortcomings of traditional simulation technology such as complex structure, poor reliability, and narrow adjustable range. It can accurately simulate point target echo signals at any orientation and distance. It is easy to use and reliable, and can be used in certain high-frequency imaging acoustic systems. Na played a very key role in the design and debugging. At the same time, the simulator has good adaptability and expansion capabilities and can be used for debugging various types of imaging sonar in the future. It has strong engineering practical value and broad application prospects.