CN108445483A - Water floats plant radar sensing system - Google Patents

Abstract

The invention relates to a radar detection system for floating plants, a radar-based monitoring system for floating and growing plants on the water surface, which utilizes the difference in microwave echo intensity between floating objects and the water surface, and uses high resolution to measure and calculate the aquatic plants flowing through the monitoring point in real time Area, and according to the area of aquatic plants flowing through different monitoring points, establish a growth model of aquatic plants, and finally realize prediction and supervision of the whole basin.

Description

Water floating plant radar detection system

technical field

The invention relates to the field of aquatic plant regulation, in particular to a radar detection system for aquatic plants.

Background technique

Floating plants have always been the focus and difficulty of river regulation, especially the exotic water hyacinth in recent years, which has strong vitality and extremely fast growth. Large pieces of water hyacinth float on the water surface and flow down the river for tens of kilometers. A large part of some water surface is covered by water hyacinth, which not only pollutes the water environment, but also affects the navigation safety of ships, and even enters the landscape waters planned by the municipal government, causing widespread concern in the society.

To this end, relevant departments have carried out relevant rectification work. Including early warning based on ship patrol, vehicle patrol, and video surveillance; clean-up work based on interception and centralized salvage and ship cruise salvage, etc., and the rectification has achieved certain results. However, affected by environmental factors such as hydrology, climate, and geography, the outbreak of water hyacinth is uncertain, resulting in a sudden outbreak of a large number of water hyacinths in certain areas and time periods, and fishing operations are temporarily stretched. Therefore, the relevant departments urgently need to establish a set of intelligent management methods for early warning and comprehensive improvement of aquatic plants, use detection sensors to obtain the distribution information of floating plants on the water surface passing through the detection points, and through the information networking and fusion of multiple detection points, integrate other parties. Meteorological and hydrological information, judging the growth trend of aquatic plants and the overall distribution of the watershed, establishing an early warning mechanism for aquatic plant management, and rationally arranging salvage operations to achieve the purpose of effectively controlling aquatic plants.

Contents of the invention

The purpose of the present invention is to overcome the deficiencies in the prior art, and provide a radar detection system for floating plants in water. The radar-based monitoring system for floating plants on the water surface utilizes the difference in microwave echo intensity between floating objects and water surfaces, and passes high Resolution, real-time measurement and calculation of the area of aquatic plants flowing through the monitoring points, and based on the area of aquatic plants flowing through different monitoring points, the growth model of aquatic plants is established, and finally prediction and supervision of the whole watershed are realized.

The purpose of the present invention is achieved through the following technical solutions:

Water floating plant radar detection system, the radar detection system is composed of antenna system, radio frequency microwave system, data processing and terminal display system;

The antenna system is composed of a transmitting antenna and a receiving antenna. The radar generates point frequency continuous wave and linear frequency modulation continuous wave signals, wherein the point frequency signal is only used for speed measurement of floating objects on the water surface, and the linear frequency modulation continuous wave signal realizes the area of floating objects on the water surface. Measurement, the signal is radiated through the transmitting antenna, and the signal enters the receiving antenna after being reflected by the aquatic plants;

The radio frequency microwave system is composed of three parts: receiving front end, data acquisition module, and frequency synthesis module. Filter output; the data acquisition module is composed of an anti-aliasing filter, ADC, FPGA and Ethernet transmission module, which is used to realize the digitization of intermediate frequency signals. The frequency synthesis component is composed of a clock reference circuit, a waveform generation circuit, a transmission channel, and an interface control circuit. , used to improve the local oscillator signal required by the receiving front end, the synchronous clock signal required by the data acquisition module, and the excitation signal for linear frequency modulation;

The data processing and terminal display system consists of a data processing module and a terminal display module to complete the online real-time calculation and display of various information of aquatic plants.

As a further improvement of this scheme, the antenna system adopts a planar array antenna system with separate transceivers, and the antenna forms a pattern characteristic of a horizontal narrow lobe and a vertical wide lobe.

As a further improvement of this program, the planar array antenna system is a waveguide planar array antenna, which adopts the structure of sub-array block design, and its operating frequency is K band, f0 ± 150MHz, and f0 is 24GHz.

As a further improvement of this scheme, the antenna gain is ≥30dB, the lobe width is ≤1° horizontally, ≤5° vertically, and the horizontal and vertical sidelobe levels are ≤-20dB; VSWR≤1.6 is required for standing waves, and the polarization mode is vertical Polarization, transceiver antenna isolation ≥ 80dB; beam pointing deviation in the frequency band electrical axis pointing deviation meets ≤ 0.2°.

As a further improvement of the program, the three-stage mixer includes a first-stage mixer, a second-stage mixer, and a third-stage mixer;

The RF excitation signal enters the first-stage mixer through the limiter, low-noise amplifier, and filter to mix to obtain the first intermediate frequency signal with a bandwidth of 7.75GHz and 300MHz;

The first intermediate frequency signal enters the second stage mixer through the filter and amplifier to obtain the second intermediate frequency signal with a bandwidth of 750MHz and 500KHz;

The second intermediate frequency signal enters the third-stage mixer after passing through the filter, amplifier, and digital control attenuator to obtain the third intermediate frequency signal with a bandwidth of 70MHz and 5MHz;

The third intermediate frequency signal is output through a filter amplifier, a numerically controlled attenuator and a low-pass filter.

As a further improvement of this program, the anti-aliasing filter is mainly used to prevent noise aliasing during ADC band-pass sampling, and its parameters are: center frequency F0=70MHz; BW-1dB=3～5MHz; BW-40dB <40MHz; BW-80dB<70MHz.

As a further improvement of this solution, the FPGA enters the three-stage decimation filter after the ADC sampling data is digitally down-converted, followed by 2-fold decimation, 5-fold decimation, and 5-fold decimation, which is equivalent to a sampling rate of 2MHz, and then passes through a After the high-pass filter, the data is packed into a fixed format and sent to the Ethernet transmission module.

As a further improvement of this scheme, the clock reference circuit generates a 100MHz signal from a constant temperature crystal oscillator, and outputs one path to the 12G source as a comb spectrum excitation source through the ADP-2-1W two-power splitter, and one path passes through the SCA-4-10 four-way The power divider supplies 3.5GHz and CRO phase-locked sources respectively; the output of LTC6946-2 receives three local oscillator signals of 820MHz; the output of LTC6946-1 transmits the excitation signal of 750MHz; the output of 13dBm signal through the amplifier is used as the acquisition clock for the signal processor.

As a further improvement of this scheme, the waveform generation circuit works as follows:

The 3.5GHz signal generated by the CRO phase-locked loop circuit and the 12G signal generated by the comb spectrum are filtered, amplified, and then mixed to generate the local oscillator signal. a vibration;

The sweeping local oscillator is divided into two channels by 3.5G, and one is amplified, frequency multiplied and filtered to generate a 7GHz local oscillator signal; the other is used as a clock for the AD9914 to generate a 600-900MHz signal, and the two signals are mixed by the HMC558 mixer and then filtered , amplification, and power distribution for the transmission channel and output to the receiving module as two local oscillators.

As a further improvement of this solution, the transmit channel is generated by a frequency synthesizer at 750MHz and mixed with a frequency-sweeping local oscillator source of 7.6-7.9GHz. After filtering and amplifying, it is mixed with a local oscillator of 15.5GHz to output a signal of 23.85-24.15GHz. , after filtering, mixing, amplifying and outputting through the isolator.

As a further improvement of this program, the performance indicators of the system are as follows:

1) Excitation: 24GHz±50MHz (23.85～24.15GHz), power: 1～1.3W, phase noise: L(1K)≤-103dBc/Hz, L(100K)≤-113dBc/Hz;

2) Local oscillator: 15.5GHz, power: 13dBm±1dBm, phase noise: L(1K)≤-108dBc/Hz, L(100K)≤-118dBc/Hz;

3) Two local oscillators: 7.75GHZ (7.6~7.9), power: 10dBm±1dBm; phase noise: better than one oscillator;

4) Three local oscillators: 820MHz, power: 10dBm±1dBm, phase noise: better than two local oscillators;

5) Clock: 100MHz, Power: 13±0.5dBm,

Phase noise: L(1K)≤-140dBc/Hz, L(100K)≤-150dBc/Hz;

6) Amplitude consistency within the FM bandwidth of the frequency sweep signal: ≤1dB;

7) Output clutter: excitation ≥ 60dBc, one oscillator ≥ 70dBc, two local oscillators ≥ 70dBc, three local oscillators ≥ 70dBc, clock ≥ 70dBc

8) Harmonic suppression: excitation ≥ 55dBc, one local oscillator ≥ 60dBc, two local oscillators ≥ 60dBc, three local oscillators ≥ 60dBc, clock ≥ 60dBc

9) Power fluctuation: ≤0.5dB

10) Power consumption: ≤30W;

11) Point frequency and linear modulation output alternately for 1s each, linear frequency modulation time 1ms, frequency modulation bandwidth 300MHz, frequency modulation linearity ≤ 2/1000.

As a further improvement of the program, the data processing module utilizes the difference in the reflectivity of rivers and aquatic plants to detect aquatic plants floating on the water surface. The area of aquatic plants is comprehensively calculated by measuring the distribution of aquatic plants flowing through the measurement section and the moving speed under the section.

As a further improvement of the program, the data processing module is as follows to the chirp continuous wave signal processing method of 1S:

The linear frequency modulation continuous wave signal realizes the area measurement of floating objects on the water surface. The echo intensity of each distance point in the section dimension is measured through the broadband signal, and the detection is carried out in the frequency domain. The area of the aquatic plants is determined by the change of the echo intensity of the presence or absence of aquatic plants. ;

Continuously send 1000 1ms linear frequency modulation continuous wave signals, process the data of the first 1ms and the data of the 1000ms;

S01: Under normal conditions, find a water surface without floating plants, measure the reflection intensity of water in each resolution unit within the azimuth beam width, and use it as the detection reference threshold;

S02: Detect the echo signal of 1 ms, take the value obtained in S01 as the reference threshold, adjust according to the actual situation, and judge whether there are floating plants in each resolution unit;

S03: Save the intensity value of each distance point, and mark the distance points with floating plants.

As a further improvement of the program, the data processing module is as follows to the point frequency continuous wave signal processing method of 1s:

S11: Calculate the power spectral density of the collected data;

S12: Take 1kHz data at an intermediate frequency of around 500KHz for threshold detection. If the threshold exceeds the threshold, it is considered to be a target. Under normal circumstances, the general water speed is less than 0.1m/s, f _d = 2v/λ, so the maximum Doppler frequency is about 16Hz, taking 1KHz on both sides with a margin in consideration of abnormal conditions such as wind, the threshold value is fine-tuned based on the detection reference threshold value obtained in S01.

S13: Calculate the target speed according to the detection frequency.

As a further improvement of this program, the data processing module calculates the target area as follows: the river width is calculated by 50 meters, divided by the distance resolution unit of 0.6 meters, and the azimuth wave velocity width is 1°, then the furthest distance in the azimuth beam is about 0.8 meters ;Calculated as follows

R = river width;

θ = azimuth beamwidth;

Then at different distances, the azimuth width covered by the azimuth beam = 2Rtan(θ/2);

S21: The target area is small

When the area of the target is less than or equal to the area of a distance resolution unit (<=0.6*0.6), the echo strength measured at this time is a bell-shaped curve with a steeper change, and the time dimension of the measured water flow in the same direction as the target at this time The length _L2 is:

Among them: 2 represents the measurement time interval of 2 seconds, v _i represents the water flow velocity obtained at each measurement time interval=water floating plant measurement speed/sinθ, n represents the number of time intervals in which targets appear continuously, and d is a fixed value, referring to The lateral distance within the azimuth beamwidth corresponding to the unit points at different distances in the section dimension;

According to the distance unit where the target appears, the length L ₁ of the target cross-sectional dimension can be obtained = length of the measured cross-sectional dimension *cosθ;

Then the area: S=L ₁ ×L ₂

S22: Large target area

The target that appears continuously in both time and distance units is regarded as a large target. At this time, the measured echo intensity change is similar to a band-pass filter. The changes on both sides are steeper, and the middle is relatively gentle. The time-dimensional target in the same direction as the water flow The length _L2 is:

L ₂ =2×v _i

Among them: 2 represents the measurement time interval of 2 seconds, and v _i represents the water velocity obtained from each measurement time interval.

According to the distance unit where the target appears, the length of the target distance dimension can be calculated L ₁ = length of the measurement section dimension*cosθ, if the distance unit of the measured target section dimension is discontinuous, then N refers to the number of distance units with water floating plants in the measured section dimension;

Then the area of floating plants measured at each time interval is: S _i =L ₁ ×L ₂

Then the target area is:

n represents the number of time intervals in which targets appear continuously, d' refers to the maximum value of the lateral distance within the azimuth beam width corresponding to the unit points of different distances in the cross-sectional dimension of the target, v refers to the last detection speed of the current target, and m refers to the repeated calculation of the area The number of time intervals, S' refers to the repeated calculation area of the water phytoplankton within the width of the beam, a total area of the flow through the measurement section is given every half hour and the result is saved.

As a further improvement of the program, the data processing module also includes calculation of data transmission, processing, storage and display time, as follows:

After A/D, the time when the data is transmitted to the data buffer area through the network cable, the input signal is a 16bit floating point number, and there are 2M sampling points in 1S, so the amount of data to be transmitted is 32Mbit;

The data transmission is planned to adopt 10Gb network, the transmission rate of 10Gb network is 10000Mb/s, and the network transmission efficiency is set to 50%, and it takes 6.4ms to transmit 32Mb data;

Write data to the hard disk, the computer hard disk storage rate is 20MB/s, it takes 200ms to store 2M points of 16bit data, read data from the hard disk to the memory, the computer hard disk read rate is 30MB/s, and it takes 140ms to read 2M points of 16bit data, from the memory to the memory The transfer rate of CPU and PCI-E bus is 8Gb/s, and the time for 32Mb data from memory to CPU is 4ms;

It takes 600ms for the CPU to perform FFT processing and threshold detection on the data, and it takes 50ms to display 1s data at the same time, so the processing time of 1s chirp signal is about 1 second.

Then there is the point frequency signal of 1s. We only take a small part of the data for processing to find the speed, and the time is about 200ms;

Therefore, the processing time of 1s linear frequency modulation signal + 1s point frequency continuous wave is about 1200ms, so the repetition time interval of 2s can complete the processing without data loss.

The beneficial effects of the present invention are: the floating plant detection radar realizes the real-time online monitoring of the area of floating plants and sundries on the water surface in the 100m wide river. The minimum detectable speed of floating objects on the water surface is as low as 0.1m/s, and the distance resolution is less than 1m; the radar has the following characteristics:

1) Real-time measurement of floating objects on the water surface;

2) High measurement accuracy;

3) Unattended, high reliability;

4) The maximum detection distance is greater than 200m;

5) Low power consumption;

6) It is suitable for flow velocity measurement of low velocity rivers.

Description of drawings

Figure 1 is a schematic diagram of the detection;

Figure 2 is a schematic diagram of the principle of cross-sectional dimension measurement;

Figure 3 is the horizontal lobe diagram of the theoretical calculation of the transceiver antenna;

Figure 4 is a theoretical calculation of the vertical lobe diagram of the transceiver antenna;

Fig. 5 is a three-dimensional lobe diagram of the theoretical meter of the transceiver antenna;

Fig. 6 is a simulation model of the transceiver antenna;

Fig. 7 is a three-dimensional simulation lobe diagram of the transceiver antenna;

Fig. 8 is a simulated horizontal lobe diagram of the transceiver antenna;

Fig. 9 is a simulated vertical lobe diagram of the transceiver antenna;

Figure 10 is the beam directing diagram of three frequency points f0-150MHz, f0, f0+150MHz;

Figure 11 is an actual beam shape diagram;

Figure 12 is a clock reference circuit;

Figure 13 is a waveform generating circuit;

Figure 14 is a transmission channel circuit;

Figure 15 is a 3500MHz CRO oscillator phase noise curve;

Figure 16 is a receiving channel circuit;

Fig. 17 is the echo input standing wave curve;

Fig. 18 is a graph of image frequency suppression degree;

Fig. 19 is a block diagram of the data acquisition module;

Fig. 20 is a schematic diagram of noise aliasing;

Figure 21 is a block diagram of the data acquisition module;

Fig. 22 Schematic diagram of calculation area of floating plants.

Detailed ways

The technical solution of the present invention is further described in detail below in conjunction with specific examples, but the protection scope of the present invention is not limited to the following description.

The aquatic plant detection radar adopts a narrow beam in azimuth to realize the minimum resolution unit of the measurement section; a wide beam is used in the elevation angle, and the energy covers the entire measurement section, and the echo intensity of each distance point in the section dimension is measured through a broadband signal. And the area of aquatic plants is determined by the change of echo intensity with or without aquatic plants.

The radar generates point frequency continuous wave and linear frequency modulation continuous wave signals, of which the point frequency signal is only used for the speed measurement of floating objects on the water surface, and the linear frequency modulation continuous wave signal realizes the area measurement of the floating objects on the water surface. After the signal is amplified by power, it is radiated by the transmitting antenna, and the signal enters the receiving antenna after being reflected by the aquatic plants, and is down-converted by the receiver. After demodulation, it becomes a digital signal, and the signal processor processes and calculates the echo strength at different distances. , and moving speed, the output section distance-flow direction distance-intensity distribution.

Its detection principle is shown in Figure 1.

There is an angle θ between the radar erection pointing and the river section,

Target section length = measured section length x cosθ;

Target flow length = target length of measurement section/sinθ.

Flow dimension measurement principle

Target length in the same direction as the river flow = resolution unit of water flow direction × number of resolution units in the detection duration period = measurement time interval × measurement speed of phytophytes × number of resolution units in the duration period/sinθ.

(1) The measurement time interval is determined by the detection radar and can be obtained accurately;

(2) The Doppler frequency fd of the target can be measured in the point-frequency continuous wave signal mode, and the measurement speed of floating plants can be calculated according to the working wavelength λ = fd×λ/2;

(3) The number of resolution units in the duration period, using the reflected echo under pure water surface conditions as the background, each resolution unit establishes the threshold in the resolution unit according to the detection background, and establishes a threshold database according to the characteristics of the water surface height, And use this threshold to judge whether there is any phytophyte in the resolution unit.

Measuring principle of section dimension

The radar transmits a chirp signal modulated from f0-150MHz to f0+150MHz. If a fixed target appears at R, the echo delay of the target at R is 2R/C. After the echo signal is de-skewed, a fixed frequency difference is formed. The relationship is shown in Figure 2:

When there are targets at different distances, different frequency differences will be formed. By making statistics on the echo intensity of different frequency points and comparing it with the echo intensity of the calm water surface, it is confirmed whether the distance point is an aquatic plant. Finally, according to the number of echo points of aquatic plants flowing through the measurement section, the area of aquatic plants is determined.

The radar generates point frequency continuous wave and linear frequency modulation continuous wave signals, of which the point frequency signal is only used for the speed measurement of floating objects on the water surface, and the linear frequency modulation continuous wave signal realizes the area measurement of the floating objects on the water surface. After the signal is amplified by power, it is radiated by the transmitting antenna, and the signal enters the receiving antenna after being reflected by the aquatic plants, and is down-converted by the receiver. After demodulation, it becomes a digital signal, and the signal processor processes and calculates the echo strength at different distances. , and moving speed, the output section distance-flow direction distance-intensity distribution.

(1) Scan form

The azimuth of the antenna is fixed, and the elevation angle can be adjusted downward, which is used to point the main beam at different distances to meet the test requirements.

(2) Parameter measurement form

Frequency domain detection, point frequency speed measurement, frequency modulation distance measurement.

(3) Antenna feeder form

Two antennas are divided into upper and lower blocks. Concentrated feed reduces loss.

(4) Launch form

Small power, solid-state emission, integrated design.

(5) Receiving form

In order to reduce the loss, the integrated design is adopted in the form of amplifying first, then frequency conversion, and then A/D. Reduce the amount of data transferred

(6) Transmission form

Network port transmission, automatic storage of low data rate AD data.

(7) Terminal form

Laptop host, C++ language implementation, adjustable screen.

(8) Structural form

Small, easy to disassemble, easy to carry, with pitch indicator, good heat dissipation.

Antenna System Solution

The continuous wave radar antenna can be in the form of a parabola or a planar array. If the transmitting subsystem adopts the T/R component scheme, the antenna should be in the form of a planar array. In addition, if the radar needs to realize multi-polarization, then the antenna can realize the multi-polarization function by using a paraboloid, but it will be very difficult and complicated to realize the multi-polarization function by using a planar array.

The parabolic antenna has the advantages of simple feeding, no grating lobes, wide frequency band, low cost and low energy consumption. The focus and difficulty of the parabolic antenna are the reflector and the feed source, while the focus and difficulty of the array antenna are the feed network and the implementation of variable polarization. For high-gain radar antennas, the number of arrays is large, and a more complex feed network is required.

From the perspective of antenna isolation, the isolation of the area array antenna will be slightly better than that of the parabolic antenna, because the area array antenna is multi-unit distributed radiation, and the marginal power is weak; The lobes formed by the distance will not be very weak at the edges.

After demonstration and analysis, the antenna system adopts the form of a planar array antenna with separate transceivers. The antenna system consists of two parts, the transmitting antenna and the receiving antenna. The transmitting antenna and the receiving antenna are placed up and down, and the transmitting antenna radiates the signal generated by the transmitter into space. The receiving antenna sends the received target echo signal to the receiver.

The main technical indicators of the antenna

(1) Working frequency, K-band, f0±150MHz; f0 is 24GHz;

(2) Antenna form, waveguide planar array antenna (sub-array block design);

(3) output channel and signal channel and coupled signal channel;

(4) The diameter of the antenna is 0.76m×0.16m (subject to meet the requirements of the index);

(5) The antenna system adopts the transceiver separation system;

(6) Antenna gain ≥ 30dB;.

(7) The lobe width is ≤1° horizontally and ≤5° vertically;

(8) Horizontal and vertical side lobe levels ≤ -20dB;

(9) Standing wave requirement VSWR: ≤1.6;

(10) The polarization mode is vertical polarization;

(11) Antenna interface and signal channel: standard BJ260 waveguide interface; coupling channel: SMA-K interface;

(12) Transceiver antenna isolation ≥ 80dB;

(13) The beam pointing deviation meets the electrical axis pointing deviation in the frequency band ≤ 0.2°;

(14) Power capacity 2W.

(15) Environmental adaptability

Working environment temperature: -10℃～+50℃.

Working environment humidity: Relative humidity: 95% (at 40°C).

(16) Reliability

Passive components, lifetime pieces.

Three defenses: waterproof, anti-mildew, anti-salt spray, no obvious mildew and corrosion;

The materials are selected with anti-aging, anti-fatigue and corrosion-resistant materials, and the medium part is subjected to corresponding anti-aging treatment;

(17) Structural design

The design of this equipment implements the idea of generalization, serialization and modularization;

Meet the anti-corrosion, anti-salt spray, electromagnetic environment and other requirements required by the project;

Fasteners are made of stainless steel to prevent corrosion of fasteners;

The structural design of the antenna array pays attention to measures such as sealing and waterproofing;

Carry out structural reliability design, use standard series of components as far as possible, and parts, parts and whole parts of the same type should have good interchangeability;

On the basis of fully ensuring the structural strength, safety and reliability, the overall weight of the structure is reduced by adopting new material technology.

Antenna Composition

In order to meet the overall detection requirements for the observation of aquatic plants, the antenna system adopts a separate antenna system for transmitting and receiving. The antenna forms a pattern characteristic of a horizontal narrow lobe and a vertical wide lobe.

The antenna system is mainly composed of a transmitting antenna, receiving antenna and antenna installation frame. The transmitting antenna is installed on the upper end of the antenna base. The antenna diameter is 0.76m wide × 0.16m high. It is used to transmit and receive signals. Aperture 0.76m wide × 0.16m high, used to receive the target echo signal and send the signal to the receiver.

The antenna system uses a waveguide planar array antenna, and each antenna is composed of a radome, a radiation array, a coupling waveguide, a feed network, and an antenna support frame. The radiating array is composed of several slotted waveguides, and on the back of the radiating waveguides there are some coupling waveguides placed vertically and intersecting, and each coupling waveguide is fed through an H-T joint or a feed network. In this way, the energy is input into the coupling waveguide through the feeding device, then coupled into the radiation waveguide by the coupling slot, and finally radiated out by the radiation slot.

Based on the demand for target detection, the antenna system adopts a continuous wave working system with separate transceivers. Because aquatic plants float on the river surface, they will flow continuously through the observed area as the water flows. In order to ensure the accuracy of detection, the radar needs to observe the target continuously, so the antenna must also work in a continuous wave system. In order to meet the working requirements of the continuous wave, the transmitting antenna and the receiving antenna are separated separately, one transmits and one receives, and they work at the same time.

In order to enhance the adaptability of the product to the climate environment, we pay attention to the environmental protection design, mainly including: rainproof, moistureproof, mildewproof, salt spray proof, rust proof, etc. The antenna system should have a radome, and the electrical performance and structural design of the radome can effectively prevent water or other sundries from entering the waveguide array. In addition, in order to avoid direct or indirect splashing of rainwater on the outer surface of important equipment, scientifically design and arrange the direction of cables to prevent rainwater from entering the interior of the equipment along the cables; in the design, all electronic equipment exposed to the outside are fully considered in sealing design to prevent rainwater Entry can cause damage to electronic equipment. The antenna structure should adopt ventilation design and other technologies to ensure the dryness of the antenna array through air flow. The connection of passive feeder equipment should adopt a sealed design to ensure the protection performance of components to the greatest extent. Use metal materials with better corrosion resistance to avoid contact corrosion of materials. When assembling metal materials that are not allowed to be in contact, take corresponding measures for anti-corrosion treatment;

To avoid water accumulation structure, the outer cover of the equipment shall be designed with proper inclination, and the plane transition shall be designed to be smooth downward, and drainage holes and vent holes shall be opened where water accumulation and moisture may be retained; the design avoids and eliminates the gap structure And lap joint structure, seal and coat the place where crevice corrosion can be formed; the outer edges of all metal parts are rounded; in order to help obtain a properly adhered paint layer or metal coating; paint is required for the surface For the parts, choose a paint that has strong bonding force and stable performance with the metal parts, such as acrylic polyurethane paint; for the metal parts that are not painted, choose a suitable electroplating process or chemical method for anti-corrosion treatment; the working conditions under the structural parts require The surface is coated and the fasteners are made of weathering steel. Welding should be avoided as much as possible for exposed structural parts, and the weld seam must be continuous. Steel structural parts must be treated with heavy-duty anti-corrosion aluminum (zinc) spraying.

Antenna System Related Calculations

(1) Gain estimation

According to the antenna gain calculation formula:

A _s —Antenna effective area

λ—antenna operating wavelength

η _x , η _y —Antenna weighted aperture efficiency

η—antenna efficiency

Substituting the known parameters into the formula, it can be calculated that the antenna gain of 0.76m×0.16m is 35dB.

(2) Selection of the number of antenna array units

The aperture of the transceiver antenna is 0.76m wide x 0.16m high, and the waveguide is a wide-side longitudinal slot array antenna. The waveguide is a non-standard waveguide, and the size is 8.4mm x 4.2mm x 0.95mm. In order to meet the requirements of the standing wave array antenna, the spacing of the element slots is half of the waveguide wavelength. That is 18.70mm. Calculate the number of row elements M in the line source direction as:

M=(L-0.5*λ _g )/(0.5*λ _g )=(760-9.35)/9.35=80.28

According to the calculation, the number of units M is taken as 80.

Since the width of the waveguide is 8.4mm, and the common wall thickness of the waveguide is 0.95mm during overall processing, the number of line source rows is 14 through calculation. That is, the entire array antenna unit is 14×80.

(3) Antenna theoretical pattern calculation

In order to obtain the required antenna sidelobe level, the amplitude of the excitation current of each antenna element in the array is weighted according to a certain illumination function (such as Chebyshev Chebyshev distribution, Taylor Taylor distribution, Hamming Hamming distribution, etc.), this method It is called the amplitude weighting method.

Usually, the lobes of the Chebyshev distributed antenna array have equal side lobes. In the aperture distribution corresponding to such lobes, large currents are often formed on the units at both ends, which is difficult to achieve in engineering practice. Taylor improved this by modifying the structure of the side lobes. His method is to move the positions of several side lobes close to the main lobe in the evenly distributed lobe, so that they have approximately equal levels, and the farther side lobes make them change according to the shape of the lobe when they are evenly distributed, so that It can avoid forming a large current on the cells at both ends. Therefore, Taylor distribution is usually used in engineering design.

Transceiver antenna diameter: 0.76m wide × 0.16m high. According to the antenna theory, in order to achieve low horizontal sidelobe level without reducing the antenna gain too much, it is more appropriate to choose the Taylor function distribution for the horizontal amplitude distribution. According to the antenna theory, for M The pattern of the planar array antenna composed of ×N same antenna elements is:

In the formula: d _x — spacing of radiation elements along the x-axis;

d _y — spacing of radiation elements along the y-axis;

K—2π/λ;

M—the number of radiation units along the x-axis direction;

N—the number of radiation units along the y-axis direction;

— radiation element pattern;

—Aperture amplitude distribution function of element array antenna;

—Aperture phase distribution function of element array antenna.

when , that is, F(θ,0), F(θ,90) are the horizontal pattern and vertical pattern of the planar array antenna, respectively.

If the antenna wants to achieve the -20dB sidelobe level of the antenna, according to the engineering design experience, the theoretical design must be carried out according to the Taylor distribution of the -25dB sidelobe level at least, and it must be realized through the precise amplitude and phase compensation of the antenna. Bring the parameters such as the number of units and spacing into the Matlab program to obtain the theoretical antenna pattern as shown in Figure 3 and Figure 4, and the three-dimensional lobe diagram of the transceiver antenna theory is shown in Figure 5.

Antenna Simulation Design

By analyzing the waveguide planar array antenna whose working frequency is 24GHz. Wherein, the size of the waveguide is 8.4mm×4.2mm×0.95mm. The waveguide wavelength _λg is 18.70mm.

Through theoretical analysis and calculation, the overall simulation calculation is mainly carried out through the commercial software CST. It mainly includes radiating waveguide and coupling waveguide parameter extraction, radiation array simulation calculation, coupled waveguide simulation calculation, feed network simulation calculation and integrated simulation calculation.

Through the conductance parameter extraction, combined with the actual amplitude distribution, the offset of the radiation slot and the inclination angle of the coupling slot are obtained. The simulation model of the transceiver antenna is established as shown in Figure 6, and the calculated three-dimensional lobe diagram is shown in Figure 7. The simulation of the transceiver antenna The lobe diagram of the horizontal plane is shown in Figure 8, and the simulated vertical lobe diagram of the transceiver antenna is shown in Figure 9.

It can be seen from the simulation results that the technical indicators of each antenna system meet the design requirements.

Antenna Performance Analysis for Broadband Signals

The radar uses a 300MHz linear frequency modulation signal, and the center frequency f0 is 24GHz. There is a transmission group delay in the waveguide slot antenna for this signal. Therefore, a deviation in beam pointing will be caused. The beam pointing at the three frequency points of f0-150MHz, f0, and f0+150MHz is simulated and calculated as shown in Figure 10.

Since the 300MHz bandwidth appears and is continuous in one signal cycle, and cannot be separated in the time domain during frequency domain detection, the deviation of beam pointing leads to broadening and deformation of the wave lobe. Finally, the actual beam shape of the antenna system for a 300MHz signal is shown in Figure 11. The beam deviation can be reduced by sub-array division. The final beam broadening is expected to be around 0.4°.

The radio frequency microwave system is composed of three parts: receiving front end, data acquisition module (hereinafter also referred to as A/D module), and frequency synthesis module. It is powered by the DC regulated power supply of the whole machine. In the support arm, work in the outdoor environment.

The frequency synthesis component provides the local oscillator signal required by the receiving front end, the synchronous clock signal required by the ADC module, and the excitation signal for linear frequency modulation.

The frequency synthesis component consists of a clock reference circuit, a waveform generation circuit (hereinafter also referred to as a frequency sweep local oscillator), a transmission channel, an interface control circuit, and a power supply processing circuit. The internal circuit is mainly composed of a constant temperature crystal oscillator with high stability and low phase noise, and a comb spectrum , microwave digital phase-locked loop, mixing filter amplification channel, digital control and power supply voltage stabilization filter circuit and so on.

The clock reference circuit is shown in Figure 12:

The clock reference circuit is a 100MHz signal generated by a constant temperature crystal oscillator, which is output through the ADP-2-1W two-way power divider and one way is used as a comb spectrum excitation source for the 12G source; one way is respectively supplied to 3.5GHz and CRO through the SCA-4-10 four-way power divider Phase-locked source; LTC6946-2 outputs and receives three local oscillator signals at 820MHz; LTC6946-1 outputs and transmits excitation signals at 750MHz; the amplifier outputs 13dBm signals for signal processors as acquisition clocks;

The waveform generation circuit is shown in Figure 13:

The local oscillator signal is generated by the CRO phase-locked loop circuit. The 3.5GHz signal and the comb spectrum generate the 12G signal. After filtering, amplifying, mixing and generating, after filtering, amplifying, and power division, it is used for the transmitting channel and receiving module.

The sweeping local oscillator is divided into two channels by 3.5G, and one is amplified, frequency multiplied and filtered to generate a 7GHz local oscillator signal; the other is used as a clock for the AD9914 to generate a 600-900MHz signal, and the two signals are mixed by the HMC558 mixer and then filtered , amplification, and power distribution for the transmission channel and output to the receiving module as two local oscillators;

The launch channel is shown in Figure 14:

The transmission channel is generated by a frequency synthesizer and mixed with the local oscillator (7.6-7.9GHz) at 750MHz. After filtering and amplifying, it is mixed with the local oscillator (15.5GHz) to output a 23.85-24.15GHz signal, and then filtered and mixed. The frequency and amplification are output through the isolator.

The performance indicators are as follows:

1) Excitation: 24GHz±50MHz (23.85～24.15GHz), power: 1～1.3W, phase noise: L(1K)≤-103dBc/Hz, L(100K)≤-113dBc/Hz;

2) Local oscillator: 15.5GHz, power: 13dBm±1dBm, phase noise: L(1K)≤-108dBc/Hz, L(100K)≤-118dBc/Hz;

3) Two local oscillators: 7.75GHZ (7.6~7.9), power: 10dBm±1dBm; phase noise: better than one oscillator;

4) Three local oscillators: 820MHz, power: 10dBm±1dBm, phase noise: better than two local oscillators;

5) Clock: 100MHz, power: 13±0.5dBm, phase noise: L(1K)≤-140dBc/Hz, L(100K)≤-150dBc/Hz;

6) Amplitude consistency within the FM bandwidth of the frequency sweep signal: ≤1dB;

7) Output clutter: excitation ≥ 60dBc, one local oscillator ≥ 70dBc, two local oscillators ≥ 70dBc, three local oscillators ≥ 70dBc, clock ≥ 70dBc;

8) Harmonic suppression: excitation ≥ 55dBc, local oscillator ≥ 60dBc, two local oscillators ≥ 60dBc, three local oscillators ≥ 60dBc, clock ≥ 60dBc;

9) Power fluctuation: ≤0.5dB

10) Power consumption: ≤30W;

11) Point frequency and linear modulation output alternately for 1s each, linear frequency modulation time 1ms, frequency modulation bandwidth 300MHz, frequency modulation linearity ≤ 2/1000.

The performance index analysis calculation is as follows:

(1) Analysis of main indicators of local oscillator

Factors affecting the phase noise index mainly include the phase noise of the reference source, the low noise of the phase detector chip, and the phase noise of the VCO. The phase noise of the final output signal within 10KHz mainly depends on the phase noise of the reference source and the noise floor of the phase detector chip, and the phase noise beyond 100KHz mainly depends on the phase noise of the VCO.

1) Key device indicators that affect phase noise:

a. Voltage controlled oscillator: ≤-130dBc/Hz@100KHz

b. Constant temperature crystal oscillator: ≤-155dBc/Hz@1KHz

c. Phase detector: -153dBc/Hz@10kHz offset@100MHz

2) Phase noise analysis of the local oscillator:

a. The calculation formula of phase noise within the loop bandwidth: floor+20Log(f0/fpD)+10LogfpD

Among them, Lfloor is the normalized low noise of the PLL chip, f0/fpD is the output frequency divided by the phase detection frequency, that is, the frequency multiplication number N, and fpD is the phase detection frequency.

Bringing the above parameters into the formula calculates:

The phase noise within the loop band is -226+20Log(3500/100)+10Log(100×106)≈-125dBc/H.

In addition, coupled with the deterioration of the actual project and the deterioration of other parameters by 2dB, it can be concluded that the phase noise in the loop band is -123dBc/Hz.

b. Calculation of the deterioration of the phase noise of the reference source (3.5GHz): 20Log(f0/fpD)＝20Log(3500/100)＝31dBc/Hz

Based on the reference phase noise of -155dBc/Hz@1KHz, the degraded phase noise is calculated as:

-155+31＝-124dBc/Hz@1KHz;

Since the phase noise after the reference deterioration is higher than the phase noise in the loop band, the final output phase noise still depends on the phase noise in the loop band, which is -123dBc/Hz@1KHz; since the 3.5GHz signal is also mixed with the 12GHz signal frequency, the final output phase noise depends on a different signal source, the 12GHz signal is in the form of frequency multiplication, and the deterioration of the comb spectrum is: -155+20Log(12000/100)+4＝109dBc/Hz@1KHz, so The phase noise of the final output oscillator is: -109dBc/Hz@1KHz.

c. Loop out-of-band phase noise analysis

The phase noise outside the loop bandwidth mainly depends on the phase noise of the VCO itself, and the specific index can be estimated according to the phase noise of the 100KHz~1MHz phase in the VCO technical index. The out-of-loop phase noise of the local oscillator is estimated as follows: 130dBc/Hz@100KHz.

2) Analysis of spurious suppression degree of local oscillator:

The spurs of the local oscillator mainly include phase detection spurs and frequency mixing spurs. Since the phase detection frequency of the 3.5GHz phase-locked loop is 100MHz, the phase detection spurs will be distributed at 100MHz away from the output frequency. The specific calculation as follows:

For the phase-locked loop of the charge pump type, phase detection spurs mainly include two aspects, one is leakage spurs, and the other is pulse spurs, and the spurious formula is as follows:

Spur＝10log(10 ^{LeakageSpur/10} +10 ^PulseSpur/10 )

Assuming that the phase detection leakage current is 1nA, its two strays are calculated separately below.

LeakageSpur＝BaseLeakageSpur+20log(Leakage/Kφ)+20logCL(s)|

＝16.0+20log(1nA/5mA)+20log|CL(s)|

＝-118+20log|CL(s)|

PulseSpur＝BasePulseSpur+40log(Fspur/1Hz)+20logCL(s)|

＝-306dBc+40log(Fcomp/1Hz)+20log|CL(s)|

＝-10+20log|CL(s)|

Spur≈PulseSpur＝-10+20log|CL(s)|

The leakage current of the phase detection is 1nA, and the BasePulseSpur is not the determined -306dBc, but around this value. Since the phase detection frequency is 100MHz, which is relatively high, phase detection spurs are mainly determined by pulse spurs, and the loop bandwidth is generally set to be less than 500kHz, so the low-pass nature of the loop filter can suppress the spurs very well. According to past experience, this program selects a loop bandwidth of about 500kHz, which can suppress spurs below -85dBc. The strays generated by frequency mixing are shown in the filter stray distribution diagram, and the spurious output of mixing filter is greater than 75dBc.

(2) Analysis of other indicators

The indicators of two local oscillators, three local oscillators, and the spurious and phase noise of the emission excitation signal are all better than those of the one oscillator, without technical difficulty. The indicators obtained according to the experimental verification are as follows:

750M phase noise index:

124dBc/Hz@1KHz,

122dBc/Hz@100KHz,

820M phase noise index:

123dBc/Hz@1KHz,

121dBc/Hz@100KHz,

7G phase noise index (3.5GHz multiplier):

-155+31＝6＝-116dBc/Hz@1KHz;

-120dBc/Hz@100KHz;

Spurious index: ≥75dBc.

The output power is determined by the output amplifiers of each local oscillator. It can be seen from the technical indicators of the amplifiers that there is a large margin, so it is not difficult to realize.

DDS bit noise metrics and spurs:

Noise index: -128dBc/Hz@1KHz,

-133dBc/Hz@100KHz.

DDS spurious: broadband spurious: 55dBc

The measured 20KHz step data in the 500KHz narrow band is 75dBc.

The main technical indicators of the above-mentioned 3500MHz CRO oscillator are shown in Figure 15.

The receiving channel circuit is shown in Figure 16:

Receive channel signal input 23.85GHz ~ 24.15GHz through the limiter, low noise amplifier, filter, a frequency mixing to get 7.75GHz bandwidth 300MHz IF, after the filter amplified into the second stage mixer to get the second IF 750MHz ( The signal bandwidth is 500KHz), after being filtered and amplified by the digital control attenuator, it enters the third-stage mixer to output the second intermediate frequency of 70MHz, and is output by the low-pass filter after being filtered and amplified by the digital control attenuator.

Its main technical indicators are:

(1) Echo frequency: 24GHz±150MHz;

(2) Local oscillator frequency: 15.5GHz;

(3) Two local oscillator frequencies: (7.75MHz±150MHz);

(4) Three local oscillator frequencies: 820MHz;

(5) Noise figure: ≤4.5dB (low, normal temperature), ≤5dB (normal temperature);

(6) The cavity is designed with a 5-position switch, which is used to adjust the gain of the receiving channel, with a step of 1dB, and the cumulative error of attenuation is ≤1dB;

(7) Channel gain: 50±1dB, when the attenuation is 0;

(8) Pin1dB≥-20dBm (when the attenuation is 20dB);

(9) Pout1dB≥+10dBm (when the attenuation is 0dB);

(10) RF filter bandwidth: BW-1dB≥300MHz (f0=24GHz);

BW-3dB≤500MHz;

Out-of-band suppression: ≥60dB(f0±2G);

(11) Echo channel image frequency suppression: ≥70dB (corresponding to the first intermediate frequency and local oscillator);

(12) The isolation of the receiving channel to the frequency synthesis component: ≥80dB;

(13) IF frequency: 70MHz;

(14) Bandwidth of intermediate frequency bandpass filter: BW-1dB≥5MHz;

BW-40dB≤40MHz;

(15) The maximum withstand power of the limiter (CW): ≥1.5W;

(16) Echo input port standing wave ratio: ≤1.5;

(17) Power consumption: ≤10W.

Analysis and calculation of technical indicators

(1) Receiver bandwidth calculation

According to the principle of chirp radar, the distance of the target is obtained by measuring the spectrum deviated from the intermediate frequency. The calculation formula is as follows:

In the formula, f _b is the difference frequency of the intermediate frequency signal; Δf is the linear frequency modulation bandwidth; R is the target distance; Tm is the modulation time; C is the speed of light.

According to the overall requirements of the radar, the linear frequency modulation bandwidth is 300MHz, the maximum target distance is 200m, and the modulation time is 1ms. It can be calculated that the maximum bias frequency fb is 400kHz. According to actual needs, reserve sufficient bandwidth, and the receiver bandwidth is designed to be 500kHz.

(2) Noise figure, gain, output P-1dB power

NF=NF1+((NF2-1)/GP1)+((NF3-1)/(GP1*GP2))+((NF4-1)/(GP1*GP2*GP3);

Receiving channel: the first stage is a waveguide conversion, and its insertion loss is 0.4dB; the second stage is a limiter of 0.75dB, the third stage is a low-noise block amplifier, and the fourth stage is an echo filter with an insertion loss of 1.5 dB, post-mixer 8dB, etc.

Noise figure, gain, output P-1dB power calculation

Noise figure: 3.73dB;

Gain: 51.3dB;

Output P-1dB compression point: +11.88dBm

(3)Limiter

The maximum input power of the low noise amplifier is 18dBm, and the limiter index parameters are shown in Table 3:

table 3

(4) echo input standing wave

The receiving input standing wave is determined by the limiter and low noise amplifier, and the echo standing wave is <1.5, and its curve is shown in Figure 17.

(5) Image frequency rejection, as shown in Figure 18:

Receiving image frequency suppression: receiving down-conversion, relatively easy to deal with clutter, a filter is added after the low-noise amplifier to filter out signals other than 23.85GHz to 24.15GHz, and a high-low pass filter is added to the intermediate frequency. The RF signal is suppressed, and a bandpass filter is added to the second intermediate frequency local oscillator. The frequency of the first oscillator is 15.5GHz, and the frequency is down-converted. Therefore, the mirror frequency is 7.15GHz to 7.45GHz. From the perspective of filter suppression, the suppression is greater than 90dBc.

A/D module

The A/D module, that is, the data acquisition module, mainly realizes the digitization of the intermediate frequency signal. Since the intermediate frequency of the receiver output signal is 70MHz, the actual useful bandwidth is only 500kHz, and in order to reduce the amount of data processed by the back-end output. Therefore, consider using under-sampling and decimating to achieve a low data rate.

As shown in Figure 19, the data acquisition module mainly includes four parts: anti-aliasing filter, ADC, FPGA and Ethernet transmission module. The interfaces include clock input XS1, intermediate frequency input XS2, power input interface XS3, communication interface XS4 and synchronization interface XS5.

The ADC requires an effective bit of 12.5 bits. The LTC2207 of Linear Company is selected. Its main performance parameters are as follows:

(1) Input voltage range (Vpp): 2.25V (11dBm);

(2) Maximum sampling frequency: 105MSPS;

(3) No false peak dynamic range (SFDR): 82dB;

(4) Noise Floor: 77.3dBFS;

(5) Valid bits: 12.9 bits.

Anti-aliasing filter design

The anti-aliasing filter is mainly used to prevent noise aliasing during ADC bandpass sampling. Anti-aliasing bandpass filter parameters are as follows: center frequency F0＝70MHz; BW-1dB＝3～5MHz; BW-40dB<40MHz; BW-80dB<70MHz

The aliasing diagram of the out-of-band noise is shown in Figure 20, and the aliasing to the in-band noise intensity is less than -77dB, which is lower than the 12.5 effective bits of the ADC.

FPGA implementation

The FPGA uses the XC7K325T-1FFG900I of the K7 series of XILINX Company, and the signal processing flow is shown in Figure 21.

The ADC sampling data enters the decimation filter after digital down-conversion. There are three stages of decimation filters (2 times decimation, 5 times decimation and 5 times decimation), which is equivalent to a sampling rate of 2MHz, and then the data is passed through a high-pass filter. Packed into a fixed format and sent to the transmission module.

The Ethernet transmission module transmits the data to the computer through the network port. The computer solves the data through the unpacking software, writes it to the hard disk according to the data length of 1s, and marks the time and signal format information. At the same time, the data can also be put into the designated memory for back-end data processing.

Data processing and terminal display system

The data processing and terminal display system consists of data processing software and terminal display software. Complete the online real-time calculation and display of various information of aquatic plants. The terminal system processes, transforms, and calculates the data detected by the radar to produce the required data and image products. The hardware selection of the system mainly considers the versatility and reliability of the hardware platform and uses a PC. The data of the radar echo is transmitted to the PC through Ethernet.

The choice of software working platform should take into account the versatility, compatibility and maintainability. It includes two aspects of the computer's operating system and the programming language of the application. The choice of software working platform directly affects the efficiency of software development, portability and the good operation of the entire system. It is based on this consideration that we choose Windows as the operating system and Visual C++ as the programming language.

The overall structure of the data processing software system

All the software of this system is built under WindowsXP/7, developed by Visual C++, and has a unified operation interface. All settings are designed in a menu-driven manner, and are completed by the system setting program, which can provide users with interactive selection or input parameters, and also allows users to save the results to disk so that various radar parameters can be set. System recall and modification.

Real-time parameter setting and timing parameter setting generate a real-time parameter table and a timing parameter table, which are called by the radar real-time processing program. The radar real-time processing controls the working status of the radar, collects data and displays the real distribution of aquatic plants. It includes front-end real-time processing software and background real-time processing software. The radar real-time processing program saves the collected raw polar radar data to disk.

The product parameter table is generated by the product parameter setting program.

The product generation table is set up by the product generation setup program.

Raw data is the starting point for the radar output product generation process. All radar output products are generated by the corresponding product generation processing program, calling the original data, and generating under the joint action of the product parameter table and the product generation table.

In the fully automatic product generation mode, the user sets the required product list each time the system processes products, and the system calls this table through batch job processing to generate and archive the corresponding image product files and data product files. So that the system distributes to the corresponding users under the control of the product distribution table or is invoked by other users through the network or other communication devices.

Adopt a unified user interface. The basic screen size is 640X480 (unit is pixel point), if the screen size is changed, the length and width will be adjusted according to the same ratio. With 16 kinds of colors as product layers, it provides convenience for radar mosaic and radar networking. The product information area in the picture includes: color comparison table, time, date, display distance, antenna azimuth, antenna elevation angle, display height, repetition frequency, radar station name, etc. The additional information display area is mainly to display some information specially required by users, such as the intensity and location information of strong echoes, the files used for secondary product display, and graphics can also be displayed, such as real-time intensity observation, which can simultaneously display the real-time observed speed echo.

The graphic display working area is to display the radar real-time scanning echo diagram and the secondary product diagram. The basic size is 480X480 pixels. The abscissa shows the time axis in seconds. The vertical axis shows the width of the river with a resolution of 1 meter. Draw the data every 1 second to display the real-time distribution and intensity of radar echoes on the river surface. As time goes by, the graphics will be scrolled and updated to display the distribution and status of radar echoes for a period of time. In the later stage, the corresponding relationship between the radar data and the distribution of water plants will be established, that is, the distribution and intensity of water plants will be demonstrated.

System settings include real-time parameter settings, timing parameter settings, product parameter settings, product generation settings, product distribution table settings, etc. All of these functions are performed by the system setup program. This program makes full use of the interface functions of WindowsXP/7, and uses the pull-down menu drive method to provide users with the ability to select, modify, and view various radar parameters. The parameters set by the user are correspondingly generated by the system setting program as a real-time parameter table, a timing parameter table, a product parameter table, a product generation table, and a product distribution table. These parameter tables are called by different application programs to control the system to perform different functions or generate different radar output products. This not only greatly facilitates the user to add different functions, but also improves the scalability of the system. Document management is to manage all data and product materials generated by the system, including automatic and human-computer interaction.

The original file is the data collected from the signal processing of the radar real-time processing program during each run, which is different from the radar output product file, and the original files formed by different working methods are different. File management is to provide a program to manage the files formed by various radar products on the disk.

In order to improve the transmission speed of output products, the system also provides compression and recovery functions for processing files. When the user sets the transmission method as the compression method, the file will be automatically compressed before the product is transmitted, and then transmitted. At the same time, when the received file is a compressed file, it will be restored to the original format automatically.

The system uses a watchdog software, through the user's parameter setting, the watchdog software runs automatically or by human-computer interaction. When running in automatic mode, the data and image files set by the user as obsolete will be deleted automatically. When running in human-computer interaction mode, the watchdog software does not automatically delete obsolete files, but gives an alarm prompt when the disk space is lower than the user's set value, so that users can use file management to save files and delete unnecessary files . The file management program provides settings for the watchdog software and data compression. The specific settings include the time to retain data, compression, and the size of the remaining space. The setting of the time to retain data tells the watchdog that the files before this time in the automatic mode can be deleted. Compression settings can be selected in two ways: compression and non-compression. In the human-computer interaction mode, if the remaining disk space is less than the set value of the remaining space, the watchdog will give an alarm message. Set the working mode of watchdog as automatic mode and human-computer interaction mode.

data processing module

Using the difference in the reflection ability of rivers and aquatic plants, detect aquatic plants floating on the water surface. The area of aquatic plants is comprehensively calculated by measuring the distribution of aquatic plants flowing through the measurement section and the moving speed under the section.

(2) 1s linear frequency modulation continuous wave signal processing method

The linear frequency modulation continuous wave signal realizes the area measurement of the floating objects on the water surface. The echo intensity of each distance point in the cross-section dimension is measured by the broadband signal, and the detection is carried out in the frequency domain, and the area of the aquatic plants is determined by the change of the echo intensity of the presence or absence of aquatic plants.

Continuously send 1000 1ms chirp continuous wave signals, process the data of the 1ms and the data of the 1000ms.

①Under normal conditions, find a piece of water surface without floating plants, measure the water reflection intensity of each resolution unit within the azimuth beam width, and use it as the detection reference threshold.

②Detect the echo signal of 1ms, take the value obtained in ① as the reference threshold, adjust according to the actual situation, and judge whether there are floating plants in each resolution unit.

③ Save the intensity value of each distance point, and mark the distance points with floating plants.

(3) 1s point frequency continuous wave signal processing method

The point frequency signal is only used for speed measurement of floating objects on the water surface.

1) Calculate the power spectral density of the collected data;

2) Take 1kHz data at an intermediate frequency of 500KHz for threshold detection. If the threshold exceeds the threshold, it is considered to be a target (under normal circumstances, the general water speed is less than 0.1m/s, f _d = 2v/λ, so the maximum Doppler frequency is about 16Hz, take 1KHz on both sides with a margin in consideration of abnormal conditions such as wind). The threshold value is fine-tuned based on ① in 1.2.

3) Obtain the target speed according to the detection frequency.

(4) Calculate the target area

The width of the river is calculated as 50 meters, divided by the distance resolution unit of 0.6 meters, and the width of the azimuth wave velocity is 1°, so the distance of the farthest point in the azimuth beam is about 0.8 meters.

1) The target area is small

When the area of the target is less than or equal to the area of a distance resolution unit (<=0.6*0.6), as shown in target 1 in Figure 22, the echo intensity measured at this time is a bell-shaped curve with a steeper change. The time dimension target length L ₂ of the water flow in the same direction is:

Wherein: 2 represents the measurement time interval of 2 seconds, v _i represents the water flow velocity (=water floating plant measurement speed/sinθ) obtained at each measurement time interval, n represents the number of time intervals in which targets appear continuously, and d is a fixed value , refers to the lateral distance within the azimuth beamwidth corresponding to different distance unit points in the section dimension, as shown in Figure 22.

According to the distance unit where the target appears, the length L ₁ of the target cross-sectional dimension = length of the measurement cross-sectional dimension * cosθ can be obtained.

Then the area: S=L ₁ ×L ₂

2) The target area is large

The target whose time and distance units appear consecutively is regarded as a large target, such as target 2, target 3 and target 4 shown in FIG. 22 . The measured echo intensity change at this time is similar to a band-pass filter, with steeper changes on both sides and gentler changes in the middle, then the time-dimension target length _L2 in the same direction as the water flow is:

L ₂ =2×v _i

Among them: 2 represents the measurement time interval of 2 seconds, and v _i represents the water velocity obtained from each measurement time interval.

According to the distance unit where the target appears, the length L ₁ of the target distance dimension (=measurement section dimension length*cosθ) can be obtained. If the distance unit of the measured target section dimension is discontinuous, then N refers to the number of distance units with water floating plants in the measured section dimension.

Then the area of floating plants measured at each time interval is: S _i =L ₁ ×L ₂

Then the target area is:

n represents the number of time intervals in which targets appear continuously, d' refers to the maximum value of the lateral distance within the azimuth beam width corresponding to different distance unit points in the cross-sectional dimension of the target, v' refers to the last detection speed of the current target, and m refers to the repeated calculation of the area The number of time intervals, S' refers to the double calculation area of the water phytoplankton within the beam width.

Gives a total area of flow through the measuring section every half hour and saves the result.

(5) Data storage

1) Raw data storage

The stored data is 16bit floating point number, with a data rate of 2M per second, 4MB per second, 14GB per hour, and 24 hours a day uninterrupted storage requires about 340GB of storage space. If you store data for three consecutive days, you need 1020GB. The hard disk of the prototype test computer is 1.2T, so it can only save the data in the last three days. After installing the hardware, if you want to store more data, you can consider wireless data transmission and transfer the data to the data center for storage. The specific implementation method needs to be further considered.

2) Storage of test results

Including the detection time (year, month, day, hour, minute, second), distance, and intensity information, the detection results are stored every second, and half an hour is the time interval (tentative, the specific results are subject to follow-up tests), and every half hour is stored as a file, giving the area value of phytophyte flowing through the measurement section every half hour.

(6) Data transmission, processing, storage and display time calculation

In order to ensure the real-time nature of data processing, it is required that the data transmission processing display storage time can adapt to the data collection time. The entire data transmission process is shown in Figure 21. The performance of the computer for the data processing of the principle prototype is tentatively set at the same performance as the existing server in the signal processing department. The processor is Intel E5, 2.4G, 8 cores, and 64G memory.

After A/D, the data is sent to the data buffer area through the network cable. The input signal is a 16bit floating point number, and there are 2M sampling points in 1S, so the amount of data to be transmitted is 32Mbi.

The data transmission is planned to adopt 10Gigabit network, the transmission rate of 10Gigabit network is 10000Mb/s, and the network transmission efficiency is set to 50% (each layer of network protocol analysis occupies bandwidth), and it takes 6.4ms to transmit 32Mb data, and write the data into Hard disk, the storage rate of computer hard disk is 20MB/s, and it takes 200ms to store 2M points of 16bit data. Reading data from the hard disk to the memory, the computer hard disk reading rate is 30MB/s, and it takes 140ms to read 2M points of 16bit data. From the memory to the CPU, the transfer rate of the PCI-E bus is 8Gb/s, the time for 32Mb data from the memory to the CPU is 4ms, and it takes 600ms for the CPU to perform FFT processing and threshold detection on the data. The display adopts 1s data to display at the same time, and the required time is about 50ms. Therefore, the processing time of 1s chirp signal is about 1 second.

Then there is the point frequency signal of 1s. We only take a small part of the data for processing to find the speed, and the time is about 200ms.

Therefore, the processing time of 1s linear frequency modulation signal + 1s point frequency continuous wave is about 1200ms, so the repetition time interval of 2s can complete the processing without data loss.

The above descriptions are only preferred embodiments of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and Modifications can be made within the scope of the ideas described herein, by virtue of the above teachings or skill or knowledge in the relevant art. And the modification and change that those skilled in the art make do not depart from the spirit and scope of the present invention, then all should be within the protection scope of the appended claims of the present invention.

Claims (16)

1. water floats plant radar sensing system, the radar sensing system is by antenna system, radio frequency microwave system, data processing and end Hold display system composition, it is characterised in that：

The antenna system is made of transmitting antenna and reception antenna, and radar generates point frequency continuous wave and linear frequency modulation continuous wave letter Number, wherein point-frequency signal is only used for the tachometric survey of floater, and linear frequency modulation continuous wave signal realizes floater Area measurement, the signal are radiate via transmitting antenna, and signal enters reception antenna after water plant is reflected；

The radio frequency microwave system is made of receiving front-end, data acquisition module, frequency synthesis component three parts, through water plant Signal after reflection accesses receiving front-end, through filter amplifier, numerical-control attenuator, low-pass filter after the processing of three-level frequency mixer Output；Data acquisition module is made of anti-aliasing filter, ADC, FPGA and Ethernet transmission module, for realizing intermediate-freuqncy signal Digitlization, frequency synthesis component are made of clock reference circuit, Waveform generating circuit, transmission channel, interface control circuit, are used for Improve the local oscillation signal needed for receiving front-end, the synchronizing clock signals needed for data acquisition module and chirped excitation letter Number；

Data processing and terminal display system are made of data processing module and terminal display module, are completed water and are floated each information of plant Online real-time measuring and calculating and display.

2. water according to claim 1 floats plant radar sensing system, which is characterized in that the antenna system is using transmitting-receiving The planar array antenna system split, antenna form horizontal narrow lobe, the pattern characteristics of vertical width lobe.

3. water according to claim 2 floats plant radar sensing system, which is characterized in that the planar array antenna system It is planar waveguide array antenna, using the structure of submatrix block design, working frequency is K-band, and f0 ± 150MHz, f0 take 24GHz。

4. water according to claim 3 floats plant radar sensing system, which is characterized in that the antenna gain >=30dB, Lobe width level≤1 °, vertical≤5 °, horizontal, vertical minor level≤- 20dB；Standing wave requires VSWR≤1.6, polarization mode For vertical polarization, isolation between transmitting and receiving antenna >=80dB；Beam position deviation electric axis in frequency band is directed toward deviation and meets≤0.2 °.

5. water according to claim 1 floats plant radar sensing system, which is characterized in that the three-level frequency mixer includes the Level-one frequency mixer, second level frequency mixer, third level frequency mixer；

The rf excitation signal enters first order frequency mixer through limiter, low-noise amplifier, filter and is mixed to obtain The first intermediate-freuqncy signal of 7.75GHz bandwidth 300MHz；

The filtered device of first intermediate-freuqncy signal, amplifier enter second level frequency mixer and obtain the second intermediate frequency of 750MHz bandwidth 500KHz Signal；

Second intermediate-freuqncy signal obtains 70MHz bandwidth after filtered device, amplifier, numerical-control attenuator into third level frequency mixer The third intermediate-freuqncy signal of 5MHz；

The filtered amplifier of third intermediate-freuqncy signal, numerical-control attenuator, low-pass filter output.

6. water according to claim 5 floats plant radar sensing system, which is characterized in that the anti-aliasing filter is main For preventing noise aliasing when ADC bandpass samplings, parameter from being：Centre frequency F0=70MHz；BW-1dB=3~ 5MHz；BW-40dB<40MHz；BW-80dB<70MHz.

7. water according to claim 6 floats plant radar sensing system, which is characterized in that the FPGA is by ADC hits According to three-level decimation filter is entered after Digital Down Convert, 2 times of extractions, 5 times of extractions, 5 times of extractions are followed successively by, 2MHz is equivalent to Sample rate, set form is then packed the data to after a high-pass filter and is sent to Ethernet transmission module.

8. water according to claim 7 floats plant radar sensing system, which is characterized in that the clock reference circuit is by perseverance Warm crystal oscillator generates 100MHz signals, does comb spectrum driving source to the sources 12G all the way through the output of two power splitters of ADP-2-1W, passes through all the way Tetra- road power splitters of SCA-4-10 are supplied respectively to 3.5GHz, CRO phase locked sources；LTC6946-2 outputs receive three local oscillation signal 820MHz； Transmitting pumping signal 750MHz is exported through LTC6946-1；Through amplifier export 13dBm signals for signal processor when acquisition when Clock.

9. water according to claim 8 floats plant radar sensing system, which is characterized in that the Waveform generating circuit work It is as follows：

Generation to be mixed after, amplification filtered with comb spectrum generation 12G signals latter by CRO phase-locked loop circuits generation 3.5GHz signals Local oscillation signal, a local oscillation signal is filtered, amplification, for transmission channel and receiving module does a local oscillator after work(point；

Frequency sweep local oscillator divides two-way to do local oscillation signal through amplification, frequency multiplication, filtering generation 7GHz all the way by 3.5G；It is done all the way to AD9914 Clock, generates 600~900MHz signals, and two kinds of signals are logical for emitting through filtering, amplification, work(point after the mixing of HMC558 frequency mixers Road and output do two local oscillators to receiving module.

10. water according to claim 9 floats plant radar sensing system, which is characterized in that the transmission channel is by frequency Synthesizer generates 750MHz and is mixed with 7.6~7.9GHz of frequency sweep local vibration source, mixed with a local oscillator 15.5GHz after filtered amplification Frequently, 23.85~24.15GHz signals are exported, are exported through isolator by filtering mixing, amplification.

11. floating plant radar sensing system according to claim 1-10 any one of them water, which is characterized in that the items of system Performance indicator is as follows：

1) it encourages：24GHz ± 50MHz, power：1~1.3W, phase noise：L (1K)≤- 103dBc/Hz, L (100K)≤- 113dBc/Hz；

2) local oscillator：15.5GHz, power：13dBm ± 1dBm, phase noise：L (1K)≤- 108dBc/Hz, L (100K)≤- 118dBc/Hz；

3) two local oscillator：7.75GHZ, power：10dBm±1dBm；Phase noise：Better than a local oscillator；

4) three local oscillator：820MHz, power：10dBm ± 1dBm, phase noise：Better than two local oscillators；

5) clock：100MHz, power：13 ± 0.5dBm,

Phase noise：L (1K)≤- 140dBc/Hz, L (100K)≤- 150dBc/Hz；Wherein 1K indicates offset signal 1KHz, i.e., Measured signal 100MHz deviates the intensity of phase noise at frequency point 1kHz, and 100K indicates the meaning, and principle is identical therewith；

6) amplitude coincidence in swept-frequency signal frequency modulation band bandwidth：≤1dB；

7) clutter is exported：Excitation >=60dBc, one local oscillator >=70dBc, two local oscillators >=70dBc, three local oscillators >=70dBc, clock >= 70dBc

8) harmonics restraint：Excitation >=55dBc, one local oscillator >=60dBc, two local oscillators >=60dBc, three local oscillators >=60dBc, clock >= 60dBc

9) power fluctuation：≤0.5dB

10) power consumption：≤30W；

11) it puts and alternately exports each 1s, linear frequency modulation time 1ms, modulating bandwidth 300MHz with linear modulation frequently, fm linearity≤ 2/1000。

12. water according to claim 11 floats plant radar sensing system, which is characterized in that the data processing module profit With the difference in river and water plant albedo, the water plant of floating on water is detected.Measure section is flowed through by measuring Water plant distribution situation, and movement speed under the section carry out the area of COMPREHENSIVE CALCULATING water plant.

13. water according to claim 12 floats plant radar sensing system, which is characterized in that the data processing module pair The linear frequency modulation continuous wave signal processing method of 1S is as follows：

Linear frequency modulation continuous wave signal realizes the area measurement of floater, and measuring section by broadband signal ties up each distance The echo strength of point, is detected in frequency domain, and the face of water plant is determined by the way that whether there is or not the variations of the echo strength of water plant Product；

The continuously linear frequency modulation continuous wave signal of 1000 1ms of hair, handles the data of the data and 1000ms of 1ms；

S01：Under normal operation, one piece of water surface for floating plant without water is looked for, each resolution cell in azimuth beamwidth is measured The reflected intensity of water, as detection benchmark threshold value；

S02：The echo-signal of 1ms is detected, the value obtained using in S01 as benchmark thresholding, according to actual conditions into Row adjustment, judges that each resolution cell has anhydrous floating plant；

S03：The intensity value for preserving each range points makes label to the range points for having water to float plant.

14. water according to claim 13 floats plant radar sensing system, which is characterized in that the data processing module pair The point frequency continuous wave signal processing method of 1s is as follows：

S11：Power spectral density is asked to gathered data；

S12：It respectively takes the data of 1kHz to carry out Threshold detection in intermediate frequency 500KHz or so, thinks there is target more than thresholding, normally In the case of, general water speed is less than 0.1m/s, f_d=2v/ λ, therefore maximum doppler frequency is about 16Hz, consideration has wind etc. abnormal In the case of there are surplus both sides, and 1KHz, the detection benchmark threshold value that threshold value is obtained using in S01 respectively to be taken to be finely adjusted as benchmark.

S13：Target velocity is found out according to detection frequency.

15. water according to claim 14 floats plant radar sensing system, which is characterized in that the data processing module pair Target area calculates as follows：River width divides for 0.6 meter by 50 meters of calculating, by Range resolution unit, 1 ° of orientation velocity of wave width, then orientation About 0.8 meter of farthest distance in wave beam；Its calculation formula is as follows

R=river width；

θ=azimuth beamwidth；

Then at different distance, orientation width=2Rtan (θ/2) of azimuth beam covering；

S21：Target area is smaller

When the area of target is less than or equal to the area of a Range resolution unit, the echo strength measured at this time is to change relatively suddenly Bell-shaped curve, the flow measured at this time time dimension target length L in the same direction₂For：

Wherein：2 represent time of measuring interval 2 seconds, v_iIt represents water velocity=water that each time of measuring interval is found out and floats plant survey Speed/sin θ is measured, n representatives continuously the time interval number of target occur, and d is a fixed value, refer to section dimension different distance list Lateral distance in the corresponding azimuth beamwidth of member point；

The range cell occurred according to target, you can find out the length L of object section dimension₁=measure section dimension length * cos θ；

Then area：S=L₁×L₂

S22：Target area is larger

For the target that time and range cell are all continuously occurred as a big target, the echo strength variation measured at this time is similar Bandpass filter, both sides variation is steeper, and centre is shallower, the then time dimension target length L in the same direction with flow₂For：

L₂=2 × v_i

Wherein：2 represent time of measuring interval 2 seconds, v_iRepresent the water velocity that each time of measuring interval is found out.

The range cell occurred according to target, you can find out the length L of target range dimension₁=section dimension length * cos θ are measured, such as The object section dimension range cell that fruit is surveyed is discontinuous, thenN, which refers to, measures the range cell that section dimension has water to float plant Number；

The water that then each time interval is measured floats plant area：S_i=L₁×L₂

Then target area is：

N, which is represented, continuously there is the time interval number of target, and d' refers to the orientation corresponding to object section dimension different distance unit spot The maximum value of lateral distance in beam angle, v' refer to current goal last time detection speed, m refer to that area computes repeatedly when Between be spaced number, S' refers to computes repeatedly area by what the water in beam angle floated plant, is provided per half an hour and flows through measurement One gross area in section simultaneously preserves result.

16. water according to claim 15 floats plant radar sensing system, which is characterized in that the data processing module is also It is specific as follows including calculating data transmission, processing, storage and display time：

For data by the time of cable incoming data buffer area, input signal is 16bit floating numbers, the total 2M sampling of 1S after A/D Point, therefore the data volume of required transmission is 32Mbit；

Data transmission is quasi- to use 10,000,000,000 nets, and the transmission rate of 10,000,000,000 nets is 10000Mb/s, if network transmission efficiency is 50%, is passed The data of defeated 32Mb need to take 6.4ms；

Hard disk is write data into, computer hard disc memory rate 20MB/s deposits 2M point 16bit data and needs 200ms, is read from hard disk Data are to memory, computer hard disc reading rate 30MB/s, read 2M point 16bit data and need 140ms, CPU, PCI-E are stored to from interior The transmission rate of bus is 8Gb/s, and the data of 32Mb are 4ms from the interior time for being stored to CPU；

CPU carries out FFT processing to data and Threshold detection needs 600ms, display to be shown simultaneously using 1s data, and required time is 50ms, therefore the linear FM signal processing time of 1s is about 1 second.

Followed by the point-frequency signal of 1s, we only take wherein a bit of data processing, and it is about 200ms to ask speed, time；

Therefore the point frequency continuous wave processing time of the linear FM signal+1s of 1s is about 1200ms, so between the repetition time of 2s Every that can complete to handle, data will not be lost.