A ring-shaped distributed FSR sensor network sea surface early warning method

Abstract

The present application relates to radar networking detection technical field, especially to a kind of encircling distributed FSR sensor network sea surface early warning method.The present application designs a kind of encircling distributed FSR system, uses fewer network nodes, and is simple to lay out, and can detect the foreign target invading in any direction.FSR node is small, low in power consumption, low in cost, and radar cross section increases rapidly in forward scattering area, so that the system is agile to the reaction of fast small target on sea surface, such as suicide attacker, drug and arms smuggling, speedboat and motorboat dangerous target, can detect target under all-weather conditions, so as to realize the monitoring and early warning of threat target.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of radar networking detection, and particularly relates to a sea surface early warning method of a distributed FSR sensor network in a surrounding shape. BACKGROUND

[0002] At present, many important sea areas and facilities, such as important activity sites, ports, wharfs, aircraft carriers, military bases and offshore drilling platforms, are threatened by small unmanned ships and frogmen who can easily carry explosive devices and weapons. Such targets have the characteristics of flexible movement, small individual size and uncertain time. In the confrontation with such targets, the radars of traditional ground-based platforms are difficult to protect a large range of sea areas due to the line-of-sight limitation; the radars of airborne or shipborne platforms can realize large-range sea area protection, but are disturbed by strong clutter, depend on weather and have high cost; and the radars of satellite platforms have powerful functions but long revisit time, so the existing systems are difficult to effectively protect exclusive economic zones, coastal borders, important ports and important sea platforms.

[0003] A new radar system, forward scattering radar (FSR), gradually enters the field of view of researchers. FSR is a bistatic radar system with a bistatic angle between 135° and 180°. FSR can receive signals in the forward scattering area of a target, in which the radar cross section (RCS) of the target rapidly increases, generally by 10 to 50 dB, and is independent of the material and shape of the target, which is of great significance to the detection of small targets and anti-stealth. FSR systems have the advantages of simple structure and low power consumption, and are also very conducive to the formation of a network to detect sea moving targets, and can work well in various environments (such as near the sea, islands, deep sea drilling platforms, etc.), and have strong survivability.

[0004] A distributed FSR sensor network is formed by arranging small sea floating forward scattering sensor nodes around important sea areas (such as important bases, platforms, scenic spots and military targets) or along key coastal lines, and cooperatively sensing, collecting and processing information of sensed objects in the network coverage area. There are mainly two kinds of distributed FSR networking methods, one is a zigzag-shaped distributed FSR system, which can form a long chain and be arranged along a specific coast or near-sea area; and the other is a parallelogram-shaped distributed FSR system, which takes a parallelogram as a basic topological unit and is mainly used for ground networking to achieve full coverage of the airspace.

[0005] However, the existing distributed FSR networking methods only have a zigzag-shaped long chain and a parallelogram-shaped airspace full coverage with a basic topological unit, and cannot meet the needs of protecting isolated sea targets such as ocean military bases and drilling platforms around the sea targets, so there is an urgent need to design a new distributed FSR method that can surround the sea targets. SUMMARY

[0006] The technical solution of the present application is to overcome the shortcomings of the prior art and provide a ring-shaped distributed FSR sensor network sea surface early warning method.

[0007] The technical solution of the present application is:

[0008] A ring-shaped distributed FSR sensor network sea surface early warning method comprises the following steps:

[0009] First, according to the elliptical coverage principle, small offshore floating forward scattering sensor nodes are arranged around important sea areas (such as important bases, platforms, scenic spots and military targets) to form a distributed forward scattering sensor network. Each node in the formed sensor network has the ability of self-positioning, transmitting / receiving signals and real-time information transmission.

[0010] Second, the sensor network formed in the first step is used to preprocess the received echo signal to obtain effective observation data.

[0011] The echo signal is: if any target passes through the coverage area, the echo signal is the superposition signal of the electromagnetic wave signal emitted by the transmitting node scattered to the receiving node and the direct wave signal; if there is no target passing through the coverage area, the echo signal is the direct wave signal.

[0012] Third, according to the linear frequency modulation characteristic of the effective observation data obtained in the second step, matched filtering is carried out, and the result after matched filtering is processed by a mean value CFAR detector to realize target detection.

[0013] In the first step, the distributed forward scattering sensor network takes an important sea platform as the center, and the radar nodes are distributed on the inner and outer concentric circles to construct a two-dimensional coordinate system in the horizontal plane, taking the center of the concentric circle as the coordinate origin, the x-axis pointing to the first node of the inner circle, the inner circle area of the concentric circle being the warning area of the sea platform, the annular area being the expected monitoring area, the inner circle radius being r1, the outer circle radius being r2; the number of nodes on the inner circle is N1, the number of nodes on the outer circle is N2, and the nodes on the inner circle are uniformly arranged along the inner circle; the angle between the first node on the outer circle and the x-axis is The coordinates of the k1th node on the inner circle can be expressed as

[0014]

[0015] The coordinates of the k2th node on the outer circle can be expressed as

[0016]

[0017] The first step, the elliptical coverage principle refers to: in the FSR sensor network, the basic coverage model is the forward scattering area between the transmitter and receiver, which can be approximated as an ellipse, the coverage model takes the baseline length L as the long axis of the ellipse, and the selection of the short axis of the ellipse depends on the length and shortness of the baseline (generally Ltan(20°)). The annular region between the inner circle and the outer circle is the monitoring area, and the coverage rate calculation formula is defined as:

[0018]

[0019] In the formula, GapArea is the coverage gap area;

[0020] In the first step, each node in the sensor network has the ability of self-positioning, signal transmission / reception and real-time information transmission, specifically including: the outer circle node can transmit and receive signals, the inner circle node only receives signals, adjacent outer circle nodes form a transmission-reception combination in pairs, and any outer circle node and the two or three inner circle nodes closest to it form a transmission-reception combination. The transmission node transmits radar signals of different frequencies, and the receiving node uses an omnidirectional antenna to receive signals;

[0021] In the second step, the method for pre-processing the echo signal includes: the signal echo is obtained at each FSR radar node, since the sensor network is composed of multiple transmission nodes and multiple receiving nodes, each transmission node transmits signals of different frequencies, and the receiving nodes are not one-to-one matched with the transmission nodes, the receiving nodes do not fixedly receive the signal echo from a certain transmitter, and can receive signals of multiple or even all transmission nodes at the same time. Assuming that there are N receiving nodes, N signal transmissions are transmitted to the information fusion center in a single cycle, and assuming that the signal received by a receiving node is mixed by the signal echoes of M transmission nodes, then M filters are needed to extract the observation data of M corresponding transmission nodes from the echo signal;

[0022] In the third step, the method for matched filtering using the linear frequency modulation characteristic of the signal specifically includes: due to the influence of the scattering coefficient phase, the signals outside the main lobe have phase differences relative to the ideal linear frequency modulation signal, therefore, only the part inside the main lobe with small scattering coefficient influence can be used as the signal for correlation processing. The signal is a real linear frequency modulation signal with unknown parameters and unknown envelope. Therefore, the estimation of the Doppler frequency is obtained first, and then the optimal matched output of the signal is obtained. When the noise is Gaussian white noise, it is equivalent to the optimal maximum likelihood estimation, and the specific implementation can be represented as:

[0023]

[0024] Where θ is the parameter to be estimated, T s is the duration of the signal, S r(t, θ) is a reference (matched) signal constructed according to the signal model. In the mathematical sense, the matched filtering is that when the reference signal is consistent with the signal parameters to be estimated, it is in phase at each time, and the optimal accumulation effect can be obtained. According to the characteristics of the forward scattering echo, a complex matched signal is constructed for matched filtering, that is:

[0025]

[0026] In the formula, w(t) is a signal envelope, γ is a Doppler frequency, and is related to the motion speed of the target:

[0027]

[0028] Where v is the target speed, is the angle between the target motion direction and the baseline, and r f is the Fresnel radius of the target. Different γ parameters are traversed, the FSR signal is matched filtered, the estimated γ makes the output signal energy maximum, and the signal optimal matching output is

[0029]

[0030] In the third step, the target is detected by the mean value CFAR detector independently by each radar station, specifically including: the target detection method used by the sensor network does not consider the influence of non-uniform clutter, but detects the target in the clutter and noise obeying Gaussian distribution. Considering using the signal optimal matching output as a detection statistic for detection, and then performing CFAR detection.

[0031] In the FSR signal processing, only a single signal is processed, so for a reference unit of 2n points, n is the number of single-sided reference units, and the detection statistic is represented as:

[0032]

[0033] In the formula, w i is the value corresponding to the ith point detection statistic E op (t), so the relationship between the corresponding normalization factor and the false alarm rate is:

[0034] T = (P fa )-1 / n-1

[0035] The relationship between the detection rate P d of the CA-CFAR detection and the false alarm rate Pf a is:

[0036]

[0037] Where SNR represents the signal-to-clutter-and-noise ratio of the accumulated echo.

[0038] Compared with the prior art, the application has the advantages and beneficial effects that:

[0039] The existing distributed FSR networking mode is only designed for networking of coastal and air defense, forming a long chain shape or a dense arrangement shape with a parallelogram as a basic topological unit. Among them, the long chain shape distributed FSR can only detect the intrusion target in one direction, and is suitable for coastal observation; the dense arrangement shape distributed FSR with a parallelogram as a basic topological unit needs a large number of nodes to cover a large area of sea area, which is not conducive to the layout of the open sea area. Therefore, these configurations cannot realize the surrounding protection of isolated sea surface areas. The application designs a ring-shaped distributed FSR system, which uses fewer network nodes, is simple to lay out, and can detect any direction of intrusion of foreign targets. The FSR node is small, low in power consumption and low in cost, and the radar cross section increases rapidly in the forward scattering area, so the system reacts quickly to fast small targets on the sea surface (such as suicide attackers, drug and arms smuggling, speedboats and motorboats, etc. dangerous targets), and can detect targets under all-weather conditions, thereby realizing the monitoring and early warning of threat targets. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 A schematic diagram of a distributed FSR sensor network node is designed;

[0041] Figure 2 A sea surface small inflatable boat target side profile in an embodiment is designed;

[0042] Figure 3 A coverage model and target crossing direction schematic diagram in an embodiment is designed;

[0043] Figure 4 A system signal-to-noise ratio schematic diagram in an example is designed;

[0044] Figure 5 A detection statistic and threshold in an embodiment is designed;

[0045] Figure 6 A CFAR detection result in an embodiment is designed. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical scheme and advantages of the application more clear, the application will be further described in detail below by specific embodiments combined with the drawings. It should be understood that the specific embodiments described here are only used to explain the application, and are not used to limit the application.

[0047] In one embodiment, a ring-shaped distributed FSR sensor network sea surface early warning method is provided, comprising the following steps:

[0048] Step one involves deploying small floating forward scattering sensor nodes around key maritime locations (such as important bases, platforms, tourist attractions, and military targets) to form a distributed forward scattering sensor network. The protected area is then covered using the elliptical coverage principle. Each node in the network has the ability to self-locate, transmit / receive signals, and transmit information in real time, collaboratively sensing, collecting, and processing information about the objects sensed within the network's coverage area.

[0049] Specifically, such as Figure 1 As shown, with the important sea surface platform as the center, radar nodes are distributed on two concentric circles, one inner and one outer. A two-dimensional coordinate system is constructed in the horizontal plane, with the center of the concentric circles as the origin, and the x-axis pointing to the first node of the inner circle. The inner circle area is the warning zone for the sea surface platform, and the annular area is the desired monitoring zone. The radius of the inner circle is r1, and the radius of the outer circle is r2; the number of nodes on the inner circle is N1, and the number of nodes on the outer circle is N2. The nodes on the inner circle are evenly arranged along the inner circle; the angle between the first node on the outer circle and the x-axis is... The coordinates of the k1th node in the inner circle can be represented as:

[0050]

[0051] The coordinates of the k2th node in the outer ring can be represented as:

[0052]

[0053] Specifically, the elliptical coverage principle includes the following: In an FSR sensor network, the basic coverage model is the forward scattering region between the transmitter and receiver, which can be approximated as an ellipse. This coverage model uses the transmit / receive baseline length L as the major axis of the ellipse, and the selection of the minor axis depends on the length of the baseline (generally Ltan(20°)). The annular region between the inner and outer rings is the monitoring area, and the coverage calculation formula is defined as follows:

[0054]

[0055] In the formula, GapArea is the area of ​​the covered gap region.

[0056] Each node in the network possesses self-localization, signal transmission / reception, and real-time information transmission capabilities. Specifically, outer-ring nodes can both transmit and receive signals, while inner-ring nodes only receive signals. Adjacent outer-ring nodes form transmit-receive pairs, and any outer-ring node forms a transmit-receive pair with its two or three nearest inner-ring nodes. Transmitting nodes transmit radar signals at different frequencies, while receiving nodes use omnidirectional antennas to receive signals.

[0057] Step two, signal preprocessing, if any target crosses the coverage area, the electromagnetic wave signal of the transmitting node is scattered by the target to the receiving node, the system pre-processes the received echo signal and extracts the effective observation data.

[0058] Specifically, the signal echo is obtained at each FSR radar node, the system is composed of multiple transmitting nodes and multiple receiving nodes, each transmitting node transmits signals of different frequencies, and the receiving nodes are not one-to-one matched with the transmitting nodes, and the receiving nodes do not fixedly receive the signal echo from a certain transmitter, and can receive signals of multiple or even all transmitting nodes at the same time. Assuming that there are N receiving nodes, N signal transmissions are transmitted to the information fusion center in a single period, assuming that the signal received by a receiving node is composed of signal echoes of M transmitting nodes, then M filters are needed to extract the observation data of the M corresponding transmitting nodes from the echo signal.

[0059] Step three, target detection, using the linear frequency modulation characteristics of the above observation data for matched filtering, through long-time correlation processing to obtain a higher signal-to-noise ratio, and through a mean value type CFAR detector, each radar station independently realizes the detection of the target.

[0060] Specifically, due to the influence of the scattering coefficient phase, the signals outside the main lobe have phase differences relative to the ideal linear frequency modulation signal, so only the part inside the main lobe with less scattering coefficient influence can be used as the signal for correlation processing. The signal is a real linear frequency modulation signal with unknown parameters and unknown envelope, so the estimation of the Doppler frequency is obtained first, and then the optimal matching output of the signal is obtained. When the noise is Gaussian white noise, it is equivalent to the optimal maximum likelihood estimation, and the specific implementation can be represented as:

[0061]

[0062] Where θ is the parameter to be estimated, T s is the duration of the signal, S r (t, θ) is the reference (matching) signal constructed according to the signal model. In the mathematical sense, matched filtering is that when the reference signal is consistent with the signal parameter to be estimated, it is in phase at each time, and the optimal accumulation effect can be obtained. According to the characteristics of the forward scattering echo, a complex matching signal is constructed for matched filtering, that is:

[0063]

[0064] In the formula, w(t) is the signal envelope, γ is the Doppler frequency, and is related to the movement speed of the target:

[0065]

[0066] Where v is the target speed, is the included angle between the target moving direction and the baseline, r f is the Fresnel radius of the target. The FSR signal is matched filtered with different γ parameters, and the estimated γ makes the output signal energy maximum, and the signal is optimally matched output as

[0067]

[0068] The target is detected by the mean value type CFAR detector independently by each radar station, and specifically includes that the target detection method used by the system does not consider the influence of the non-uniform clutter, but detects the target in the clutter and noise obeying Gaussian distribution. The signal optimally matched output is used as a detection statistic for detection, and then CFAR detection is performed.

[0069] In the FSR signal processing, only a single signal is present, so for a 2n-point reference unit, the detection statistic is represented as:

[0070]

[0071] Therefore, the relationship between the corresponding normalization factor and the false alarm rate is:

[0072] T = (P fa )-1 / n-1

[0073] The detection rate P d of the CA-CFAR detection and the false alarm rate P fa are related as:

[0074]

[0075] Wherein SNR represents the signal-to-clutter-and-noise ratio of the accumulated echo.

[0076] In this embodiment, the system simulation parameters are set as follows: the frequency band is 1251 MHz, the total number of nodes is 10, the number of nodes in the inner and outer rings is 5, the inner ring radius is 10 km, the outer ring radius is 20 km, the included angle is π / 5, the transmission power is 40 W, the transmission antenna gain is 11 dB, the receiving antenna gain is 23 dB, the loss is 10 dB, and the antenna height is 0.9 m. The target parameters are set as follows: the target is a small inflatable boat on the sea, the side profile is as shown in Figure 2 , the target width is 4 m, the height is 1.6 m, the moving speed is 15 m / s, and the target moving direction is perpendicular to the baseline between the inner ring node 1 and the outer ring node 1. The simulation result is the analysis and processing result of the signal echo received by the inner ring node 1. Figure 3 The system coverage model is given, and in this example, the shadow area is the void area, so the coverage rate of this example can be calculated as 93.06%. Figure 4The system signal-to-noise ratio schematic diagram is given, the minimum detection signal-to-noise ratio is set to 10dB, and it can be seen that the forward scattering area meets the system detection requirement. Figure 5 and Figure 6 The detection statistics and threshold, and the CFAR detection result are given respectively.

[0077] Therefore, the example can cover more than 90% of the sea surface protection area, and realize effective detection of the fast small target on the sea surface.

[0078] The above content is a further detailed description of the present application in combination with specific embodiments, and cannot be considered as limiting the specific implementation of the present application to these descriptions. For ordinary skilled persons in the technical field to which the present application belongs, some simple deductions or replacements can be made without departing from the concept of the present application, and all of them shall be considered as falling within the protection scope of the present application.

Claims

1. A method for early warning of sea surface by a ring-shaped distributed FSR sensor network, characterized in that It comprises the following steps: First, the forward scattering sensor nodes are arranged in a circular manner on the sea to form a distributed forward scattering sensor network; Second, the received echo signals are preprocessed to obtain effective observation data using the sensor network formed in the first step; Third, the effective observation data obtained in the second step are subjected to matched filtering according to the linear frequency modulation characteristics, and the results after the matched filtering are subjected to mean value CFAR detector to realize the detection of targets and complete the sea surface early warning of the circular distributed FSR sensor network; The first step is to construct a two-dimensional coordinate system in the horizontal plane, with the concentric circle center as the coordinate origin, and the radar nodes are distributed on the inner and outer concentric circles. x The first node points to the inner circle, the concentric inner circle area is the warning area of the sea surface platform, the annular area is the expected monitoring area, the inner circle radius is , the outer circle radius is ; the number of nodes on the inner circle is , and the number of nodes on the outer circle is ; the nodes on the inner circle are evenly arranged along the inner circle; the angle between the line connecting the first node on the outer circle and the origin and the x axis is ; then the coordinates of the th node on the inner circle are represented as: Outer ring The coordinates of each node are represented as follows: 。 2. The sea surface early warning method of the circular distributed FSR sensor network according to claim 1, characterized in that: In the first step, the small floating forward scattering sensor nodes are arranged in a circular manner on the sea around the important place according to the elliptical coverage principle.

3. The sea surface early warning method of the circular distributed FSR sensor network according to claim 2, characterized in that: The important place on the sea includes important bases, platforms, scenic spots or military targets.

4. The sea surface early warning method of the circular distributed FSR sensor network according to claim 2 or 3, characterized in that: In the first step, each node in the formed sensor network has the ability of self-positioning, transmitting / receiving signals and real-time information transmission.

5. The sea surface early warning method of the circular distributed FSR sensor network according to claim 1, characterized in that: The first step is that the elliptical coverage principle refers to: in the FSR sensor network, the basic coverage model is the forward scattering area between the transmitter and the receiver, which is approximately an ellipse, and the coverage model takes the baseline length of the transmitter and the receiver as the long axis of the ellipse L The short axis of the ellipse depends on the length and shortness of the baseline, and the annular area between the inner circle and the outer circle is the monitoring area. The coverage rate calculation formula is defined as: In the formula, is the area of the gap region covered.

6. The sea surface early warning method of the circular distributed FSR sensor network according to claim 1, characterized in that: In the first step, each node in the sensor network has the ability of self-positioning, transmitting / receiving signals and real-time information transmission, specifically including that the outer ring nodes can transmit and receive signals, the inner ring nodes only receive signals, the adjacent outer ring nodes form a transmitting-receiving combination in pairs, any outer ring node and the two or three inner ring nodes closest to it form a transmitting-receiving combination, the transmitting nodes transmit radar signals of different frequencies, and the receiving nodes receive signals using omnidirectional antennas.

7. The sea surface early warning method of the circular distributed FSR sensor network according to claim 6, characterized in that: In the second step, if any target passes through the coverage area, the echo signal is the superposition signal of the electromagnetic wave signal scattered by the target to the receiving node and the direct wave signal; if no target passes through the coverage area, the echo signal is the direct wave signal. The method for preprocessing echo signal: the signal echo is obtained at each FSR radar node, the sensor network is composed of multiple transmitting nodes and multiple receiving nodes, each transmitting node transmits signals of different frequencies, and the receiving nodes and the transmitting nodes are not one-to-one matched, the receiving nodes are not fixed to receive signals from a certain transmitter only, and can receive signals of multiple or even all transmitting nodes simultaneously, assuming that there are N receiving nodes in total, N signal transmissions are transmitted to the information fusion center in a single period, and assuming that the signal received by a certain receiving node is composed of signal echoes of M transmitting nodes, then M filters are needed to extract observation data of M corresponding transmitting nodes from the echo signal.

Images