Radar unit for determining the position and / or speed of an object in road traffic
A radar unit with a sparse array of irregularly arranged antenna elements and non-regular Fourier transformation addresses the challenge of reducing antenna elements in radar systems, improving object detection for autonomous driving safety.
Patent Information
- Application Number
- DE202024105530
- Authority / Receiving Office
- DE · DE
- Patent Type
- Utility models
- Current Assignee / Owner
- Priority Date
- 2023-07-28
- Filing Date
- 2024-05-15
- Publication Date
- 2024-12-05
- Estimated Expiration
- 2034-05-31
AI Technical Summary
Radar systems for autonomous driving require a large number of antenna elements to achieve high angular resolution, leading to challenges in reducing the physical aperture and increasing the feasibility of autonomous driving systems.
A radar unit with a sparse array of antenna elements arranged in an irregular pattern, combined with a non-regular Fourier transformation, to determine the position and speed of objects in road traffic, reducing the number of antenna elements while maintaining detection accuracy.
The solution allows for reliable object detection with fewer antenna elements, enhancing driving safety by providing accurate position and speed information for advanced driver assistance systems.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
[0001] The present invention relates to a radar unit, in particular a vehicle radar unit, for determining a position, in particular direction and / or distance, and / or a speed of an object in road traffic, comprising an antenna unit with a plurality of antenna elements, wherein the antenna elements are arranged in an irregular pattern in which output signals can be emitted with the aid of the irregularly arranged antenna elements, in which reflection signals can be detected with the aid of the irregularly arranged antenna elements, and comprising a computing unit which is designed and / or configured to evaluate the detected reflection signals in order to determine the position and / or speed of the object by means of a non-regular Fourier transformation which can evaluate the signals from the antenna elements arranged in the spatially irregular pattern.
[0002] Radar systems are known to have multiple antenna elements for transmitting and receiving signals. However, the number of antenna elements is quite high.
[0003] Modern advanced driver-assistance systems (ADAS) are designed to improve driver safety and enable autonomous driving. Ultimately, such systems aim to enable autonomous driving. Key sensor technologies include ultrasound, lidar, cameras, and radar. Lidar and cameras provide high-resolution images but struggle in adverse weather conditions, while radar, while more robust, has lower image quality due to its low angular resolution. To increase the feasibility of radar systems for autonomous driving, large physical apertures are required, which presents a challenge when increasing the aperture of radar arrays.
[0004] A typical frequency-modulated continuous wave (FMCW) radar achieves high angular resolution with a large aperture. Antenna elements of the radar system or radar unit typically form uniform linear arrays (ULAs) separated by half a wavelength, but this can result in arrays with hundreds of antenna elements.
[0005] Reducing the number of elements without changing the aperture results in sparse arrays. The design of such arrays has gained attention because the antenna elements are strategically positioned to avoid ambiguous detection.
[0006] The object of the invention is to reduce the number of antenna elements in a radar system, while still being able to reliably detect an object in road traffic.
[0007] The problem is solved by a radar unit, a vehicle, a computer program and a computer-readable storage medium with the features of the independent claims.
[0008] A radar unit, in particular a vehicle radar unit, is proposed for determining a position and / or a speed of an object in road traffic by means of a radar unit in a vehicle. The radar unit can also be a vehicle radar unit. The position of the object can, for example, be a direction and / or distance. Direction and / or distance can be related to the radar unit or the vehicle. The direction means, for example, that it can be determined whether the object is in front of the vehicle or to the left or right of the vehicle. The distance to the radar unit and thus to the vehicle is also measured or determined. By determining the position and / or a speed of an object in road traffic, driving safety is increased and accidents are avoided. The method described here can therefore be used in driver assistance systems.The objects can also be other road users, such as other cars, trucks, bicycles, pedestrians, or solid objects such as trees, poles, or pillars. The radar unit can be located, for example, in a bumper, particularly in the front and / or rear area of the vehicle, and / or another vehicle component. These driver assistance systems are also referred to as "Advanced Driver Assistance Systems," or ADAS for short.
[0009] The radar unit comprises an antenna unit with multiple antenna elements, wherein the antenna elements are arranged in an irregular pattern. Furthermore, the antenna elements are sparsely arranged. This means that there are fewer antenna elements compared to antenna units whose antenna elements are arranged in a regular pattern. Such antenna units are referred to as "sparse (antenna) arrays." If the antenna elements are arranged in a regular pattern, for example, they are all spaced apart from one another at a distance corresponding to half the wavelength of the output signals.
[0010] Furthermore, output signals are emitted using the irregularly and / or sparsely arranged antenna elements. The output signals are electromagnetic waves.
[0011] In addition, reflection signals are detected using the irregularly and / or sparsely arranged antenna elements. After the output signals are emitted, they encounter objects, if any are present. The electromagnetic waves are reflected by the object, sending reflection signals back to the antenna unit. These reflection signals are detected by the antenna elements, which are arranged in an irregular and / or sparse pattern.
[0012] Furthermore, the radar unit comprises a computing unit that is designed and / or configured to evaluate the detected reflection signals to determine the position and / or speed of the object using a non-regular Fourier transform. The non-regular Fourier transform can evaluate the reflection signals detected by the antenna elements arranged in the spatially irregular pattern. The non-regular Fourier transform is also referred to as a "non-uniform Fourier transform." With the help of the non-regular Fourier transform or a non-regular fast and / or discrete Fourier transform, a tool is provided that can evaluate the reflection signals at a sufficient speed, wherein the reflection signals were detected by the antenna elements in the irregular and / or sparse pattern.The non-uniform fast and / or discrete Fourier transform is also known as nuFFT (non-uniform fast Fourier transformation). The position and / or velocity of the object is determined using a corresponding algorithm.
[0013] An improvement results when an antenna unit is used in which the antenna elements are arranged on grid points of a, in particular two-dimensional, grid, with some grid points being unoccupied.
[0014] It is useful if grid points arranged next to each other have a distance of λ / 2.
[0015] It is advantageous if the antenna elements are arranged in a circle around the grid points, with the circle having a radius of 5% - 20% of λ / 2.
[0016] It is advantageous if the antenna elements are arranged such that the array or arrangement of antenna elements provides a clear response and preferably coincides with or approximates the grid points, with some elements remaining unoccupied. The grid spacing is preferably λ / 2.
[0017] It is advantageous if the antenna elements are arranged in such a way that they deliver clear results, especially after the non-uniform fast Fourier transformation. This should avoid generating any additional signal peaks other than those from objects.
[0018] It is beneficial to use an antenna unit with a sparsely populated grid. An antenna unit with a sparse distribution of antenna elements can also be used. The antenna elements are best suited to a sparsely populated grid.
[0019] It is advantageous if the positions of the antenna elements are stored in a data storage and the positions of the antenna elements are retrieved from the data storage to perform the non-regular Fourier transform. The positions of the antenna elements are relative to a reference point, for example, the center of the antenna unit. The positions of the antenna elements can be used as parameters for the non-regular fast Fourier transform or the corresponding algorithm. The positions of the antenna elements allow the non-regularity of the antenna elements to be taken into account when determining the position and / or distance of objects.
[0020] It is beneficial to use the non-regular Fourier transform and the antenna elements arranged in an irregular pattern to determine the direction, especially an angle, of the object to the antenna unit.
[0021] It is advantageous to use the non-regular Fourier transform and the antenna elements arranged in an irregular pattern to determine the distance of the object to the antenna unit.
[0022] It is beneficial if a non-regular fast Fourier transform algorithm is used to determine the position, especially the direction, and / or the distance of the object.
[0023] It is advantageous if the antenna elements are operated using a MIMO (“multiple input multiple output”) method.
[0024] It is useful to use an antenna unit with antenna elements arranged in such a way that, in a virtual arrangement of the antenna elements, all antenna elements are spaced apart from one another. The antenna elements of the array or antenna unit are arranged such that, in the virtual array setting in the MIMO (Multiple Input Multiple Output) configuration, the following two requirements are met: a) the elements fit into an evenly spaced grid or pattern (preferably at a distance of λ / 2) and b) two virtual antennas in the same location are avoided (this only provides redundant information to the system). This is achieved in the context of arrays with minimal redundancy. They simultaneously minimize blurring in the field of view and redundancies in the information acquired by the array.
[0025] Furthermore, it is advantageous if at a time a selected antenna element transmits the output signal and at least some other, in particular all, antenna elements detect the reflection signals.
[0026] An advantage results when a sequence is carried out in which at respective times a selected antenna element transmits the output signal and at least some other, in particular all, antenna elements detect the reflection signals, wherein at successive times other antenna elements transmit the output signals and at least some other, in particular all, antenna elements detect the reflection signals.
[0027] It is advantageous if the output signals have at least one output pulse, wherein the output pulse is frequency-modulated so that a frequency of the output pulse, in particular from an initial frequency, increases preferably linearly by a signal bandwidth (B).
[0028] According to an advantageous development of the invention, it is useful if the output signals are generated with a frequency between 76 and 81 GHz.
[0029] It is advantageous if the output signals have multiple output pulses. These output pulses are also referred to as chirps.
[0030] It is advantageous if several product values are formed by multiplying the respective output signals and the respective reflection signals.
[0031] According to an advantageous development of the invention, it is useful if the product values are filtered, in particular by means of a low-pass filter, so that filter values are formed.
[0032] It is advantageous if quadratic terms are neglected during the simplification and / or geometric relationships between the antenna elements are taken into account.
[0033] It is advantageous if the output signals are continuously transmitted and / or the reflected signals are continuously detected. This is the operating mode of a continuous wave radar or FMCW radar.
[0034] An improvement results when the approximate values are shifted to a common point in time. The MIMO operating mode, as well as the coherence and stationarity of the radar unit, enable the time shift to be implemented.
[0035] It is advantageous if the approximate values ẋ̂ mkn (t) are Fourier transformed with respect to time, preferably using a fast Fourier transform. X^mkn[fR]=F¯[xmkn](fR)=∫−∞∞xmkn(t)exp(−j2πfRt)dt=γkexp(−j2πλ(dxRm+dxTn)sin(θk))∫0Tcexp(−j2π(fRk−fR)t)dt X^mkn[fR]=γkexp(−j2πλ(dxRm+dxTn)sin(θk))exp(−j2π(fRk−fR)Tc)sin(π(fRk−fR)Tc)π(fRk−fR)
[0036] It is advantageous if a maximum of X̂mkn [f R ] is determined. Using the sinc function given above, the maximum is where f Rk = f R This allows the distance to be determined.
[0037] It is advantageous if the approximate values ẋ̂ mkn (t) are Fourier transformed with respect to position, preferably using a fast Fourier transform.
[0038] Furthermore, it is advantageous if the approximate values x̂ mkn (t) must first be Fourier transformed with respect to time and then Fourier transformed with respect to position.
[0039] It is beneficial if, after applying the Fourier transformation with respect to time, the non-regular discrete Fourier transformation with respect to space is applied, with X^^k[fR,fΘp] =
[0040] It is also advantageous if, after determining a position, in particular direction and / or distance, and / or speed of an object in road traffic, the corresponding result is output. The result of determining the position, direction, distance or separation and / or speed can, for example, be output in such a way that it is shown on a display in the vehicle. This provides the driver of the vehicle with feedback about the position and / or speed of the object and allows them to adjust their driving behavior accordingly. Additionally or alternatively, the result can advantageously also be output to a higher-level vehicle control system. This higher-level vehicle control system can then, for example, initiate braking to avoid an accident.
[0041] Furthermore, a computer program is proposed which, when executed by the computing unit, causes it to carry out a method according to one or more method steps of the preceding and / or following description.
[0042] Also proposed is a computer-readable storage medium comprising a computer program, wherein the computer program, when executed by the computing unit, causes the computing unit to carry out a method according to one or more method steps of the preceding and / or following description.
[0043] The radar unit or radar system can execute the method according to one or more method steps of the preceding and / or following description. In particular, the radar unit or radar system comprises the antenna unit with a plurality of antenna elements, wherein the antenna elements are arranged in an irregular and / or sparse pattern. The radar unit or radar system further comprises the computing unit, which can execute the method with at least one method step of the preceding and / or following description.
[0044] It is advantageous if the radar unit comprises a microcontroller. In this case, the microcontroller can be the computing unit. Alternatively, it is advantageous if the computing unit is integrated into the microcontroller. The calculations required for the method can be carried out quickly and efficiently by the microcontroller and / or the computing unit of the radar unit. A computer program can be installed and / or arranged on the microcontroller, which causes the microcontroller and / or the computing unit to carry out the method for determining a position, in particular direction and / or distance, and / or a speed of an object in road traffic and / or to carry out the calculations required for this purpose.Additionally or alternatively, the microcontroller can be designed in such a way, for example by means of a corresponding architecture, that the microcontroller can carry out the method for determining a position, in particular direction and / or distance, and / or a speed of an object in road traffic and / or the calculations required therefor.
[0045] The microcontroller and / or the computing unit can be designed as a so-called “system on a chip”.
[0046] The method steps explained in the preceding and / or following description can thus be carried out by means of the microcontroller and / or the computing unit. These method steps comprise, in particular and at least, controlling the components to generate the output signals with the aid of the antenna unit. The antenna unit can also advantageously be used to detect the reflection signals, and the microcontroller and / or the computing unit can advantageously carry out and / or control the evaluation and / or processing of the electrical signals based on the reflection signals. In addition, the microcontroller and / or the computing unit can output a result of the position, distance, and / or speed measurement, which is then displayed, for example, on a vehicle display as position, distance, and / or speed information, for example, relative to a car traveling ahead.
[0047] Additionally or alternatively, the microcontroller and / or the processing unit can also be connected to a higher-level control system of the vehicle. For example, the microcontroller and / or the processing unit can transmit the result of the position, distance, and / or speed measurement to the vehicle's higher-level control system. Based on this result, the higher-level control system can, for example, control a braking system and initiate a braking operation to avoid a rear-end collision.
[0048] The microcontroller and / or the computing unit are preferably designed and / or configured to monitor and / or control the radar unit components involved in the method, and advantageously also to evaluate and / or process the signals. Additionally or alternatively, the microcontroller and / or the computing unit can output the results of the position, distance, and / or speed measurements and / or transmit them to other components in the vehicle.
[0049] The microcontroller and / or the computing unit can advantageously be implemented as a semiconductor chip and / or a microprocessor. The microcontroller and / or the computing unit can also be integrated into an embedded chip.
[0050] It is also advantageous if the computing unit and / or the microcontroller are connected to the antenna unit. Furthermore, the computing unit and / or the microcontroller can also be connected to the antenna elements. This allows the computing unit and / or the microcontroller to transmit the output signals and receive the reflected signals for evaluation. The computing unit and / or the microcontroller can transmit the output signals directly to the antenna unit for transmission. Alternatively, the computing unit and / or the microcontroller can cause another element of the radar unit, for example, the signal generator explained below, to transmit the output signals.
[0051] It is also advantageous if the radar unit includes a data storage unit. The data storage unit can store the data required for the process.
[0052] Additionally or alternatively, it is advantageous if the radar unit has a signal generator. The signal generator can, for example, receive control signals from the processing unit and / or the microcontroller. The signal generator can then transmit corresponding signals to the antenna unit, which then transmits the output signals.
[0053] Additionally or alternatively, it is advantageous if the radar unit has a multiplier. Using the multiplier, signals can be combined and / or processed together to obtain processed signals for the method described here.
[0054] Additionally or alternatively, it is advantageous if the radar unit includes a low-pass filter. This filter can be used to filter signals.
[0055] Additionally or alternatively, it is advantageous if the radar unit includes a switching unit. The switching unit can form a connection between the signal generator and a respective antenna element, so that the signal from the signal generator is routed to a specific antenna element. The switching unit can switch back and forth between the antenna elements. The switching unit can be a demultiplexer, for example.
[0056] Furthermore, it is advantageous if the data memory, the signal generator, the multiplier, the low-pass filter, and / or the switching unit are integrated into the computing unit and / or the microcontroller. At least one of the elements mentioned here can be integrated into the computing unit and / or the microcontroller. Additionally or alternatively, at least one of the elements mentioned here can be arranged outside the computing unit and / or the microcontroller. For example, the data memory can be embodied as a separate component, such as a chip.
[0057] Additionally or alternatively, it is advantageous if the data memory, the signal generator, the multiplier, the low-pass filter, and / or the switching unit are connected to the computing unit and / or the microcontroller. For example, the data memory can be designed as a separate component from the computing unit and / or the microcontroller. Data connections and / or signal connections can be arranged between the elements mentioned here, so that the elements can communicate with each other and / or exchange data and / or signals.
[0058] Furthermore, it is advantageous if the processing unit, microcontroller, data memory, signal generator, multiplier, low-pass filter, and / or switching unit are arranged on a circuit board. This interconnects all elements. Additionally or alternatively, the antenna unit can also be coupled to the circuit board. It is also possible for the antenna unit to be connected to the circuit board, for example, via a connector located on the circuit board, using a data cable and / or signal cable.
[0059] A vehicle with a radar unit is also proposed. The radar unit can be configured according to a feature of the preceding and / or subsequent description. The radar unit can also be configured, in particular, to include the computing unit, in such a way that it can execute at least one method step of the preceding and / or subsequent description.
[0060] The automotive radar system or radar unit uses a so-called Linear Frequency Modulated Continuous Wave (LFMCW) radar in a Multiple Input Multiple Output (MIMO) configuration. Here, signal processing, which is generally performed using beamforming methods, is modeled as a discrete Fourier transform (DFT) expression. In the context of far-field approximation, both methods are essentially equivalent.
[0061] The DFT calculation for distance information conjugate to the time variable is a standard procedure in this field. However, applying this identical process (DFT) to the angular variable results in a nonlinearity in the conjugation between the angles and angular frequencies. Therefore, it is important to ensure accurate sampling of the angular frequency information. With a field of view (FOV) of + / - 90 degrees, a spacing of half a wavelength between the elements is required. This adherence to the Nyquist criterion guarantees accurate signal sampling and thus prevents the introduction of ambiguities during the sampling phase.
[0062] The angular resolution of the array is determined by the array aperture in conjunction with the wavelength. From a practical perspective, it is often advantageous to combine a large aperture with a sparsely distributed array. However, this leads to a violation of the Nyquist criterion, which affects the quality of the resulting processing. In most cases, the DFT can still be calculated because we are only working with phasors or limited terms, but target identification may become ambiguous or even erroneous. Here, the term "array" is used to refer to the arrangement of antenna elements, or rather, the antenna unit.
[0063] For sparse systems or antenna arrays, it may be useful to apply the formalism and techniques of compressed sensing. This approach incorporates a novel set of signal processing and recovery methods anchored in "greedy algorithms" such as orthogonal matching pursuit (OMP). Another approach to consider is ensuring well-controlled DFT convergence when treating minimum redundancy arrays (MRA). Although this approach results in a higher geometric noise background, strategic antenna positioning can minimize sidelobe levels across the entire FOV of interest. This can be achieved through comprehensive system tuning and careful array design methods.
[0064] With the introduction of nonlinearity in angular coordinate space and the sparse, irregular positioning of the antenna elements for sampling angular frequencies, the problem can be formulated as a non-uniform density functional theory (NuDFT) calculation. NuDFT can be computed more quickly using an FFT-based method—the NuFFT algorithm. The approach can be extended to 3D imaging by incorporating the elevation angle, whose formulation and processing is similar to the azimuth angle. The FOV is also adjusted according to requirements in this case.
[0065] The NuFFT algorithm may require an oversampling step followed by interpolation, where the signal is aligned to the grid coordinates. This phase is the most resource-intensive part of the algorithm and where signal accuracy is most likely to be compromised. To streamline this step, the array can be designed so that the elements of the sparse array in the virtual environment are directly aligned with the oversampling grid. This grid can be calculated during the array design phase. This alignment optimizes processing when the NuFFT grid aligns with the Uniform Linear Array (ULA) grid in the virtual environment. The result is that the NuFFT requires fewer additional points and oversampling is minimized. Depending on the specific constraints and design decisions, oversampling can be increased but limited (e.g.,to a lambda / 4 grid spacing, doubling the points used by the NuFFT algorithm). Despite this complexity, this approach still offers significant advantages in terms of algorithm complexity (order(V*logV)) and the speedup associated with FFT-based processing.
[0066] Finally, the use of a NuDFT offers flexibility when dealing with real-world radar systems, which may exhibit slight alignment errors in the positions of the antenna elements. During this phase of signal processing, calibration information can be readily applied to the signal. In such scenarios, it is important to carefully select the NuFFT interpolation function and method to match the radar signal shape, thus ensuring precise and efficient signal processing.
[0067] The array is designed to meet NuFFT processing requirements, enabling both an efficient configuration that avoids a large oversampling grid and signal degradation. This allows for the entire signal processing to be implemented under the same formalism with minimal formulation effort.
[0068] A random, sparse array is the worst case for applying the NuFFT algorithm. Even if the NuDFT is well-defined and the arrangement is unambiguous, processing efficiency cannot be significantly improved by such an arrangement. To apply the NuFFT algorithm, it is usually good to find a grid spacing that accommodates all or most of the antenna elements. Often, in a random arrangement, pairs of elements hardly share a common multiple. A sufficiently dense grid is then provided so that all signals can be interpolated to some of the grid points. In this step, most of the signal is distorted, and the signal accuracy is lost.
[0069] Among the numerous sensors in modern vehicles, automotive radar systems provide high-resolution estimates of target position and velocity. Furthermore, higher angular accuracy can be achieved by using larger radar apertures that efficiently combine multiple transmitters and receivers. Sparse antenna arrays, i.e., sparsely and / or irregularly arranged antenna elements, are designed to provide such high angular resolution while avoiding ambiguities caused by irregular spacing between the antenna elements. Such arrays require a high computational effort for real-time applications. Taking into account the non-uniform or irregular structure of the signal under consideration, we demonstrate that processing methods based on non-uniform fast Fourier transforms (NuFFT) can efficiently and accurately estimate the direction of arrival (DOA) of targets.
[0070] A MIMO (multiple input multiple output) configuration creates large virtual arrays with a smaller number of antenna elements.
[0071] Fast processing of FMCW radar signals typically involves a Fourier transform (FT) of the digitized mixed signal for each channel. For ULA arrays with a spacing of λ / 2, Fourier transforms can also calculate the direction of arrival (DOA) by integrating over the virtual array dimension. Multidimensional FFT processing strategies are implemented in commercial solutions, but are limited by sampling theorems that require equidistant sample points and are challenging for sparse arrays.
[0072] Newer methods of non-uniform fast Fourier transform (NuFFT) enable the processing of data on irregular grids. NuFFT enables the direct, efficient multidimensional processing of FMCW radar signals with sparse arrays. NuFFT is crucial for processing synthetic aperture radar (SAR) data and generates high-resolution radar images by efficiently processing unevenly sampled data and compensating for Doppler shift, resulting in focused images of targets.
[0073] We consider a signal that is transmitted at a time t by an antenna element n such that its waveform sn(t)=Acos(2π(f0t+12St2)) This waveform is subject to linear frequency modulation (LFM) in time. The frequency starts at a value f0 and increases linearly to f0 + B, where B is the signal bandwidth. This frequency modulation forms a pulse or chirp of duration T C with a gradient S=BTC.
[0074] When a target k is in front of the radar unit or radar system, part of the emitted radiation is scattered back to the system. A receiving antenna m receives an echo signal from the target k in the form S mk (t) = A k s n (t - τ). The received signal is the transmitted signal s n (t), which is delayed by τ and is subject to a damping determined by the amplitude coefficient A k described.
[0075] The transmitted signal s n (t) and the received echo s mk (t) from target k are mixed by direct multiplication so that xmkn (t) = s n (t) * s mk (t). Only the low frequency, the so-called beat frequency f IF , remains after the high frequency component from x mkn (t) was filtered out. The resulting filtered mixed signal x̃ mkn (t) = γ IF cos(2πf IF t + ϕ IF ) is again a periodic function. This filtered mixed signal can also be called the filter value. The beat frequency f IF is the difference between the transmit and echo frequencies. Since the echo delay time τ = 2cR is proportional to the distance R traveled by the wave, f IF = S * τ. For automotive applications, the delay is short and therefore the expression for the intermediate frequency phase simplifies to ϕ IF = 2πf0τ - πSτ 2 ≅ 2πf0τ. The amplitude term γ IFtakes into account the different losses of the transmitted signal intensity and the radar cross-section of the target k at this wavelength. In the following, the delay is referred to as τ mkn , and the complex amplitude as γ k since the predominant contribution is the target k. In practice, it is common to consider both quadratures of the mixed signal separately, so that x mkn (t) = γ k exp(j2πf0τ mkn )exp(j2πSτ mkn t) is.
[0076] Both terms of the signal depend mainly on the delay time τ mkn of propagation between the transmitters, the targets and the receivers. This affects both the intermediate frequency f IF as well as the term of the constant phase ϕ IF The time delay results from the total propagation distance with the speed c from the transmitting antenna n and the backscattering by a target k to the receiving antenna m, so that τmkn=Rmknc=Rmk+Rknc applies.
[0077] The positions of the transmitter n and the receiver m are at a distance dx Tn or dx Rm from a center of the array. The distance between the antennas m and n and a target k with respect to the center of the array is given by the cosine law and has the following expressions. Rkn≅Rk−dxTn sin(θk) Rmk≅Rk−dxRm sin(θk)
[0078] The summation of both terms gives the total approximate distance between the antennas and the target k, R mkn = R kn + R mk = 2R k - (dx n + dx m ) sin θ k . Simplifying with dx RTmn = dx Rm + dx Tn , one obtains the expression for the time delay τmkn=Rmknc=2Rkc−dxRTmnsinθkc.
[0079] The received signal x mkn (t) has the following approximate expressions: x^mkn(t)=γkexp(−j2πλ(dxRmsin(θk)))exp(−j2πλ(dxTnsin(θk)))exp(j4πSRkct) with λ=cf0, the wavelength of the radiation. Additional constant phase contributions independent of θ can be captured in the complex amplitude γek. Starting from (3), which describes the signal expression of a single target k for a given pair n and m of transmitting and receiving antennas, respectively, we can formulate the received signals more succinctly in the case of multiple antennas and targets. The present system uses time domain multiplexing (TDM), in which only one antenna transmits at a given time and all receiving antennas listen simultaneously. This approach enables the use of the multiple input multiple output (MIMO) formalism.
[0080] In the TDM method, the signals are not transmitted simultaneously. However, due to the stationarity of the generated signals, they can be converted to a transmission time for processing.
[0081] The antenna position coordinates of the simulated sparse array are multiples of λ / 2. According to the MIMO approach, the total signal x mn (t), which is transmitted by an antenna n and received by an antenna m, is collected and converted into a vector x(t). The signal received by an antenna m is the coherent sum of the echo contributions of all K targets in front of the antenna, such that xmn(t)=∑k=1Kxmkn(t), At a given time t, the received signal is the vector x(t) = [x 11 (t), x 21 (t), ..., x (M-1)N (t), x MN (t)] T ∈ C MNx1, where the superscript T is the transposition operator. The entire received signal includes the forward and return signal between all combinations of transmitters and receivers and is transferred into the signal vector: x(t)=∑k=1KγkA(θk)sk(t)+w(t)
[0082] The sampling vector A(θ k ) describes the response of the array with respect to the azimuth position of the targets. A complex additive white Gaussian noise vector such that w(t) ∈ C MNx1 , is added to the signal model. The temporal waveform s(t) is the FMCW composite echo signal. For a given target k of interest, we can determine the direction of departure (DoD) a T (θ k ) and the direction of arrival (DOA) a R (θ k ) as follows: aR(θk)=[e−j2πλdxR1sin(θk),...,e−j2πλdxRMsin(θk)]T aT(θk)=[e−j2πλdxT1sin(θk),...,e−j2πλdxTNsin(θk)]T
[0083] They add up to the sampling vector A(θ k) = a R (θ k ) a T (θ k ) T ∈ C MNx1 , which encompasses all combinations of transmit and receive antennas and covers the entire virtual antenna area. This vector encodes the topography of the array. The complete sampling vector is then as follows: A(θk)=[e−j2πλdxRT11sin(θk), e−j2πλdxRT12sin(θk),...,e−j2πλdxRTmnsin(θk),...,e−j2πλdxRTMNsin(θk)]
[0084] In the following, a received signal vector x(t) is considered, as described in equation (4). Since the intermediate frequency signal consists of a sum of sine terms, it is convenient to first perform the Fourier transform on the time variable. Numerically, this amounts to performing a fast Fourier transform (FFT) on the temporal data of each virtual array element. The azimuth spectrum with parameter θ is then estimated for each range bin. This is typically done by digital beamforming (DBF). The expression is P(θ)=A∗(θ)RxA(θ) where R x is the covariance matrix of the signal x(t). Peaks from the resulting scan are searched for as potential DOA (direction of arrival) targets.
[0085] In the case of a ULA (uniform linear array), the DOA calculation can be performed directly by applying the FFT over the dimension of the virtual antenna positions. It is shown that for a sparse array, an equivalent strategy can be applied for FMCW radars by using the NuFFT (non-uniform fast Fourier transform) instead of the regular FFT. Next, the identification of the sources of the received signal power (the targets) by calculating the 2D range-azimuth spectral density is discussed. Considering a 1D time signal y(t) with its Fourier transform Ŷ(f), the power spectral density (PSD) is the squared modulus of the Fourier transform, S yy (f) = |Ŷ(f)| 2 , where t and f are conjugate Fourier variables of time and frequency, respectively.
[0086] The PSD over the time variable is initially calculated with a single component x mkn(t) of x(t). We also assume that we have a target k and that the noise contributions are negligible. Starting from (3), the frequency fRk=2SRkc defined so that, x^mkn(t)=γkexp(−j2πλ(dxRm+dxTn)sin(θk))exp(j2πfRkt)
[0087] The FT of ̂ẋ̂ mkn (t) with the parameters f R and t as conjugate Fourier variables is: X^mkn[fR]=F¯[xmkn](fR)=∫−∞∞x^mkn(t)(t)exp(−2πfRt)dt==γk exp(−j2πλ(dxRm+dxTn)sin(θk))∫0Tcexp(−j2π(fRk−fR)t)dt
[0088] We restrict the signal to the duration of a pulse or chirp (from t0 = 0 to Tc), i.e., the period during which the signal is measured. Calculating the integral part of (11) yields: X^mkn[fR]=γkexp(−j2πλ(dxRm+dxTn)sin(θk))exp(−j2π(fRk−fR)Tc)sin(π(fRk−fR)TC)π(fRk−fR)
[0089] The range frequency f Ris directly proportional to the range parameter R with fR=2SRc. If you find a frequency f R = f Rk , where the spectral density of X̂ mkn [f R ] is maximum, the range R k of a target k.
[0090] The processing is then extended to the azimuth direction θ. The extreme positions of the antennas in the virtual array define the aperture A = max(dx RTmn ) - min(dx RTmn ). The origin of the array is at its center and Δ = dx RTmn and fθk=sin θkλ are defined. For now, Δ ∈ [-A / s, A / 2] is a continuous variable. X^k[fR]=γk exp(−j2πλΔfθk)exp(−jπ(fRk−fR)Tc)sin(π(fRk−fR)Tc)π(fRk−fR)
[0091] The adjoint Fourier transform is applied to equation (13) with Δ and f θ applied as conjugate parameters. X^^[fR,fΘ] = ∫−∞∞ ∫−A / 2A / 2exp(−2πΔfθk)exp(j2πΔfθ)dΔ
[0092] Calculating the integral as before and taking all K targets into account results in: X^^[fR,fΘ]=∑k=1Kγkexp(−j2π(fRk−fR)Tc)sin(π(fΘ−fΘk)A)π(fΘ−fΘk)sin(π(fR−fRk)Tc)π(fR−fRk)
[0093] In practice, the signal points of (16) are recorded for specific, discrete values. In particular, the signal is sampled at regular time intervals Δt by a digitizer, so that a specific recorded time is given by t = t0 +ℓ Δt, where ℓ is the sample number and t0 is the start time. This allows the calculation of the previous Fourier integrals by discrete Fourier transforms (DFT). The knowledge of f R = f Rk directly indicates the range of the detected target with Rk=fRkc2S According to the theory of regular sampling, the radar system has a maximum detection range of c4SΔt≃250 m m and a distance resolution of c2S≃15 cm. cm.
[0094] Similar to range, the DOA of a target is determined by determining f θ = f θkdetermined, and it becomes possible to determine the angular parameter associated with the various targets. The radiation across aperture A is sampled by the various virtual antennas. Large arrays are required to obtain high-resolution DOA estimates. It is attractive to sample the signal with strategically positioned antennas. DOA calculation requires that the estimate of a target at a given angle be unambiguous across the entire field of view. This can be optimized by minimizing the sidelobe levels (SLL) of a sparse array with irregular antenna positioning.
[0095] An array with minimal redundancy is created. It has 59 degrees of freedom and a redundancy value of 1.08. Calculating the array factor shows that the detection of a target in front of the radar is unambiguous, even though the SLL is higher than for a ULA with the same aperture size.
[0096] The DOA of targets is determined directly by beamforming using Equation (8). This DOA estimation method has a quadratic complexity with the number of evaluated azimuth angles. In analogy to the ULA case, it is also possible to calculate the azimuth spectral density from Equation (14) using non-uniform discrete Fourier transform methods. For a given sparse array with antenna positions dx RTmn NuFFT efficiently calculates p Fourier coefficients f θp , which are within [−1λ,1λ] distributed and provided by the following non-uniform DFT X^^k[fR,fΘp] =
[0097] Δ mn can be assigned to the positions of the antenna elements dx RTmnThe NuFFT algorithm utilizes the fast Fourier transformation (FFT) in its core computation by using an oversampling factor and an interpolation step. The oversampling factor ensures that the unevenly distributed samples are correctly represented in the frequency domain, while the interpolation step bridges the gap between the unevenly distributed points and the oversampled grid required for applying a regular FFT. Consequently, the computational complexity of the NuFFT is of order (V*logV), where V represents the number of Fourier coefficients after oversampling. This complexity is comparable to the standard FFT and makes the NuFFT a computationally efficient method.
[0098] To illustrate the previous discussion, signal data is generated using equation (4) with the exact time delays τ, while processing is performed using the approximate expressions of equation (2). The range FFT is first applied to the time traces, followed by DBF or azimuth NuFFT to the range domains across the virtual array channels.
[0099] The NuFFT is calculated numerically using a dedicated library. For the sparse array, the results of non-uniform Fourier processing are similar to those obtained with DBF. Individual targets are well resolved, as are the two extended obstacles. This illustrates the applicability of non-uniform Fourier transform-based processing for the angular resolution of sparse FMCW radar systems. Velocity estimation could also benefit from the use of non-uniform sampling, since sampling is not continuous in TDM-MIMO. We anticipate that the direct use of processing methods that account for irregularities in the sampling parameters, whether spatially or temporally sparse, will be extended to additional dimensions to efficiently capture range, velocity (Doppler), azimuth, and elevation by performing a 4D NuFFT.
[0100] Example values are f0 = 70 to 80 GHz, B = 1 to 4 GHz, T c = 5 to 75 µs and / or Δt = 2 to 20 ns. Advantageous values are f0 = 77 GHz, B = 1 GHz, T c = 33 µs and / or Δt= 10 ns.
[0101] With the advent of larger arrays consisting of sparsely distributed antennas, radar systems aim to simultaneously provide high-resolution range, velocity, and angle estimates for potential targets in front of a vehicle. This results in the need to process large amounts of data to achieve real-time target acquisition. For this purpose, efficient processing methods are needed. This work demonstrates that known azimuth FFT-based processing methods can be extended to a sparsely and unevenly populated field for an FMCW radar system. For this purpose, the angle FFT is replaced by an angle NuFFT, which accounts for the scatter and irregularity of the array and the signal structure. Both the range and azimuth processing are then computed with the same complexity as the regular FFT.In the automotive sector, the presented approach could be extended to also improve the processing of the Doppler effect in MIMO radars.
[0102] Based on the MIMO signal model in matrix form, x(t)=∑k=1KγkA(θk)sk(t)+w(t), the covariance matrix R x of the signal vector can be easily calculated. It is later used for a variety of operations. This approach reduces the amount of data by removing the time dependence.
[0103] The covariance matrix is: Rx=E[x(t)xH(t)] Rx=AE[s(t)sH(t)]AH+σ2I0 with σ 2 the noise power and I0 of the identification matrix and the unit matrix, respectively. From this, we can then calculate the pseudo-periodogram as estimated by the digital beamforming approach, which has an intensity peak when θ = θ k , P(θ)=A*(θ)RxA(θ)
[0104] While this method is more general, it is slower than the Fourier transform-based approach. Beamforming involves scanning for a good delay that produces a peak. This approach is closer to cross-correlation, whereas the FFT-based spectral density is an autocorrelation of the signal, which can still be efficiently evaluated numerically.
[0105] In the following section, we provide a brief overview of the common steps involved in applying NuFFT, using the NuDFT expression. Since it is a new numerical approximation method, there are many variations. The usual approach is to adjust the signal to match the conditions to be used with the regular FFT algorithm.
[0106] The signal model is rewritten using our notations as follows: xmkn(t)=γkexp(−j2πfΘΔmn)exp(j2πfRkt)
[0107] This results in the following NuDFT formulation: X^^k[fR,fΘp] =
[0108] Given the non-uniform DFT (NuDFT = non-uniform discrete Fourier transform) expression according to equation (23), to simplify the expression and without losing generality with respect to the problem at hand, the distance contributions are not taken into account and the signal of interest (for a single target k and a virtual antenna v) can be defined as y v (t) = exp(-j2πf θk dx v ). This results in the following NuDFT: Xk[fp]=∑v=1Vexp(j2πdxv(fp−fk))
[0109] Where X [f p ] the DFT coefficient at frequency point f p , V is the number of virtual antennas, dx vis the position of the v-th antenna in the virtual antenna grid, f k is the angular frequency of the k-th target, and f p is the frequency point in the DFT.
[0110] The number of virtual antennas V is the product of the number of transmitter antennas and receiver antennas, V = N×M. To calculate the NuFFT along the array aperture direction, it is important that it is sampled along the 1D aperture. As described by the signal model according to equation (3), in the far field, the phase effect of the position of the transmitter antenna and the receiver antenna can be expressed in a single variable dx v This is advantageous because it can reduce the calculation along the aperture (to obtain the DOA) to a single variable instead of the two positions of the transmitter and receiver.
[0111] For a MIMO system and in the far field, the virtual array, antenna array, or antenna grid, is formed by considering all possible pairs of transmit and receive antennas. With M transmit antennas and N receive antennas, the result is a virtual array with M×N elements. Because the information from all antenna combinations is utilized, this larger virtual array, antenna array, or antenna grid, can provide better resolution and greater coverage compared to a physical array with the same number of elements.
[0112] In the signal model according to equation (3), this amounts to factoring the angle contributions −j2πλ(dxRmsin(θk)) and −j2πλ(dxTnsin(θk)) with factor dx RTmn = dx Rm + dx Tn , the virtual array coordinate, which is parameterized by the number of transmitters n and receivers m. The signal model is then x^mkn(t)=γkexp(−j2πλ(dxRTmnsin(θk)))exp(j4πSRkct) using the virtual field coordinate dx RTmn This enables the extension of regular radar processing methods for MIMO arrays.
[0113] The NuFFT algorithm can accept the following elements as input (in a 1D formulation): - Inconsistent data points y v , for v = 1, 2, ... , V . - Corresponding non-uniform locations dx v , for v = 1, 2, ... , V . - Desired Fourier frequencies f p(k) , for k = 1, 2, ... , K.
[0114] The first step of the algorithm is rasterization. This step is essentially an interpolation process. The rasterized data Y p serve as an approximation of the original signal at the newly defined grid points. We map our patchy data onto a uniform grid. For each patchy data point y vat location dx v we calculate its contribution to the neighboring grid points. The rasterized data Y p are calculated as follows: Yp=∑v=1Vyvexp(−j2πfp(k)dxv)ϕ(uvp) where uvp=dxv−pR is the offset function and R is the oversampling factor. The screening kernel ϕ(u) using a Kaiser-Bessel function is given by: ϕ(u)={1WI0[πW21−(2uW)20 otherwisefor|u| <W2 where W is the kernel width and I0 is the modified Bessel function of the first kind.
[0115] The Kaiser-Bessel function is typically used as a mesh kernel because of its advantageous properties.
[0116] Properties in maintaining signal integrity during redistribution.
[0117] Next, and because the gridded data is uniformly distributed, we perform a standard Fast Fourier Transform (FFT) on the gridded data. We map the grid index p to the Fourier frequency index k by fp(k)=pR: Y^p(k)=FFT{Yp}
[0118] Finally, the convolution can be corrected using the Fourier transform of the lattice kernel.
[0119] Since the raster function convolves with the signal during the rasterization process, in this step the Fourier-transformed data is deconvolved by dividing it by the Fourier-transformed raster function, G(f p(k) ), are divided. We perform a deconvolution by dividing the FFT result by the Fourier transform of the gridding kernel G(f p(k) ) is shared: Xk=Y^p(k)G(fp(k)) where G(f p(k) ) is the Fourier transform of the grid kernel and is given by: G(fp(k))=11−(2fp(k)W)2 for |f p(k) | ≤ W / 2 and otherwise 0.
[0120] The corrected Fourier transform X k is a numerical estimate of the original signal data at the desired Fourier frequencies f p(k) , for k = 1, 2,..., K.
[0121] A short list of the different symbols and their meaning: - y v : Inconsistent data points. - dx v : Non-uniform places corresponding to yv. - f p(k) : Desired Fourier frequencies. - Y p : Rasterized data. - Ŷ p(k) : FFT of the rasterized data. - X k : Corrected Fourier transform, the final result. - and vp : Offset function. - R: Oversampling factor. - ϕ(u): Gridding kernel function. - G (f p(k) ): Fourier transform of the lattice kernel. - W : Half width of the raster kernel in the frequency domain.
[0122] The following explains linear frequency modulation for so-called LFMCW (linear frequency modulated continuous wave) systems. This involves linear frequency modulation and a continuous wave mode of operation. The signal is a sequence of frequency chirps. Linear frequency modulation (chirp) has the following form: f(t) = f0 + St, where f0 is the carrier frequency, S = B / T c is the modulation rate and t is the time variable.
[0123] The phase of the signal of the transmitting antenna n or of a transmitting antenna element n is the time integral of the frequency ϕn(t)=∫0tf(τ)dτ=∫0t(f0+Sτ)dτ ϕn(t)=f0t+12St2
[0124] It follows that the transmitted and received signals are as follows: xn(t)=A cos(2π(f0t+12St2))xmk(t)=A cos(2π(f0(t−τmk)+12S(t−τmk)2))
[0125] The received signal X mk (t) is around the flight time τ mk between the individual targets and the receiver. The index n represents a transmitting antenna element, namely the nth transmitting antenna element. The index m represents a receiving antenna element, namely the mth receiving antenna element. The index k represents an object.
[0126] The received signal is multiplied directly with the transmitted signal to mix or dechirp the received signal. The result is a mixed signal xmix(mkn)(t)=xn(t)xmk(t) with two contributions at high and low frequencies. The high-frequency contribution is then filtered out, leaving only the low beat frequency of the signal, i.e., the frequency difference between the two signals. xmix(mkn)(t)=xn(t)xmk(t) xmix(mkn)(t)≈12cos(2π(f0τmk−Sτmkt+12Sτmk2))
[0127] The quadratic term in τ mk can be neglected since the delay τ mk does not make a significant contribution in the far field of the radar system or since τ mk is short. This means that xmix(mkn)(t)≈12cos(2π(f0τmk−Sτmkt)
[0128] This section describes 3D imaging with the radar system. In addition to range and azimuth, elevation information is considered. Elevation is analogous to azimuth and functions similarly even with a sparse array. Both dimensions are orthogonal, and therefore the expression and contributions have similar expressions (each with its own variables). The field of view with respect to elevation is also typically much smaller.
[0129] The 3D spectral density taking into account the elevation at the angular frequency f Φ and in connection with the contributions of objectives in f Φk and over an aperture or stop B is: X^^[fR,fΘ,fΦ]=∑k=1K γk exp(jπ(fRk−fR)Tc)sin(π(fR−fRk)Tc)sin(π(fΘ−fΘk)A)sin(π(fΦ−fΦk)B)π(fR−fRk) π(fΘ−fΘk) π(fΦ−fΦk)
[0130] We point out that for a complete 3D description of the signal DFT at hand, it may be necessary to introduce the two electrical angles depending on the usual spherical coordinates and to define the two independent frequencies f Θ and f Φ and the expressions given below to obtain the representation with a 2D sparse antenna array.
[0131] As a rule, a signal model can be assumed that has the range R k , the azimuth θ k and the elevation ϕ k with respect to the target k and which is as follows: x^mkn(t)=γkexp(−j2πλ(dxRm+dxTn)sin(θk)cos(ϕk)))exp(−j2πλ(dyRm+dyTn)sin(ϕk)))exp(j4πSRkct)
[0132] The virtual array or antenna grid is extended in 2D in the plane perpendicular to the radiation propagation. The coordinates of the virtual array can be indexed with two variables, one for the horizontal coordinates of the antenna elements dx RTmn = dx Rm + dx Tn and the other for the vertical coordinates dy RTmn = dy Rm + dy Tn Then we introduce a variable change for the angular coordinates so that they become independent for the two orthogonal directions. We define p k = sin(θ k )cos(ϕ k )and q k = sin(ϕ k ). With this change, the signal model (with dx RTmn = dx and dy RTmn = dy): x^mkn(t)=γkexp(−j2πλdx pk)exp(−j2πλdy qk)exp(j4πSRkct)
[0133] With this equation, the variables associated with both angular directions are separated, and the same scheme as described above is possible using the NuFFT for the conjugate variables dx and p_k as well as the conjugate variables dy and q k applicable.
[0134] The inverse transformation to recover θ k and ϕ k also comes directly from the knowledge of p k and q k . We have ϕ k = arcsin (q k ) and θk=arcsin(pk1−qk2). This allows the use of sparse 2D arrays with the NuFFT processing described above for the 1D case and the performance of 3D imaging in the static case and 4D imaging in the dynamic case with range, Doppler velocity, azimuth and elevation.
[0135] Finally, the processing presented here is fully compatible with 4D imaging using a sparse array. The time component of the signal model can be supplemented by the contribution of the Doppler velocity phase. The signal model is then converted to x mkn (t) = x̂ mkn (t)e j2πΔft , where Δf=2 vλ is the frequency shift resulting from the radial velocity. In most radar systems, the signal is sampled by the digitizer at regular time intervals. The range and Doppler velocity both depend on the time variable t and can be processed using a regular joint 2D FFT for the fast (range) and slow (Doppler velocity) contributions of the signal. We can also recall that the regular FFT is a limiting case of the NuFFT with regular sampling intervals. This allows the radar system to operate even with irregular time samples by using a 2D NuFFT for the time variable.
[0136] Finally, the proposed processing formalism can encompass 4D processing (azimuth, elevation, range, velocity) with a sparse array and potentially irregular samples in the time domain in a single formalism using a 4D NuFFT. In the usual case of regular sampling of the signal by the acquisition system, the NuFFT reduces to the regular FFT.
[0137] Further advantages of the invention are described in the following exemplary embodiments. They show: Fig. 1 a schematic view of a radar unit for determining a position and / or a speed of an object, Fig. 2 an excerpt from the signal generation, signal acquisition and evaluation and Fig. 3 a frequency-time diagram of the output signal and the reflection signal.
[0138] Fig. Figure 1 shows a schematic view of a radar unit 2 for determining the position and / or speed of an object 1. Object 1 is, for example, a vehicle, a pedestrian, a bicycle, a tree, or a pillar. Object 1 can be detected using radar unit 2. This is used by driver assistance systems to make road traffic safer. In particular, accidents can be avoided or prevented.
[0139] The radar unit 2 is also arranged on and / or in a vehicle. The radar unit 2, also called a radar system, comprises a computing unit 17 and / or a data memory 18. The computing unit 17 can execute the method described here, and the data required for this purpose, such as a program and / or the positions of antenna elements 4-10, can be stored in the data memory 18.
[0140] The radar unit 2 further comprises an antenna unit 3 having a plurality of antenna elements 4-10. The antenna elements 4-10 are arranged in an irregular and / or sparse pattern. However, this pattern is not random. Rather, the antenna elements 4-10 are arranged such that the signals detected thereby can be evaluated using the method described here, in particular using non-uniform fast Fourier transformation.
[0141] Output signals 11 can be transmitted using the antenna elements 4-10 arranged in an irregular pattern. These output signals 11 strike the object 1 and are reflected by it. They return to the radar unit 2 as reflected signals 12 and are detected by the antenna elements 4-10 arranged in an irregular pattern. The antenna elements 4-10 can thus be operated as transmitting and receiving antennas.
[0142] According to the antenna unit 3 shown, it also includes unoccupied antenna positions. In an antenna unit 3 with antenna elements 4-10 arranged in a regular pattern, all antenna positions are occupied. These can then be spaced at a distance of λ / 2, where λ is the wavelength of the signals 11, 12. Such antenna units 3 are also referred to as ULAs, which stands for uniform linear arrays.
[0143] However, in the method described here, in particular a computer-implemented method, an antenna unit 3 is used, wherein the antenna elements 4-10 are arranged in an irregular pattern. The antenna elements 4-10 are arranged here such that, in a virtual arrangement of the antenna elements 4-10, in particular in MIMO (multiple input multiple output) operation, ambiguities in the detection of the object 1 are prevented. In some scenarios, two objects 1 can be detected in a random pattern after evaluation, even though only one object 1 is present. The method described here uses so-called sparse (antenna) arrays.
[0144] Furthermore, an unoccupied antenna position 13 is shown here. The unoccupied antenna positions 13 (the antenna unit 3 has several unoccupied antenna positions 13) form an irregular pattern of antenna elements. Additionally or alternatively, the antenna elements 4-10 are arranged sparsely. This requires fewer antenna elements 4-10 compared to ULAs (uniform linear arrays).
[0145] To determine the position and / or speed of object 1, a non-regular Fourier transformation, in particular a non-regular discrete Fourier transformation and / or a non-regular fast Fourier transformation algorithm, is performed. This method can evaluate the signals, in particular the detected reflection signals 12, that are detected by the antenna elements 4-10 arranged in an irregular pattern.
[0146] An angle 14 can be determined as the position of object 1, which is also referred to as θ in the previous description. The angle 14 or θ is measured here relative to a center line 16 of the radar unit 2 and / or the antenna unit 3. If the radar unit 2 is arranged, for example, in the front bumper of a car, the object 1 is located to the right in front of the car.
[0147] Additionally or alternatively, a distance 15 to object 1 can also be determined using the method described here. The distance 15 is also referred to as R in this description. The distance 15 or R is measured here between object 1 and the intersection point between center line 16 and radar unit 2 and / or antenna unit 3.
[0148] Fig. Figure 2 shows a section relating to the transmission, reception, and evaluation of the signals, particularly the reflection signals. The output signals 11, which are transmitted by the antenna elements 4-10, are generated with the aid of a signal generator 19. The signal generator 19 does not directly generate the output signals 11, but rather generates corresponding electrical signals, which are converted into the output signals 11 by the antenna elements 4-10. The signal generator 19 can be connected to the processing unit 17 via a connection not shown here. Consequently, the processing unit 17 can transmit control commands to the signal generator, so that the latter can cause the antenna elements 4-10 to transmit the output signals 11.
[0149] Furthermore, a switching unit 22 is shown here, which forwards the signals coming from the signal generator 19 to the corresponding antenna elements 4-10. The switching unit 22 can form a connection to each antenna element 4-10, so that the signal from the signal generator 19 can be routed to the corresponding antenna element 4-10. The switching unit 22 can switch back and forth between the antenna elements 4-10. The switching unit 22 can be a demultiplexer, for example.
[0150] At object 1, which is a car here, they are reflected and return as reflection signals 12 and are detected again by the antenna elements 4 - 10. As electrical signals, the output signals 11 and the reflection signals 12 have the form s n (t) and s m (t), as they are in Fig. 2. The antenna elements 4-10 convert electrical signals into the output signals 11. Furthermore, the antenna elements 4-10 convert the reflected signals 12 into corresponding electrical signals. These two signals are fed to a multiplier 20, which multiplies the two signals s n (t) and s m (t), representing the output signals 11 and the reflection signals 12. The index n represents an output signal 11 and the index m represents a reflection signal 12.
[0151] With the help of a low-pass filter 21, the multiplied signals s n (t)*s m (t)=x mn(t) filtered. This is followed by the computing unit 17, which executes the further process. Alternatively, the multiplier 20 and / or the low-pass filter 21 and / or the signal generator 19 can also be integrated into the computing unit 17. Furthermore, a connection, for example for data exchange, can also be present between the computing unit 17 and the multiplier 20, the low-pass filter 21, the signal generator 19, the switching unit 22 and / or the data memory 18 (not shown here).
[0152] The computing unit 17 can also be embodied as a microcontroller or a microprocessor. The computing unit 17 can also be integrated into the microcontroller or the microprocessor. The elements shown here can be interconnected via data connections and / or signal connections. Additionally or alternatively, at least some of the elements shown here can be integrated into the computing unit 17 and / or the microcontroller.
[0153] However, at least some of the elements shown here can also be arranged outside the computing unit 17 and / or the microcontroller. For example, the data memory 18 can be arranged outside and / or formed as a separate chip.
[0154] Furthermore, the elements mentioned and shown here can be arranged on a circuit board. The antenna unit 3 and / or the antenna elements 4-10 can also be arranged on the circuit board. Alternatively, the antenna unit 3 and / or the antenna elements 4-10 can also be connected to the circuit board via a data connection and / or signal connection, for example, in the form of a cable.
[0155] Fig.Figure 3 shows a frequency-time diagram of the output signal 11 and the reflection signal 12. In the diagram shown here, the output signal 11 and the reflection signal 12 are shown as a pulse, although this pulse is not to be understood as a temporal pulse. The pulse is, as shown here, an increase in frequency from f o to f0 + B. After the maximum frequency f0 + B is reached, the output signal 11 can start again at f0. Since the reflection signal 12 is based on the output signal 11, the reflection signal 12 has essentially the same shape. The output signal 11 and reflection signal 12 shown here are also referred to as chirp, as is also partially done in the previous description. The output signal 11 and the reflection signal 12 have a time duration T C Here, τ denotes the time until the reflection signal 12 is detected again after the output signal 11 has been transmitted. τ=2Rc with c the speed of light a connection is made to the distance 15 or R. List of reference symbols 1 object 2 radar units 3 Antenna unit 4 first antenna element 5 second antenna element 6 third antenna element 7 fourth antenna element 8 fifth antenna element 9 sixth antenna element 10 seventh antenna element 11 output signals 12 reflection signal 13 unoccupied antenna position 14, θ angle 15, R Distance 16 Center line 17 Computing unit 18 data storage 19 Signal generator 20 multiplier 21 low-pass filter 22 Switching unit
Claims
[1] Radar unit, in particular vehicle radar unit, for determining a position, in particular direction and / or distance, and / or a speed of an object (1) in road traffic, with an antenna unit (3) with a plurality of antenna elements (4 - 10), wherein the antenna elements (4 - 10) are arranged in an irregular pattern, wherein output signals (11) can be emitted with the aid of the irregularly arranged antenna elements (4 - 10) and wherein reflection signals (12) can be detected with the aid of the irregularly arranged antenna elements (4 - 10), and with a computing unit (17) which is designed and / or configured to evaluate the detected reflection signals (12) to determine the position and / or the speed of the object (1) by means of a non-regular Fourier transformation which can evaluate the signals, in particular the detected reflection signals (12), from the antenna elements (4 - 10) arranged in an irregular pattern. [2] Radar unit according to the previous claim, characterized by that the antenna elements (4 - 10) are arranged on grid points of a, in particular two-dimensional, grid, wherein some grid points are unoccupied. [3] Radar unit according to one or more of the preceding claims, characterized by that grid points arranged next to each other have a distance of λ / 2. [4] Radar unit according to one or more of the preceding claims, characterized bythat the antenna elements (4 - 10) are arranged in a circle around the grid points, the circle having a radius of 5% - 20% of λ / 2. [5] Radar unit according to one or more of the preceding claims, characterized by that the antenna unit (3) has a sparse grid. [6] Radar unit according to one or more of the preceding claims, characterized by that positions of the antenna elements (4 - 10) are stored in a data memory (18) and the positions of the antenna elements (4 - 10) can be obtained from the data memory (18) for carrying out the non-regular Fourier transformation. [7] Radar unit according to one or more of the preceding claims, characterized bythat the computing unit (17) is designed and / or configured such that it can determine the direction, in particular an angle (14; θ), of the object (1) to the antenna unit (3) with the aid of the non-regular Fourier transformation and the antenna elements (4 - 10) arranged in an irregular pattern. [8] Radar unit according to one or more of the preceding claims, characterized by that the computing unit (17) is designed and / or configured such that it can determine a distance (15; R) of the object (1) to the antenna unit (3) with the aid of the non-regular Fourier transformation and the antenna elements arranged in an irregular pattern. [9] Radar unit according to one or more of the preceding claims, characterized bythat the computing unit (17) is designed and / or configured such that it can apply a non-regular fast Fourier transformation algorithm to determine the position, in particular the direction, and / or the distance (15; R) of the object (1). [10] Radar unit according to one or more of the preceding claims, characterized by that the computing unit (17) is designed and / or configured such that it can operate the antenna elements (4 - 10) by means of a MIMO method. [11] Radar unit according to one or more of the preceding claims, characterized by that the antenna elements (4 - 10) are arranged such that in a virtual arrangement of the antenna elements (4 - 10) all antenna elements (4 - 10) are spaced apart from one another. [12] Radar unit according to one or more of the preceding claims, characterized bythat the computing unit (17) is designed and / or configured such that at a time a selected antenna element (4 - 10) can transmit the output signal (11) and at least some other, in particular all, antenna elements (4 - 10) can detect the reflection signals (12). [13] Radar unit according to one or more of the preceding claims, characterized by that the computing unit (17) is designed and / or configured such that it can carry out a sequence in which, at respective points in time, a selected antenna element (4 - 10) can transmit the output signal (11) and at least some other, in particular all, antenna elements (4 - 10) can detect the reflection signals (12), wherein at successive points in time, other antenna elements (4 - 10) can transmit the output signals (11) and at least some other, in particular all, antenna elements (4 - 10) can detect the reflection signals (12). [14] Radar unit according to one or more of the preceding claims, characterized by that the radar unit (2) and / or the computing unit (17) is designed and / or configured such that the output signals (11) have at least one output pulse, wherein the output pulse is frequency-modulated such that a frequency of the output pulse, in particular from an initial frequency, increases preferably linearly by a signal bandwidth (B). [15] Radar unit according to one or more of the preceding claims, characterized by that the radar unit (2) and / or the computing unit (17) is designed and / or arranged such that output signals (11) with a frequency between 76 and 81 GHz can be generated. [16] Radar unit according to one or more of the preceding claims, characterized bythat the radar unit (2) and / or the computing unit (17) is designed and / or arranged such that the output signals (11) have a plurality of output pulses. [17] Radar unit according to one or more of the preceding claims, characterized by that the computing unit (17) is designed and / or arranged in such a way that it calculates several product values x mkn (t) by multiplication of the respective output signals s n (t) and the respective reflection signals s mk (t) can form. [18] Radar unit according to one or more of the preceding claims, characterized by that the product values x mkn (t), in particular by means of a low-pass filter (21), so that filter values x̃ mkn (t) can be formed. [19] Radar unit according to one or more of the preceding claims, characterized bythat the computing unit (17) is designed and / or arranged such that it calculates the filter values x̃ mkn (t) can be simplified so that approximate values x̂ mkn (t) can be formed. [20] Radar unit according to one or more of the preceding claims, characterized by that the computing unit (17) is designed and / or arranged in such a way that it can neglect quadratic terms during the simplification in order to preferably ϕ IF the filter values x̃ mkn (t) and / or take into account geometric relationships between the antenna elements (4 - 10). [21] Radar unit according to one or more of the preceding claims, characterized by that the computing unit (17) is designed and / or arranged in such a way that it calculates the approximate values ̂ẋ̂ mkn (t) to a common point in time. [22] Radar unit according to one or more of the preceding claims, characterized bythat the computing unit (17) is designed and / or arranged in such a way that it calculates the approximate values ̂ẋ̂ mkn (t) with respect to time, preferably a fast Fourier transformation can be applied. [23] Radar unit according to one or more of the preceding claims, characterized by that the computing unit (17) is designed and / or arranged in such a way that it has a maximum of X̂ for determining the distance (15; R) mkn [f R ] can be determined. [24] Radar unit according to one or more of the preceding claims, characterized by that the computing unit (17) is designed and / or arranged in such a way that it calculates the approximate values ̂ẋ̂ mkn (t) can be Fourier transformed with respect to position, preferably using a fast Fourier transform. [25] Radar unit according to one or more of the preceding claims, characterized bythat the computing unit (17) is designed and / or arranged in such a way that it calculates the approximate values ̂ẋ̂ mkn (t) can first be Fourier transformed with respect to time and then Fourier transformed with respect to position. [26] Radar unit according to one or more of the preceding claims, characterized by that the computing unit (17) is designed and / or arranged in such a way that, after the application of the Fourier transformation with respect to time, the non-regular discrete Fourier transformation with respect to location can be applied, with X^^k[fR,fΘp] = [27] Radar unit according to one or more of the preceding claims, characterized bythat the computing unit (17) is designed and / or configured such that, after determining a position, in particular direction and / or distance, and / or a speed of an object (1) in road traffic, it can output the corresponding result. [28] Radar unit according to one or more of the preceding claims, characterized by that the radar unit (2) comprises a microcontroller which is the computing unit (17) or in which the computing unit (17) is integrated. [29] Radar unit according to one or more of the preceding claims, characterized bythat the computing unit (17) and / or the microcontroller are designed and / or configured in such a way that they can control and / or control the radar unit (2), can determine the position, in particular direction and / or distance, and / or the speed of the object (1) in road traffic, can evaluate and / or process the signals occurring in the process and / or can output the associated result(s). [30] Radar unit according to one or more of the preceding claims, characterized by that the computing unit (17) and / or the microcontroller is connected to the antenna unit (3), in particular the antenna elements (4 - 10). [31] Radar unit according to one or more of the preceding claims, characterized by that the radar unit (2) comprises a data memory (18), a signal generator (19), a multiplier (20), a low-pass filter (21) and / or a switching unit (22). [32] Radar unit according to one or more of the preceding claims, characterized by that the data memory (18), the signal generator (19), the multiplier (20), the low-pass filter (21) and / or the switching unit (22) are integrated in the computing unit (17) and / or in the microcontroller and / or that the data memory (18), the signal generator (19), the multiplier (20), the low-pass filter (21) and / or the switching unit (22) are connected to the computing unit (17) and / or to the microcontroller. [33] Radar unit according to one or more of the preceding claims, characterized by that the data memory (18), the signal generator (19), the multiplier (20), the low-pass filter (21), the switching unit (22), the computing unit (17) and / or the microcontroller are arranged on a circuit board and / or that the antenna unit (3) is arranged on the circuit board or, in particular by means of a connector arranged on the circuit board, is coupled to the circuit board. [34] Vehicle with a radar unit (2) designed according to one or more of the preceding claims. [35] Computer program which, when executed by a computing unit (17), causes the computing unit (17) to determine a position, in particular direction and / or distance, and / or a speed of an object (1) in road traffic by means of a radar unit (2) in a vehicle, wherein the radar unit (2) is designed according to one or more of the preceding claims. [36] Computer-readable storage medium comprising a computer program, wherein the computer program, when executed by means of a computing unit (17), causes the computing unit (17) to determine a position, in particular direction and / or distance, and / or a speed of an object (1) in road traffic by means of a radar unit (2) in a vehicle, wherein the radar unit (2) is designed according to one or more of the preceding claims.
