Slow Phase Time Division Multiple Access (SP-TDMA) Modulation Scheme for MIMO Radar

Abstract

A millimeter-wave automotive radar is disclosed that includes a plurality of millimeter-wave transmitters configured to simultaneously transmit phase-encoded signals according to a phase modulation scheme, and a plurality of subarrays operatively coupled to respective ones of the plurality of millimeter-wave transmitters, each subarray including a predefined number of transmitter elements, and configured to transmit the phase-encoded signals received from a respective one of the millimeter-wave transmitters from other subarrays in different, mutually orthogonal time slots using time division multiplexing.

Description

[Technical Field]

[0001] The present disclosure relates to radar systems, and more particularly to a Slow Phase-Time Division Multiple Access (SP-TDMA) transmit modulation scheme in a Multiple Input Multiple Output (MIMO) radar. [Background technology]

[0002] It is well known to those skilled in the art that multiple-input / multiple-output (MIMO) technology, which involves measuring the magnitude and phase of signals received by N receivers (multiple outputs) for each of M transmitters (multiple inputs), can be used to create a "virtual" receive array that is larger than the physical receive array, thereby improving the angular resolution of an object. There are various modulation schemes for MIMO that aim to orthogonalize the signals transmitted by individual transmitters so that they can be separated at the receiver to form a virtual array. These include, but are not limited to, time division multiplexing (TDM), frequency division duplex (FDD), Doppler division multiple access (DDMA), and binary phase modulation (BPM). In TDM MIMO, transmitters transmit signals one at a time, and each receive element receives the signals from each transmitter sequentially in time, allowing the signals to be separated according to which transmitter they originate from. When the spacing between the transmitting and receiving elements is set appropriately, the signals at each receiver can be rearranged to correspond to the transmitter from which they originated, and the phase difference of each received radiation wave will make it appear as if there are equivalently more receiving elements than actually exist, thereby creating "virtual" elements.

[0003] Figure 1 shows a MIMO virtual array that consists of (M × N) elements but uses only (M + N) physical antenna elements, much larger than the physical number of transmitters and receivers required. The angular resolution of the virtual array is therefore much finer than that of the physical array if not used in the MIMO implementation.

[0004] With further reference to FIG. 1, a first transmitter, TX1, transmits a signal via an antenna with a wide antenna radiation pattern; the signal strikes an object at an angle θ and is reflected; the reflected wavefront across the four receivers experiences a progressive phase delay between elements of Δφ=(2π / λ)*d*sin(θ) radians, where λ is the wavelength of the transmit frequency and d is the distance between the receive antennas. Similarly, when a second transmitter, TX2, transmits a signal, it experiences a similar progressive phase delay between elements. However, the spacing between elements TX1 and TX2 adds an additional progressive phase to these (2π / λ)*4d*sin(θ) radians phase terms, where 4d is the separation distance between the transmit antennas of TX1 and TX2. Separating the signals across the receive array (and writing in terms of Δφ) results in the following relative phase: {0,Δφ,2Δφ,3Δφ,4Δφ,5Δφ,6Δφ,7Δφ} This proves to be comparable to the conventional case. Such systems are popular, and several automotive mmWave (76-81 GHz) radar transceiver chips suitable for MIMO are currently commercially available, each with a varying number of transmit and receive channels that can be further increased using multiple serial chips.

[0005] Figure 2 shows a conventional MIMO radar, where 4 × transmitter is at distance d t The receiver is offset by 4 × the distance d rThe received signal is offset by a factor of 10. Such MIMO radars typically use Frequency Modulated Continuous Wave (FMCW) modulation, which is another common type of modulation known to those skilled in the art. An FMCW waveform, also known as a chirp, is a complex sine wave whose frequency varies over time. FMCW radars transmit chirps at a period called the Pulse Repetition Interval (PRI), most often in a sawtooth configuration (typically increasing linearly with time), although other chirp types exist. The resulting target echo from a single object in the scene will contain a delayed and attenuated copy of the transmitted chirp. Mixing the received signal with the transmitted chirp results in a complex sine wave. This waveform is known as a beat signal, and its frequency is directly proportional to the distance to the detected object.

[0006] Figure 3 shows multiple chirps collected within a single "frame," allowing Doppler frequency variation in the "slow time" dimension to be determined. When multiple objects are present in the scene, the mixer produces a signal consisting of the sum of multiple sinusoids of varying amplitude, phase, and frequency originating from these objects. The spectrum of frequencies output from the mixer can provide the distances of the multiple objects. Beat frequency estimation is typically performed in the digital domain after digitally sampling the beat signal. Because the beat frequency is much smaller than the radar bandwidth, a slow analog-to-digital converter (ADC) can be used. By sampling the beat signal and placing the samples for each chirp in a separate column of a matrix, the row index of the matrix corresponds to "fast" time across a single chirp, and the column index corresponds to "slow" time across multiple chirps. By employing a Fast Fourier Transform (FFT) to calculate the Discrete Fourier Transform (DFT) of each column of the matrix, the range of objects in the radar field of view can be calculated by determining their beat frequency, and by applying a further FFT along the rows of the matrix, the velocity of the objects can be calculated through detection of their Doppler frequency. The use of these two FFTs is commonly referred to as a 2D FFT (also known as a range-Doppler FFT), which allows the determination of objects in terms of both range and velocity. Performing a 2D FFT has the advantage of reducing the noise level through matched filtering of the object's beat frequency and Doppler frequency. Obviously, the number of objects that fit into the same range-velocity binary is usually small, depending on the radar's range and velocity resolution.

[0007] Figure 4 shows a range-velocity plot. While this plot provides a lot of useful information, it lacks detail about the object's angular position. Angular position is determined by taking the range-Doppler binaries from each receiver channel's 2D FFT output (including both real and virtual channels via TDM) and performing a third "angle" or 3D FFT across each receiver's 2D FFT output. Note that this is typically done after a Constant False Alarm Rate (CFAR) thresholding step, which retains only binaries with a signal-to-noise ratio above a certain threshold. The object's velocity can be determined from the Doppler frequency shift of the detected signal. The Doppler effect, the well-known change in wave frequency associated with changes in relative motion between the source and the observation point, appears in the detected radar beat signal if the object is moving relative to the radar. The beat signal is digitized into discrete range binaries. Because the chirp duration is very short, the change in IF frequency is too small to shift the object's beat signal to a different range binary across successive chirps. However, by comparing the phase of the sinusoidal frequency components across successive chirps, the velocity of the object can be estimated. The phase delay of the signal reflected from a target at range R is given by

number

[0008] For a moving object, the phase changes by Δφ across successive chirps, corresponding to a change in distance ΔR:

number

number

[0009] Figure 5 shows the time -1 , v max , and v max Shown is an example of a range-Doppler profile for a target at higher velocities, where the Doppler phase increases to >π and "wraps" to appear in the Doppler binary corresponding to negative velocities. This illustrates the effect of having an "ambiguous" phase; in the case of the last plot, unless the phase wrap is known, the decoded velocity calculation will be incorrect.

[0010] One way to further increase the angular resolution of radar systems is to use techniques such as sparse arrays, but it is well known that angular resolution is proportional to the number of available channels (or virtual channels in MIMO radars). This means that using current approaches would require more radar transceiver chips, making them more expensive and therefore prohibitive for automotive / mass-production applications that require low cost.

[0011] Patent document 1 (US2021 / 239791A1) discloses combining two or more antennas at a common port on each transmit port of a radar transceiver chip. Each antenna is designed to operate in a different frequency band, with a frequency gap between these bands. The antennas use a different design concept: one antenna is on a PCB and the other is a waveguide. To keep the signals arriving at each antenna separate, this frequency separation is set to at least one-tenth of the operating band. Similarly, it is implied that the antennas have orthogonal polarizations so that the signal entering the other path from one antenna is reduced by at least 20 dB.

[0012] Patent Document 2 (US2022 / 107402A) discloses a radar device including multiple transmit antennas, each transmitting a transmit signal, and a radar transmitter that applies a Doppler shift to the transmit signal transmitted from each of the multiple transmit antennas. It is disclosed that each TX antenna is provided with a different phase code. Therefore, there is a need for a MIMO configuration that increases radar range, increases maximum unambiguous Doppler velocity, and improves angular resolution without increasing the number of radar transceiver chips. [Prior art documents] [Patent documents]

[0013] [Patent Document 1] U.S. Patent Application Publication No. 2021 / 239791 [Patent Document 2] US Patent Application Publication No. 2022 / 107402 Summary of the Invention

[0014] The present invention relates to a MIMO radar as defined in the appended claims.

[0015] In one aspect, a millimeter-wave digital automotive radar is provided, including a plurality of millimeter-wave transmitters configured to simultaneously transmit phase-encoded signals according to a phase modulation scheme, and a set of antenna elements corresponding to each of the millimeter-wave transmitters, forming a plurality of sets of antenna elements for the corresponding plurality of millimeter-wave transmitters, wherein each set of antenna elements sequentially transmits the phase-encoded signal received from its respective transmitter in a different mutually orthogonal time slot using time division multiplexing, such that each antenna of the set transmits a similar phase-encoded signal in a different time slot to other antennas of the set, and each antenna element is assigned an index number within each corresponding set, such that antenna elements with similar index numbers across the plurality of sets of antenna elements simultaneously transmit during different time slots using different slow phase codes. The millimeter-wave digital automotive radar further includes a plurality of millimeter-wave receivers coupled to the antennas, wherein the millimeter-wave receivers receive phase-encoded signals reflected from objects away from the radar according to a time division multiple access (TDMA) modulation scheme.

[0016] In one embodiment of the present invention, antenna elements with similar index numbers across multiple sets of antenna elements form corresponding antenna subarrays, and the antenna elements of each antenna subarray transmit simultaneously during different time slots using different slow phase codes.

[0017] In one embodiment of the present invention, multiple antenna subarrays transmit through at least one antenna element a phase-encoded signal received from a respective millimeter wave transmitter during each of the orthogonal time slots, such that antenna elements of multiple subarrays transmitting simultaneously during a first time instance continue to transmit simultaneously for different subsequent time instances using different slow phase codes according to a predefined scheme.

[0018] In one embodiment of the present invention, multiple mmWave transmitters are implemented via multiple transmit channels of at least one transceiver chip.

[0019] In one embodiment of the present invention, the multiple antenna subarrays are connected to multiple millimeter-wave transmitters via respective multiple single-input multiple-output millimeter-wave switches, with the input of each switch connected to a respective millimeter-wave transmitter and the multiple outputs of each switch connected to antennas of different antenna subarrays.

[0020] In one embodiment of the present invention, the number of antenna subarrays is the same as the number of outputs of each switch, and the number of antenna elements in each subarray is the same as the number of mm-wave transmitters.

[0021] In an embodiment of the present invention, when the number of switches is four and the number of throw positions of each switch is three, the first throw position of each switch is connected to a respective antenna of the first subarray, the second throw position of each switch is connected to a respective antenna of the second subarray, and the third throw position of each switch is connected to a respective antenna of the third subarray, and the same throw positions of each switch are activated simultaneously to enable simultaneous transmission from the antennas of each antenna subarray.

[0022] In one embodiment of the present invention, the throw positions of each switch are activated in a common predefined sequence to enable transmission of phase-encoded signals from the respective transmitters in mutually orthogonal time slots according to a TDMA modulation scheme.

[0023] In one embodiment of the present invention, the antenna elements of either the mm-wave transmitter, the mm-wave receiver, or both are non-equidistantly spaced.

[0024] In one embodiment of the present invention, the phase modulation is selected from either Doppler Division Multiple Access (DDMA) and Binary Phase Modulation (BPM).

[0025] In one embodiment of the present invention, the millimeter-wave radar further includes a plurality of millimeter-wave transceivers that receive and demodulate phase-encoded signals transmitted by a plurality of millimeter-wave transmitters based on a cyclic sum of subbands, identify the millimeter-wave transmitter of each received signal, and then identify the transmitting antenna element of the millimeter-wave transmitter based on the arrival time of the demodulated signal.

[0026] In one embodiment of the present invention, the millimeter-wave automotive radar further comprises a further plurality of antenna subarrays coupled to respective plurality of millimeter-wave receivers configured to receive signals and simultaneously apply phase codes to the received signals according to a phase modulation scheme, the phase codes being in mutually orthogonal time slots separate from other subarrays using time division multiplexing, and the number of millimeter-wave receivers being the same as the number of antenna elements in each antenna subarray.

[0027] In one embodiment of the present invention, a method of operating a millimeter-wave automotive radar is provided, the method including: enabling multiple millimeter-wave transmitters to simultaneously transmit phase-encoded signals according to a phase modulation scheme; providing a set of antenna elements for each millimeter-wave transmitter to form multiple sets of antenna elements for a corresponding plurality of millimeter-wave transmitters; and enabling the antenna elements in each set to sequentially transmit the phase-encoded signals received from their respective transmitters in separate, mutually orthogonal time slots using time division multiplexing, wherein each antenna of a set transmits a similar phase-encoded signal in a different time slot to other antennas in the set, and each antenna element is assigned an index number within each corresponding set, allowing antenna elements with similar index numbers across the multiple sets of antenna elements to simultaneously transmit during different time slots using different slow phase codes.

[0028] In another aspect, a millimeter-wave digital automotive radar is provided, comprising: a plurality of millimeter-wave transmitters configured to simultaneously transmit phase-encoded signals according to a phase modulation scheme; a plurality of antenna subarrays operatively coupled to the plurality of millimeter-wave transmitters, each antenna subarray including the same number of antenna elements as the number of millimeter-wave transmitters, and each antenna subarray configured to transmit the phase-encoded signals received from its respective millimeter-wave transmitter in separate, mutually orthogonal time slots from the other antenna subarrays using time division multiplexing; and a plurality of millimeter-wave receivers coupled to the antennas, the plurality of millimeter-wave receivers receiving the phase-encoded signals reflected from objects remote from the radar according to a TDMA modulation scheme.

[0029] Various embodiments of the present invention disclose a MIMO radar that utilizes RF switches in conjunction with MIMO radar transceiver chips to increase the number of transmit channels available for a given number of MIMO radar transceiver chips. In this way, higher angular resolution can be achieved without significantly increasing the cost of the radar system, since the cost of the switches is much lower than the cost of the radar transceiver chips. By transmitting simultaneously on multiple channels, the effects of losses due to the switches are eliminated. By incorporating TDMA encoding and DDMA (forming a type of SP-TDMA), the signals from each transmitter and the throwing positions of the switches can be determined to form a virtual array. Using an RF switch with a 1:S pole-to-throw ratio in combination with a MIMO radar transceiver chip with a total of M transmit and N receive channels on the chip to form SP-TDMA MIMO increases the number of MIMO transmit channels and antennas to (S x M), improving angular resolution compared to an equivalent TDM / BPM / DDMA radar using a radar transceiver chip with an equivalent number of channels (assuming the same array topology, e.g., both MIMO arrays are "packed" uniform linear arrays, or both are sparse array columns of the same type) with M transmit and N receive channels.

[0030] Also, there is an equivalent radar transceiver chip having the same number of M transmit channels and N receive channels, but the radar range is extended compared to a MIMO radar system without switches. Further, the present disclosure increases the maximum unambiguous Doppler speed of the radar compared to this TDM MIMO radar system if S < M.

[0031] The maximum unambiguous Doppler speed of the radar also increases when compared to a TDM MIMO radar system using the same radar transceiver chip(s) and switches (or equivalent combination), and realizes the same number of (S×M) MIMO physical transmit channels and N MIMO physical receive channels. Further, this reduces the possibility that multiple targets with different speeds at the same distance obscure each other compared to a DDMA MIMO radar system using the same number of (S×M) transmit channels and N receive channels.

[0032] Thus, a modulation method is provided that can simultaneously use multiple transmitters together with a switch. (The important concept here is the simultaneity that expands the distance). This switch means that more TX channels can be used to improve the angular resolution. To add TX channels that can be transmitted simultaneously, it is a clearer way to use another transceiver chip, but the switch is much cheaper.

[0033] Also provided is a computer program including program instructions for causing a computer program to execute the above method that can be embodied on a recording medium, a carrier signal, or a read-only memory.

Brief Description of the Drawings

[0034] The present invention will be more clearly understood from the following description of its embodiments given by way of example only, with reference to the accompanying drawings. [Figure 1] Shows a conventional 8-receiver element antenna configuration and an equivalent MIMO configuration. [Figure 2] This is an example of a single-chip radar system for MIMO, with four transmit channels and four receive channels. [Figure 3] FIG. 1 shows an example of a commonly used type of FMCW radar chirp and the resulting range and velocity estimates using a 2D FFT. [Figure 4] To calculate the angle, we show an additional FFT over all receivers with a 2D FFT for each receiver. [Figure 5] Examples of FMCW transmission sequences and range-Doppler profiles are shown for a single stationary object, an object moving at vmax, and an object moving at a speed greater than vmax but less than twice the value of vmax. [Figure 6] An example of a time division multiple access (TDMA) FMCW modulation scheme is shown. [Figure 7] An example of a DDMA modulation scheme using three transmitters is shown. [Figure 8] An example of empty subband-based DDMA with NTX=4 and Nsub=6 is shown. [Figure 9] 1 shows one possible type of DSP chain for empty subband-based DDMA. [Figure 10] Combining TDMA and DDMA in accordance with an embodiment of the present invention presents a new scheme known as slow phase time division multiple access (SP-TDMA) scheme. [Figure 11A] FIG. 1 shows a block diagram of an example MIMO radar system employing SP-TDMA scheme using a single radar transceiver chip with 4×TX channels, 4×RX channels across three time slots, and a 4×SP3T switch to increase the number of MIMO transmit antennas to 12, in accordance with an embodiment of the present invention. [Figure 11B]FIG. 1 shows a block diagram of an example MIMO radar system employing SP-TDMA scheme using a single radar transceiver chip with 4×TX channels, 4×RX channels across three time slots, and a 4×SP3T switch to increase the number of MIMO transmit antennas to 12, in accordance with an embodiment of the present invention. [Figure 11C] FIG. 1 is a block diagram of details of an example of an exemplary MIMO radar system in which an SP-TDMA scheme is implemented in the receiver, in accordance with an embodiment of the present invention. [Figure 12] An example of an SP-TDMA transmit modulation scheme using DDMA phase encoding is shown, with four transmitters and four single-pole, triple-throw switches (throws labeled A, B, and C). [Figure 13] A phase diagram representation of three potential unambiguous phase values ​​for an unambiguous phase measurement is shown, with no phase wrap (φv1), a +2π radian phase wrap (φv2), and a -2π radian phase wrap (φv3) φv_neas. [Figure 14] For the system of FIG. 11, an example of a wavefront arriving at a subsection of the virtual array (the subarray corresponding to only TX1) is shown. [Figure 15] In the system in Figure 11, three azimuth FFT outputs from the coherent peak method for a target moving at azimuth angle -30° and <-vmax are shown. [Figure 16] 1 illustrates an embodiment of a SP-TDMA (with DDMA phase encoding) radar DSP chain for use with the present invention. [Figure 17] A second embodiment of a SP-TDMA (with DDMA phase encoding) radar DSP chain is shown, where CFAR is performed earlier in the DSP chain to reduce the processing requirements for demodulation, angular rate and velocity processing. [Figure 18] A third embodiment of a radar DSP chain for SP-TDMA (with DDMA phase coding) using the Chinese Remainder Theorem is shown. DETAILED DESCRIPTION OF THE INVENTION

[0035] Figure 6 shows an example of a time division multiple access (TDMA) FMCW modulation scheme. In a TDM modulation scheme, when multiplexed signals come from different transmitters, each transmitter is transmitted in a separate time slot. This means that the digitized received signals corresponding to each transmitter can be easily separated in time without applying complex demultiplexing schemes, and a virtual array can be constructed from the signals received in each time slot. However, the number of time slots is limited to the number of transmitters, N. TX This comes at the expense of maximum unambiguous velocity, since it increases the chirp repetition period in each transmission channel, reducing the maximum detectable velocity by:

number

[0036] Figure 7 shows an example of a DDMA modulation scheme using three transmitters. In the case of DDMA MIMO, each TX channel transmits simultaneously, so the transmit (TX) power is 10*log compared to TDM modulation. 10 (N TX ) dB increase, which increases the Effective Isotropic Radiated Power (EIRP) by 10*log 10 (N TX ) dB increase. In this method, orthogonality between TX channels is achieved by applying a unique cyclic phase code to each TX channel using a phase shifter, which sequentially changes the phase of each chirp. If the number of TX channels is increased by N TX , and there are K chirps per frame, n TX The phase of the k-th chirp on the k-th TX channel is given by:

number

[0037] This is N TX= 3 TX channel (where the phase is wrapped or reconstructed to within ±π radians of the signal's minimum phase). In the range-Doppler profile, which is the Doppler profile at a specific range, or more specifically, for a specific range binary, three subbands are highlighted with different patterns. The target is displayed in the subbinary 5 of each subband.

[0038] By using the DDMA modulation method, a single target can be observed with each peak being N due to the slow time-phase encoding of the chirp. t N compatible with one of the transmitters TX This gives K = 3 distinct peaks along the Doppler dimension. The Doppler dimension represents the Doppler phase, with the Doppler peak corresponding to TX1 (phase coded at 0 radians across all K chirps) representing the true velocity of the target, and the peaks corresponding to TX2 and TX3 being phase shifted by 2π / 3 radians and -2π / 3 radians respectively, which is the phase shift increment of the DDMA code applied to these channels. The Doppler domain is divided into K / N TX Size N TX subbands, each with a Doppler binary signal of K / N TX It is repeated for each binary. K / N for each subband TX Binaries are called subbinaries, e.g. N TX Each of the subbands is from 1 to K / N TXThe Doppler peak has subbinaries numbered from 1 to 10, with subbinaries numbered from 1 to 10. This leads to various challenges. First, the Doppler peak corresponding to the target's true velocity is the one corresponding to TX1, which has a phase encoding of 0 radians. In this example, the Doppler peak is the local maximum of all Doppler phases for a particular range binary in the Doppler FFT output. However, which Doppler peak this is is not known without further processing (techniques such as the Chinese Remainder Theorem (CRT) can be applied). Furthermore, if multiple targets are at the same range but have different velocities, they may appear in the same range-Doppler subbinary.

[0039] Figure 8 shows the N TX =4 and N sub This figure shows an example of empty subband-based DDMA with n = 6, which recovers clear velocity and maps Doppler binaries to the correct TX channel. TX Apply a phase code to the k-th chirp of the k-th TX channel.

number

[0040] formula N sub , (N sub- N TX ) to the divisor N TX A single target is generated by increasing 2π / N sub Spaced in phase in radians, N in the Doppler dimension TX For this approach to work, the total number of chirps, or Doppler binaries per frame, K, must be greater than or equal to the number of Doppler subbands used, N, to generate an integer number of binaries in each subband. sub It must be divisible by (6 in this case).

[0041] K=24, N TX =4, N subUsing the example of FIG. 8 where =6, each Doppler subband is sub For example, we have n = 4 Doppler subbinaries, so that subbinary 3 in subband 2 corresponds to binary 17 in the Doppler spectrum. In this case, the target return will produce four distinct peaks in the same subbinary in adjacent subbands of the Doppler spectrum, each corresponding to one TX channel, while the same subbinary in the remaining two subbands is empty. This allows us to solve for the ith range binary and the jth Doppler subbinary to find which subbinaries contain detections and which are empty, N sub Different hypotheses result.

number

[0042] Referring to Figure 8, N TX =4,N sub = 6, there are six hypotheses to test for each ith distance, jth Doppler subbinary.

number

[0043] Here, the static target appears at Doppler subbinary 3 in subbands 1, 2, 3, and 4. Taking the magnitude of the Doppler peak as 1, we obtain the following hypothetical value:

number

[0044] Hypothesis 1 has the largest sum and is therefore selected as the correct answer. Therefore, subbinary 3 of Doppler subband 1 is mapped to TX1, subbinary 3 of Doppler subband 2 is mapped to TX2, subbinary 3 of Doppler subband 3 is mapped to TX3, and subbinary 3 of Doppler subband 4 is mapped to TX4. Demodulation of the DDMA subbands is enabled after TX1 is mapped, since the slow time phase encoding added to TX2, TX3, and TX4 is known.

[0045] One additional advantage of empty subband-based DDMA over TDM is that the maximum clear rate after the DDMA demodulation step is v max =λ / 4T c and 1 / N as in the TDM MIMO case. TX The point is that there is no reduction by a factor of .

[0046] Figure 9 shows one possible type of DSP chain for empty subband-based DDMA, where a 2D FFT is performed on the data from each receive channel, resulting in N RX This provides a set of range-Doppler profiles. DDMA demodulation is then performed to map each range-Doppler subbinary to a corresponding TX channel and recover the true velocity of that range-Doppler subbinary. Following this, angle processing is performed before a Constant False Alarm Rate (CFAR) threshold is calculated and a local maximum algorithm can be implemented to determine which range-Doppler angle binaries contain target detections. Further angle processing allows a 4D point cloud (showing the 3D position and velocity of all targets detected in the frame) to be calculated and displayed. Those skilled in the art will recognize that the order in which CFAR and angle processing are performed is not fixed and that it may be more efficient in terms of processing time or priorities to perform the CFAR detection step before the DDMA demodulation and / or angle processing step.

[0047] FIG. 10 illustrates a novel scheme known as slow phase time division multiple access (SP-TDMA) 1000, which combines TDMA and DDMA modulation schemes, in accordance with an embodiment of the present invention. SP-TDMA 1000 is a hybrid modulation scheme that divides multiple transmitters into small subarrays of multiple transmitter elements, assigns a time slot to each subarray, and transmits simultaneously in the assigned time slots. The subarrays simultaneously transmit phase-encoded signals received from their respective millimeter-wave transmitters through at least one transmitter element during each time instance, and multiple subarray transmitter elements that simultaneously transmit during the first time instance continue to transmit simultaneously across subsequent time instances. Simultaneous transmission on multiple channels of the subarrays increases the system's transmit equivalent isotropically radiated power (EIRP) due to wave coherence, thereby increasing the maximum distance at which a radar can detect a target. By applying phase coding so that the elements of each subarray are mutually orthogonal (in a manner similar to DDMA), digital signal processing (DSP) techniques can be used to separate the signals corresponding to each TX channel and create a MIMO virtual array. In this way, orthogonality is achieved by combining phase coding and TDMA. The phase modulation scheme outlined here combines the phase coding used in Doppler division multiple access (DDMA) with the time slotting scheme used in TDMA. However, it should be understood that other phase modulation schemes, such as binary phase modulation (BPM), can also be used. BPM may be simple to implement but difficult to demodulate and may require the implementation of a CRT-like rate expansion method. Other variations on time slot allocation, such as sequentially multiplexing time slots or using nonsequential sequences with varying chirp order to facilitate alternative Doppler expansion methods, are also feasible.

[0048] 11A and 11B show an exemplary MIMO radar system 1100 that implements an SP-TDMA scheme at the transmitting end in accordance with an embodiment of the present invention. The MIMO radar system 1100 may be a millimeter-wave automotive radar that may be designed for use in an automotive four-dimensional radar imaging system. The MIMO radar system 1100 may also be used in any radar system, such as for defense or combat, but is more particularly applicable to automotive applications.

[0049] The MIMO radar system 1100 includes a single radar transceiver chip 1102 with four receive channels and four transmit channels TX1, TX2, TX3, and TX4. Each transmit channel is also called a millimeter-wave transmitter, and each receive channel is also called a millimeter-wave receiver.

[0050] The four transmit channels TX1, TX2, TX3, and TX4 are configured to simultaneously transmit phase-encoded signals according to a phase modulation scheme. The MIMO radar system 1100 includes sets of antenna elements 1107(1), 1108(2), and 1109(3), which sequentially transmit the phase-encoded signals received from each transmitter TX1 in separate, mutually orthogonal time slots using time division multiplexing, such that each antenna in the set transmits a similar phase-encoded signal in a different time slot relative to the other antennas in the set. Thus, a set of antenna elements is provided for each transmit channel. Each antenna element is assigned an index number of 1, 2, or 3 within each set, such that antenna elements with similar index numbers across multiple sets of antenna elements simultaneously transmit during different time slots using different slow phase codes. Multiple millimeter-wave receivers are coupled to the antennas, and receive phase-encoded signals reflected from objects away from the radar according to a time division multiple access (TDMA) modulation scheme.

[0051] In one example where antenna elements with similar index numbers across multiple sets of antenna elements form corresponding antenna subarrays, and antenna elements in each antenna subarray transmit simultaneously during different time slots using different slow phase codes, antenna elements with index number 1 form a first antenna subarray 1106a, antenna elements with index number 2 form a second antenna subarray 1106b, and antenna elements with index number 3 form a third antenna subarray 1106c.

[0052] The four transmit channels TX1, TX2, TX3, and TX4 are connected to the inputs of first, second, third, and fourth Single Pole 3 Throw (SP3T) switches 1104a, 1104b, 1104c, and 1104d (throws labeled A, B, and C). Thus, the number of switches 1104a, 1104b, 1104c, and 1104d is the same as the number of transmit channels on the radar transceiver chip 1102.

[0053] 11B, first, second, and third antenna subarrays 1106a, 1106b, and 1106c are connected to transmit channels TX1, TX2, TX3, and TX4 via switches 1104a, 1104b, 1104c, and 1104d, respectively. Each antenna subarray includes the same number of antenna elements as the number of transmit channels, and the number of antenna subarrays is the same as the number of switch outputs. In one embodiment of the present invention, the antenna elements of a given antenna subarray are configured to transmit simultaneously.

[0054] Therefore, when the number of outputs of each switch is 3, a total of three antenna subarrays are used. Also, when the number of transmission channels is 4, the total number of antennas in each subarray is 4.

[0055] In an embodiment of the present invention, at any given time, the output / throw positions of the switches are connected to antenna elements of different subarrays 1106a, 1106b, 1106c. For example, throw position A of switches 1104a through 1104d is connected to antennas of the first subarray 1106a, throw position B of switches 1104a through 1104d is connected to antennas of the second subarray 1106b, and throw position C of switches 1104a through 1104d is connected to antennas of the third subarray 1106c.

[0056] Thus, the four transmit channels TX1, TX2, TX3, and TX4 transmit simultaneously, applying phase encoding. All transmitters are powered on at the same time and transmit during the first chirp. All linked switch positions are activated simultaneously for each transmit channel. All switch positions simultaneously move to the next position while maintaining the same DDMA encoding. This step is repeated until all switch positions have been completed. This entire process is repeated continuously during operation. Each switch position is activated in a predefined common sequence to enable transmission of the phase-encoded signal from each transmitter according to the TDMA modulation scheme.

[0057] In one example, all three switches' A positions are activated simultaneously to allow the antennas in the first antenna subarray 1106a to transmit simultaneously, all B positions are activated simultaneously to allow the antennas in the second antenna subarray 1106b to transmit simultaneously, and so on. Thus, the four transmit channels transmit according to a DDMA modulation scheme, and the three switches A, B, and C in each transmit channel transmit according to a TDMA modulation scheme. Those skilled in the art will appreciate that the switches in the above example may be added as an alternative to the receive channels, or to both the transmit and receive channels.

[0058] Those skilled in the art will appreciate that the SP-TDMA scheme can be applied to similar systems with different numbers of transmit and receive channels (such as a two-radar transceiver chip system) or different switch configurations, e.g., SP2T, SP4T, etc. In the MIMO radar configuration 1100, the number of transmit channels can be increased from four chip transmit outputs to twelve by switching three outputs of the SP3T switch. In practical terms, such an arrangement can be used to scale from a 4xTX channel, 4xRX channel system using TDM modulation (using only the radar transceiver chip) to a larger 12xTX channel, 4xRX channel system using TDM modulation (using the radar transceiver chip and the SP3T switch). However, this increases the number of distinct velocity hypotheses that must be solved, and also eliminates the transmission gain / system range that could be achieved by simultaneously transmitting on multiple channels using a modulation scheme such as DDMA. Here, DDMA phase encoding is implemented (using a form of TDMA) across three time slots, with up to four transmit channels (i.e., four channels on the radar transceiver chip) simultaneously transmitting at once. Furthermore, each transmit channel TX1, TX2, TX3, TX4 is differently phase coded according to DDMA or other applicable phase modulation schemes, for example BPM.

[0059] Furthermore, it will be appreciated by those skilled in the art that the SP-TDMA scheme can also be implemented on the receiver side.An exemplary MIMO system implementing the SP-TDMA scheme on the receiver side is shown with reference to Figure 11C.

[0060] FIG. 12 shows an example of the first 6 chirps of an SP-TDMA transmit modulation scheme with DDMA phase coding in a MIMO radar configuration 1100.

[0061] In the case of the MIMO radar configuration 1100, the phase codes and time slots are assigned as shown in FIG. throw The switch shifts the phase of the DDMA modulation scheme N phases before the next phase shift is implemented. throwRepeated N times, simultaneously TX The switch is N throw The throw position can be switched through throw Repeat the DDMA modulation over the time slots. In this way, a hybrid of DDMA and time division multiple access is implemented, whereby for each set of time slots (i.e., N throw The DDMA scheme (corresponding to each set of switch throws) is demodulated by the cyclic sum approach detailed above. Alternatively, the Chinese Remainder Theorem (CRT) or other suitable means can be used to recover the velocity and map the Doppler subbinaries to the correct transmission channel for demodulating the signal. By demodulating the DDMA, N throw The data received in each of the set of time slots can be mapped back to the original transmitter, and then the complete virtual array can be mapped by considering which transmit antenna element is used in which time slot. The phase encoding chosen here is similar to that used in DDMA, but it will be recognized that other types of phase encoding can be used without limitation. Similarly, all of the switch positions shown in FIG. 11 are shown in the same position, cycling repeatedly from A to B to C in all cases for ease of illustration. However, this need not be the case, as long as the position cycled through each switch is constant.

[0062] In the SP-TDMA transmission modulation scheme, the maximum clear rate is v max / N throw while the number of transmission channels is limited to N throw The angular accuracy, gain, and range of the radar system increase proportionally. max DSP methods that may be used to recover include, but are not limited to, the coherent peak method, where azimuth and / or elevation information is taken into account, or CRT, where subframes with different coprime-chirp repetition intervals may be used to recover the true velocity.

[0063] To work out which of the possible velocities detected is correct, i.e., the "true velocity," this embodiment uses a coherent gain method. In this example, the virtual array is assumed to be a "filled" array of uniformly spaced elements of d=λ / 2 (λ is the wavelength of the operating frequency), and coherent azimuth FFT processing can be used (though those skilled in the art will understand that azimuth elevation processing or the Chinese Remainder Theorem, or an "unfilled" array with non-uniform element spacing, could be used). For example, referring to Figures 11 and 12, when using an SP3T switch, there are three time slots, ±v max,ext For velocities within, there are three possible Doppler phase values ​​for the ambiguous phase measurement, v max,ext =v max *N throw These phase values ​​are shown graphically in Figure 13 using a phasor diagram and correspond to the following values: φ v,unwrapped ∈{φ v1 ,φ v2 ,φ v3}, in this case: φ v1 =φ v,meas φ v2 =φ v,meas +2π φ v3 =φ v,meas -2π At this time, φ v,meas is the measured Doppler phase, and φ v1 , φ v2 and φ v3are the three possible phase values, with and without phase wrapping. For further illustration, Figure 14 shows the wavefront arriving at a subsection (corresponding to only TX1) of the hypothetical array of the radar system shown in Figure 11. Here, unless the detected object is in the boresight (directly in front of the radar), the signal wavefront takes time to reach subsequent antenna elements, and that delay is proportional to the angle of incidence θ of the wavefront from the boresight. This is represented by the added phase Δφ. If the detected object is moving with a relative velocity v to the radar, then in a subsequent time slot, when the switch is moved from position A to position B (and back to position C), there will be a further change in phase Δφ proportional to the object's velocity, even though a switching delay time must be taken into account. v In this way, a complete virtual array can be formed, and three possible phase "hypothesis" values, φ, are calculated across the virtual array. v1 , φ v2 , φ v3 and perform an azimuth FFT for each, and the FFT with the highest peak is φ v corresponds to the true, correct, "unwrapped" value of , since correction with the true phased value generally gives the most coherent FFT peak.

[0064] In one embodiment of the present invention, the virtual antenna array configuration may be other than a simple "fill" array along the azimuth plane. A fill array with overlapping elements includes elements that occupy the same positions in the virtual array in subsequent time slots. Phase comparison of overlapping elements provides an unambiguous measurement of Doppler phase and requires less complex processing than coherent peak or CRT methods, facilitating alternative (or complementary) velocity enhancement methods for time-multiplexed received signals, allowing faster object detection.

[0065] In one embodiment of the present invention, a rectangular virtual array is provided with both azimuth and elevation components, and angular resolution of object positions is achieved in both the azimuth and elevation directions.

[0066] In one embodiment of the present invention, a non-uniform array (e.g., a sparse array) is provided. The array is either linear in one plane or rectangular in shape occurring in both the azimuth and elevation planes. A larger virtual array aperture may be created, resulting in higher angular resolution for a comparable number of chips.

[0067] The coherent peak method is further illustrated in Figure 15, where v<-v at an azimuth angle of -30°. max A radar target moving with a velocity v of 0.0 ...

number

[0068] 16 shows a DSP chain for demodulating signals received by multiple receive channels of a MIMO radar 1100 according to a first embodiment of the present invention. In step 1602, ADC data is generated by sampling the IF signal of each receive channel to generate a three-dimensional data array of size number of samples per chirp x number of chirps x number of Rx channels. In step 1604, N throw Time division is performed by separating the sampled mixer data according to time slots, resulting in the following data structure:

number

[0069] In step 1606, a range-Doppler 2D FFT is performed along the first two dimensions of the data structure obtained in step 1604. The total number of 2D FFTs performed is N Rx ×N throw and the output is size samples per chirp × number of chirps / N throw ×N throw× number of Rx channels. In step 1608, DDMA demodulation is performed for all range-Doppler subbinaries, N Rx ×N throw A separate process is performed for each range-Doppler profile, mapping each range-Doppler subbinary to its corresponding TX channel and recovering the true ambiguous measured velocity for that range-Doppler subbinary. In step 1610, coherent peak / azimuth / velocity processing is performed using the associated elements of the virtual array to recover the true unambiguous extended velocity for each range-Doppler subbinary. In step 1612, a CFAR is calculated, and optionally a local maximum algorithm can be performed, to determine which range-Doppler azimuth binaries contain valid target detections. In step 1614, a list of detections is generated. In step 1616, elevation processing is performed on only those range-Doppler binaries flagged as detections during step 1612. In step 1618, a 4D point cloud (showing the 3D location along with the velocity of all targets detected in the frame) is calculated and displayed. Note that elevation detections can also be used for velocity extension; therefore, step 1616 becomes azimuth processing in this case. If both azimuth and elevation detection are used for velocity expansion, steps 1602 through 1614 are performed simultaneously for both azimuth and elevation, and step 1616 is deleted.

[0070] 17 shows a DSP chain for demodulating signals received by multiple receive channels of a MIMO radar 1100 according to a second embodiment of the present invention. In step 1702, ADC data is generated by sampling the IF signal of each receive channel to generate a three-dimensional data array of size number of samples per chirp x number of chirps x number of Rx channels. In step 1704, N throw Time division is performed by separating the sampled mixer data according to time slots, resulting in the following data structure:

number

[0071] In step 1706, a range-Doppler 2D FFT is performed along the first two dimensions of the data structure obtained in step 1704. The total number of 2D FFTs performed is N Rx ×N throw and the output is size samples per chirp × number of chirps / N throw ×N throw × the number of Rx channels. In step 1708, a non-coherent sum is performed, in which the absolute magnitude value data for each range-Doppler subbinary is taken and summed over all N Rx ×N throw Summed over channels, size per chirp number of samples × number of chirps / N throw In step 1710, a CFAR is calculated using the non-coherently summed data, and an optional local maximum algorithm can be performed to determine which range-Doppler binaries contain target detections. In step 1712, DDMA demodulation is performed on only the binaries containing detections, and the range-Doppler subbinaries corresponding to each detection are mapped to corresponding transmit channels to construct a virtual array. In step 1714, coherent peak / azimuth / velocity processing is performed using the virtual array on only the range-Doppler binaries containing detections. In step 1716, elevation processing is performed. In step 1718, a 4D point cloud (showing the 3D location along with velocity of all targets detected in the frame) is calculated and displayed.

[0072] In the embodiment of FIG. 17, the CFAR calculation is performed early in the DSP chain, reducing the processing requirements for demodulation, angular rate estimation, and velocity estimation, thus improving the speed of calculations, but potentially reducing the overall number of detections made.

[0073] 18 shows a DSP chain for demodulating signals received by multiple receive channels of a MIMO radar 1100 according to a third embodiment of the present invention. In step 1802, ADC data is generated by sampling the IF signal of each receive channel to generate a three-dimensional data array of size number of samples per chirp x number of chirps x number of Rx channels. In step 1804, N throw Time division is performed by separating the sampled mixer data according to time slots. In steps 1806 and 1808, the ADC data is split into subframes 1 and 2 transmitted at different chirp repetition rates. The output is two 4-dimensional data arrays. In step 1810, a range-Doppler 2D FFT is performed along the first two dimensions of the data arrays obtained in steps 1806 and 1808. The total number of 2D FFTs performed is N Rx ×N throw ×N sub-frame and the output is size samples per chirp × number of chirps / N throw × Number of Rx channels × N throw In step 1812, a non-coherent sum is performed in which, for each subframe, absolute magnitude value data is taken for each range-Doppler subbinary and summed over all N Rx ×N throw The data are summed over the channels. In step 1814, the CFAR is calculated and a local maximum algorithm can be performed to determine which range-Doppler angle binaries contain target detections. In step 1816, DDMA demodulation is performed on only those binaries that contain detections. In step 1818, velocity expansion is performed using the Chinese Remainder Theorem (CRT) using the two sets of DDMA demodulated data arrays corresponding to subframes 1 and 2. In step 1820, angle processing is performed, and in step 1822, a 4D point cloud (showing the 3D location along with the velocity of all targets detected in the frame) is calculated and displayed.

[0074] In the embodiment of FIG. 18, sub - frames with different co - prime chirp repetition intervals are implemented so that the Chinese Remainder Theorem can be used for speed expansion instead of the coherent peak method.

[0075] · To perform SP - TDMA modulation, by using a switch with a polarization ratio of S and an insertion loss of L, the same number and type of transceiver chips are used, and there are several advantages as follows compared with TDM or DDMA MIMO without using a switch. · Compared with TDM, since the transmission gain increases by S / L times (assuming L < S), the signal - to - noise ratio is improved and the maximum reach distance of the radar increases. Furthermore, the number of elements in the virtual array aperture increases by S times, leading to a further increase in distance due to the increase in array gain and an improvement in angular resolution due to the increase in array aperture size. · Compared with DDMA MIMO, the transmitted EIRP decreases by 1 / L times, but the virtual array aperture size increases by S times, leading to an increase in array gain and an improvement in angular resolution. Even in this case, if S > L, the distance can be extended to provide an increase in array gain larger than the increase in transmitter loss.

[0076] Furthermore, by performing SP - TDMA modulation using a switch in a MIMO radar, compared with the same MIMO radar that performs pure TDM modulation using a switch, the SP - TDMA MIMO system increases the EIRP by transmitting simultaneously on multiple channels, and as a result, the reach distance increases. Also, in the SP - TDMA system, since the number of transmission time slots used is small, the original unambiguous maximum speed is high, so there are fewer speed hypotheses to be solved to further expand the maximum speed. As a result, the complexity of the DSP method is reduced and the time efficiency is high.

[0077] While the preceding description uses a switch, it should be noted that the present invention is not limited to systems that use switches. This configuration is used herein solely to increase the number of MIMO channels without increasing the number of radar transceiver chips. This technique can also be applied to systems using multiple radar transceiver chips (master-slave configuration) to increase the number of MIMO channels without using a switch. Assuming the same type of radar transceiver chip, this technique can be used to extend the range of a radar system over a switched SP-TDMA system. This combines eliminating switching losses, increasing the number of simultaneously transmitting transceiver channels to achieve higher transmit EIRP, and increasing the number of transceiver receive channels to provide a larger virtual array and higher system gain. This latter feature also has the benefit of increasing radar resolution. This multiple transceiver technique offers the following advantages over either TDM or DDMA systems (using the same radar transceiver chip): Compared to TDM, increased coherent gain from transmitting on multiple channels simultaneously results in greater range. Simultaneous transmission on multiple channels also results in a higher inherent maximum unambiguous rate, shorter chirp repetition times for the same virtual array transmit channel, and less computationally intensive rate-enhancing techniques, such as coherent peaking, require fewer hypotheses to solve, potentially resulting in greater robustness in low SNR and multipath situations. Compared to DDMA, the number of simultaneously transmitting channels is smaller, resulting in a slightly shorter range. However, the aforementioned SP-TDMA system has fewer subbands compared to DDMA systems, which reduces the probability that multiple targets will appear in the same subbinary of different sets of subbands, thereby reducing the risk of one or more targets being hidden and undetected. Furthermore, with DDMA, the number of simultaneously transmitting channels is increased, resulting in finer phase steps, making chirp-based phase encoding more susceptible to the chip-accuracy capabilities of phase code generation. Therefore, the present invention increases the probability of detecting all targets within the radar's field of view. This feature is particularly useful in systems with a large number of TX channels, as it leads to an increased number of subbands in DDMA systems. Therefore, a DDMA system with NTX=12 requires a requirement of 16 subbands (assuming Nsub=z*2x, where z is an integer, typically 1 or 3, but not limited to this), which increases the probability of missing these detections. However, by using the SP-TDMA modulation scheme, only five sub-bands (including one empty sub-band for demodulation) are required using three time slots, reducing the risk of obscuring one or more targets. In one embodiment of the present invention, the CRT method can be used for demodulation to find the correct velocity, rather than the empty sub-band-based DDMA technique and coherent peak method described above.

[0078] It should be noted that in the SP-TDMA embodiment using a switch described above, the switch is placed between the transmit channels and the transmit antenna elements of the radar transceiver chip, but the same result of increasing the number of MIMO transceiver channels beyond those available from the transceiver alone can be achieved by placing the switch instead between the receive antenna elements and the receive channels of the radar transceiver chip. In this case, the number of transmit antennas is equal to the number of transmit channels on the radar transceiver chip, and phase modulation is performed on the transmit signals as in the previous example, but the number of receive antennas is N RX From N throw *N RXand switching is performed on the receive antennas. This increases the MIMO virtual array size, improving the range and angular resolution performance of the system as described in the previous switching example. Otherwise, the DSP chain is the same for both embodiments, with the received data separated according to time slots before the phase code demodulation, CFAR detection, and angle processing steps.

[0079] By placing switches between the transmit channels of the radar transceiver and the transmit antenna, and between the receive antenna and the receive channels of the radar transceiver, it is possible to further increase the number of channels of the SP-TDMA MIMO transceiver without increasing the number of radar transceiver chips. In this case, the transmit phase modulation and switching scheme can be N throw Time slot to N throw 2 The DSP chains are otherwise the same, but the received data is separated according to time slots before the phase code demodulation, CFAR detection and angle processing steps. This method reduces the radar's inherent maximum unambiguity rate compared to switching only transmit or receive, but increases angular resolution and range.

[0080] In this specification, the terms "comprise", "comprises", "comprised", "comprising" or variations thereof and "include", "includes", "included", "including" or variations thereof are considered interchangeable and all are to be given the broadest possible interpretation and vice versa.

[0081] The invention is not limited to the embodiments described herein, which may vary both in structure and detail.

Claims

1. a plurality of millimeter wave transmitters configured to simultaneously transmit phase encoded signals according to a phase modulation scheme; a set of antenna elements corresponding to each millimeter wave transmitter, forming a plurality of sets of antenna elements for the corresponding plurality of millimeter wave transmitters, each set of antenna elements sequentially transmitting the phase-encoded signals received from its respective transmitter in different mutually orthogonal time slots using time division multiplexing, such that each antenna of the set transmits a similar phase-encoded signal to other antennas of the set in a different time slot, and each antenna element is assigned an index number within each corresponding set, such that antenna elements with similar index numbers across the plurality of sets of antenna elements simultaneously transmit during different time slots using different slow phase codes; a plurality of millimeter-wave receivers coupled to an antenna, the millimeter-wave receivers receiving the phase-encoded signals reflected from objects remote from the radar in accordance with a Time Division Multiple Access (TDMA) modulation scheme; A millimeter-wave digital automotive radar equipped with

2. 2. The millimeter-wave digital automotive radar of claim 1, wherein the antenna elements of like index numbers across multiple sets of the antenna elements form corresponding antenna subarrays, and the antenna elements of each antenna subarray transmit simultaneously during different time slots using different slow phase codes.

3. 3. The millimeter-wave automotive radar of claim 2, wherein the plurality of antenna subarrays transmit through at least one antenna element the phase-encoded signal received from a respective millimeter-wave transmitter during each orthogonal time slot, such that the antenna elements of the plurality of subarrays transmitting simultaneously during a first time instance continue to transmit simultaneously for different subsequent time instances using different slow phase codes according to a predefined scheme.

4. 10. The millimeter-wave automotive radar according to any of the preceding claims, wherein the multiple millimeter-wave transmitters are implemented via multiple transmission channels of at least one transceiver chip.

5. 5. The millimeter-wave automotive radar according to claim 2, wherein the plurality of antenna subarrays are connected to the plurality of millimeter-wave transmitters via respective plurality of single-input multiple-output millimeter-wave switches, an input of each switch being connected to a respective millimeter-wave transmitter, and a plurality of outputs of each switch being connected to antennas of different antenna subarrays.

6. 6. The millimeter-wave automotive radar according to claim 2, wherein the number of antenna subarrays is the same as the number of outputs of each switch, and the number of antenna elements in each subarray is the same as the number of millimeter-wave transmitters.

7. 7. The millimeter-wave automotive radar according to claim 2, wherein when the number of switches is four and the number of throw positions of each switch is three, the first throw position of each switch is connected to a respective antenna of the first subarray, the second throw position of each switch is connected to a respective antenna of the second subarray, and the third throw position of each switch is connected to a respective antenna of the third subarray, and the same throw positions of each switch are activated simultaneously to enable simultaneous transmission from the antennas of the respective antenna subarrays.

8. 8. The millimeter-wave automotive radar according to claim 2, wherein the throw positions of the switches are activated in a predefined common sequence to enable transmission of phase-encoded signals from the respective transmitters in mutually orthogonal time slots according to a TDMA modulation scheme.

9. 9. The millimeter-wave automotive radar according to claim 2, wherein the antenna elements of either the millimeter-wave transmitter or the millimeter-wave receiver, or both, are arranged at non-equidistant intervals.

10. 10. The millimeter-wave automotive radar according to claim 2, wherein the phase modulation is selected from the group consisting of Doppler Division Multiple Access (DDMA) and Binary Phase Modulation (BPM).

11. 10. The millimeter-wave automotive radar according to claim 9, further comprising a plurality of millimeter-wave transceivers that receive and demodulate the phase-encoded signals transmitted by the plurality of millimeter-wave transmitters based on a cyclic sum of the subbands, identify the millimeter-wave transmitter of each received signal, and then identify a transmitting antenna element of the millimeter-wave transmitter based on an arrival time of the demodulated signal.

12. 12. The millimeter-wave automotive radar according to claim 2, further comprising: a further plurality of antenna subarrays coupled to a respective plurality of millimeter-wave receivers configured to receive signals and simultaneously apply phase codes to the received signals according to a phase modulation scheme, the phase codes being in mutually orthogonal time slots separate from the other subarrays using time division multiplexing, and the number of millimeter-wave receivers being the same as the number of antenna elements in each antenna subarray.

13. A method of operating a millimeter wave automotive radar, comprising: enabling a plurality of millimeter wave transmitters to simultaneously transmit phase encoded signals according to a phase modulation scheme; providing a set of antenna elements for each millimeter wave transmitter to form a plurality of sets of antenna elements for a corresponding plurality of millimeter wave transmitters; enabling the antenna elements of each set to sequentially transmit phase-encoded signals received from their respective transmitters in separate, mutually orthogonal time slots using time division multiplexing, with each antenna of the set transmitting a similar phase-encoded signal in a different time slot to other antennas of the set, each antenna element being assigned an index number within each corresponding set, allowing antenna elements of similar index numbers across multiple sets of antenna elements to simultaneously transmit during different time slots using different slow phase codes; A method comprising:

Images