Time-of-flight sensor and sensing method

Abstract

A time-of-flight sensor includes a time-of-flight (TOF) processor having a digital TOF port, a digital input port, and a digital output port, the TOF processor including a phase detector including a cyclic rotation de-multiplexer (DEMUX), a first summer coupled to a first DEMUX output, a second summer coupled to a second DEMUX output, a third summer coupled to a third DEMUX output, a fourth summer coupled to a fourth DEMUX output, and a phase estimator coupled to outputs of the first, second, third, and fourth summers and having a phase estimate output; a driver having a digital driver port coupled to the digital TOF port and a driver output port; and an analog-to-digital converter (ADC) having an output port coupled to the digital input port of the digital TOF processor.

Description

[0001] This application is a divisional application of Chinese invention patent application No. 201910699841.9, filed on July 31, 2019, entitled "Time-of-Flight Sensor and Sensing Method".

[0002] Cross-reference to related applications

[0003] This application claims the benefit of co-pending U.S. Provisional Patent Application Serial No. 62 / 712,952, filed on July 31, 2018, which is incorporated herein by reference. Background Technology

[0004] Time of flight (TOF) is a property of an object, particle, or wave (e.g., a sound wave or an electronic wave) that relates to how long it takes for the object, particle, or wave to travel through a medium. TOF technology can be used for a variety of purposes, including ranging and 3D imaging.

[0005] 3D Time-of-Flight (TOF) technology has revolutionized the machine vision industry by providing 3D imaging through the use of low-cost CMOS pixel arrays along with active modulated light sources. Its compact design, ease of use, high accuracy, and high frame rate make TOF cameras an attractive solution for a wide range of applications, such as computer vision, drones, and robotics.

[0006] 3D time-of-flight (TOF) cameras work by illuminating a scene with a modulated light source and observing the reflected light. The phase shift between illumination and reflection is measured and converted into distance. Typically, the illumination comes from a solid-state laser or LED operating in the near-infrared range (~850 nm), invisible to the human eye. An imaging sensor, designed to respond to the same spectrum, receives the light and converts the photon energy into an electric current. Note that the light entering the sensor has both an ambient component and a reflected component. Distance (depth) information is embedded only in the reflected component. Therefore, a high ambient component reduces the signal-to-noise ratio (SNR).

[0007] To detect the phase shift between illumination and reflection, the light source is pulsed or modulated using a continuous wave (CW) source, typically a sine or square wave. Square wave modulation is more common because it is easier to implement using digital circuitry. However, sine waves may result in less distortion.

[0008] The pulse method is simple and direct. The light source illuminates for a short period (Δt), and at each pixel, the reflected energy is sampled in parallel using two out-of-phase windows C1 and C2 with the same Δt. The accumulated charges Q1 and Q2 during these sampling periods are measured and used to calculate the distance. In contrast, the CW method acquires multiple samples for each measurement, typically at least four samples per cycle of the modulation frequency. Using this technique, the phase angle φ between illumination and reflection, as well as the distance d from the object, can be calculated.

[0009] CW measurements are based on the fact that the phase (which cycles once every 2π (“phase wrapping”)) means that the distance will also have aliasing. The distance where aliasing occurs is called the ambiguous distance (damb). Due to distance wrapping, the dimb is also the maximum measurable distance. As mentioned earlier, if you want to extend the measurable distance, you must reduce the modulation frequency, which reduces the accuracy of the distance measurement.

[0010] There are commercially available Time-of-Flight (TOF) systems that utilize the aforementioned processing. However, existing TOF systems are not integrated solutions; they typically involve several common components programmed with firmware and software to perform TOF measurements, thereby deriving depth and 3D data. Such systems are often bulky and expensive.

[0011] Furthermore, due to the limitations of the light source and sampling of the detector in previous TOF sensors, square wave signals were typically used, which led to greater ambiguity and depth error.

[0012] Due to phase entanglement, there is a trade-off between distance measurement range and measurement accuracy. Phase entanglement refers to the ambiguous distance represented by signals reflected from an object. High-precision methods require high-frequency modulation, which limits the range of objects that can be detected.

[0013] When performing Time-of-Flight (TOF) measurements based on reflected waves, there may be interference from other light sources with periodic modulation. These other light sources may be unrelated to the TOF sensor, or they may be other TOF sensors operating in the same physical space. Other TOF sensors operating in the same physical space are particularly problematic because the detected light may be direct light from those other TOF sensors.

[0014] These and other limitations of the prior art will become apparent to those skilled in the art after reading the following description and studying the accompanying drawings. Summary of the Invention

[0015] Several exemplary embodiments will now be described with reference to the accompanying drawings, wherein similar reference numerals are provided for similar parts. These exemplary embodiments are intended to be illustrative and not to limit the invention. The drawings include the following figures:

[0016] Figure 1 This is a block diagram of a first example embodiment of a time-of-flight (TOF) sensor;

[0017] Figure 2 This is a block diagram of a second example embodiment of a TOF sensor;

[0018] Figure 3 This is a flowchart of a method for operating a TOF sensor;

[0019] Figure 4 This is a schematic diagram illustrating the operation of the TOF sensor at the first frequency f1 and the second frequency f2;

[0020] Figure 5 This is a block diagram of a third example embodiment of a TOF sensor;

[0021] Figure 6 This is a block diagram of a Maximum Length Sequence (MLS) generator;

[0022] Figure 7 It is a flowchart of a method for distinguishing the phase-shifted reflected waveform of a transmitted waveform from other waveforms;

[0023] Figure 8A This is a schematic diagram showing an MLS signal;

[0024] Figure 8B This is a schematic diagram showing an MLS-modulated cosine wave; and

[0025] Figure 9A and 9B This is a schematic diagram of a cosine wave pulse modulated by MLS. Detailed Implementation

[0026] exist Figure 1 In this embodiment, the Time-of-Flight (TOF) sensor 10, illustrated by way of example and not limitation, includes a TOF processor 12, a driver 14, an analog-to-digital converter (ADC) 16, and a signal conditioner 18. In some embodiments, the TOF sensor 10 is coupled to or further includes an interface 20, a waveform transmitter 22, and a waveform receiver 24. For example, the interface 20 may be an external light-emitting diode (LED) or laser interface, the waveform transmitter may be an LED or a laser, and the waveform receiver 24 may be a photodetector. In an example embodiment, the TOF sensor 10 is implemented as an integrated circuit 26. Other components, such as the interface 20, the waveform transmitter 22, and / or the waveform receiver 24, may also be implemented as part of the integrated circuit 26.

[0027] The TOF processor 12 includes a digital TOF port 28, a digital input port 30, and a digital output port 32. The driver 14 includes a digital driver port 34 coupled to the digital TOF port 28 of the TOF processor 12. The ADC 16 has an output port 36 coupled to the digital input port of the digital TOF processor. The signal conditioner 18 couples a waveform receiver 24 to the input of the ADC 16. The interface 20 couples the driver 14 to the waveform transmitter 22. It should be noted that the TOF processor 12 can act as a correlator with a correlated waveform (e.g., derived from a reflected waveform) as an input.

[0028] It should be appreciated that, in this example embodiment, integrated circuit 26 provides control signal 40 to drive (or modulate) light source 22. Control signal 40 is preferably coupled to additional driving electronics (e.g., FET or MOSFET) of driver 14 to amplify the power to the light source (LED, laser diode, etc.), or it may independently provide power to drive light source 22.

[0029] Control signal 40 generates a clock that can be used to determine the phase difference between the modulated waveform emitted by light source 22 and the reflected waveform received by photodetector 24. As a non-limiting example, photodetector 24 may be a single-photon avalanche diode (SPAD), silicon photomultiplier tube (SiPM), silicon photodiode, III-V group photodiode, photoconductor, etc. In some embodiments, photodetector 24 may be an array of photodetectors capable of generating images. Photodetector 24 may be associated with a lens or lens barrel (not shown) to enhance light collection and facilitate the generation of images of a scene or object / target, similar to a conventional camera system. The photodiode is read out individually by electronics in the sensor system. The readout may include a low-noise amplifier and a filter. The readout may also provide the necessary reverse bias voltage. After gain and amplification and filtering, the signal is sampled and digitized by a fast ADC, and the data is sent to circuitry for calculation. Figure 2 A block diagram of the controller / microcontroller that performs this calculation is shown, and the photodiode and readout are shown again for easier reference.

[0030] exist Figure 2 In this example, the TOF sensor 10', illustrated by way of example and not limitation, includes a TOF processor 12', an ADC 16, and a signal conditioner 18', wherein similar reference numerals refer to similar components. In this non-limiting example, the TOF processor 12' includes a demultiplexer (DMUX) 44, a counter 46, a first summer 48, a second summer 50, a third summer 52, a fourth summer 54, a first clock control register 56, a second clock control register 58, a third clock control register 60, a fourth clock control register 62, and a CORDIC rotator 64. The summers 48-54, the clock control registers 56-62, and the CORDIC rotator 64 collectively include a phase estimator 66.

[0031] In this example, DEMUX 44 is 1:2 n(1:4) Demultiplexer, where n = 2. Therefore, DEMUX 44 is an n-bit (2-bit) controlled demultiplexer comprising two control ports S0 and S1, a first DEMUX output Y0, a second DEMUX output Y1, a third DEMUX output Y2, and a fourth DEMUX output Y3. In this example, counter 46 is a 2-bit counter with clock input CLK and outputs S0 and S1. ADC 16 is similar. Figure 1 Operating like a similar component, the signal conditioner 18' includes an operational amplifier (OPAMP) 68, which has a feedback resistor 70 and a filter 72, such as an anti-aliasing filter (AAF).

[0032] As described above, the phase estimator 66 includes summers 48-54, clock control registers 56-62, and a CORDIC rotator 64. In this non-limiting example, the first DEMUX output Y0 is coupled to the input of the first summer 48, the second DEMUX output Y1 is coupled to the input of the second summer 50, the third DEMUX output Y2 is coupled to the input of the third summer 52, and the fourth DEMUX output Y3 is coupled to the input of the fourth summer 54. In this example, the first summer 48 sums the positive real part (∑+REAL) of the demultiplexed signal, the second summer 50 sums the negative imaginary part (∑-IMAG) of the demultiplexed signal, the third summer 52 sums the negative real part (∑-REAL) of the demultiplexed signal, and the fourth summer 54 sums the positive imaginary part (∑-IMAG) of the demultiplexed signal. Registers 56-62 latch the summation value of summers 48-54 for each clock cycle of the system clock CLK.

[0033] It should be noted that the CORDIC rotator 64 is merely one example of a phase estimator. Other examples of phase estimators include polynomial approximations of the arctangent function, or lookup table approximations of the arctangent function followed by linear interpolation. For example, see Ukil, Abhisek & Shah, Vishal & Deck, Bernhard. (2011), Fast computation of arctangent functions for embedded applications: A comparative analysis. 10.1109 / ISIE.2011.5984330, and P. Markstein, "A fast-start method for computing the inverse tangent," 17th IEEE Symposium on Computer Arithmetic (ARITH'05), 2005, pp.266-271, all of which are incorporated herein by reference. In this non-limiting example, the CORDIC rotator 64 digitally implements Volder's algorithm as a CORDIC (Coordinate Rotation Digital Computer) algorithm to efficiently compute hyperbolic and trigonometric functions. CORDIC is an example of a bitwise algorithm that iteratively converges to the answer by processing one bit (or digit) at a time. In the absence of a hardware multiplier, CORDIC is closely related to methods known as pseudo-multiplication and pseudo-partitioning. The only operations required to converge to the answer are addition, subtraction, shifting, and table lookup. Therefore, the CORDIC algorithm belongs to the category of shift-and-add algorithms. The design and fabrication of CORDIC rotators are well known to those skilled in the art.

[0034] The example CORDIC algorithm implemented by the CORDIC rotator 64 can be explained as follows. Assume the received signal current contains a known frequency but a phase... Unknown sine component.

[0035]

[0036] Where I(t) is an analog signal from the amplifier and filter with an A0 DC level, and m is the ratio of the AC signal amplitude to the DC level.

[0037] If there is a reference signal (e.g., a control signal to the LED, or an electrical signal from the oscillator modulating the light source), then

[0038] R(t) = exp(2πif m t)

[0039] The phase can be calculated as:

[0040]

[0041] It is understandable that a higher value of K represents a longer integration time, which improves the signal-to-noise ratio (SNR) and computation in the presence of noise. The accuracy; N(t), assuming the ADC operates at a frequency f s =4f m If the incoming signal I(t) from the detector is sampled, then

[0042]

[0043] Wherein, the reference signal is

[0044]

[0045] The phase shift of the reflected signal relative to the emitted (reference) signal is as follows:

[0046]

[0047] It should be noted that the computation is decomposed into four discrete sums to eliminate the need for multiplication in the correlation process. Therefore, the mathematics behind the algorithm for efficiently performing complex correlations is as follows:

[0048]

[0049] or

[0050]

[0051] It will become apparent that by choosing f s =4f m This eliminates the need for any multiplication in the CORDIC rotator 64. The arithmetic complexity is reduced to one real number accumulation per pixel per sample and one 4-quadrant arctangent per pixel per relevant interval (as a non-limiting example), which can be implemented using CORDIC.

[0052] The maximum depth at which a distance can be detected unambiguously can be written as Where c is the speed of light, f m It is the modulation frequency. In d max The phase then repeats itself in a phenomenon known as “phase entanglement,” potentially leading to ambiguous values.

[0053] In this example, the standard deviation of the constraint estimation error is as follows:

[0054]

[0055] Where N is the number of samples read by the ADC within each computation cycle or integration time (e.g., the number of samples summed together for calculating the phase as described above). SNR is the signal-to-noise ratio for one of those readings, and f is the modulation frequency. It will be noted that the higher the frequency, the lower the error in estimating the time of flight. However, due to phase entanglement, d max It also decreases.

[0056] Figure 3 Flowchart 68 shows an example procedure for estimating phase using complex correlation and CORDIC derotation. Procedure 68 begins with operation 70, and in operation 72, f... m The amplitude modulation (AM) is applied to power the transmitter (e.g., LED 24'). Next, in operation 74, the variables are initialized. In operation 76, before reading samples s from the ADC, 1 / (4*f) is applied. m After a waiting period of 1 second, the sample s is increased to Q(k), and k is incremented by 1. Next, in operation 78, it is determined whether k = 5. If not, operation 76 is repeated. If k is indeed equal to 5 (e.g., operation 76 has been performed 4 times in the i-th iteration), operation 84 determines whether variable i is equal to variable N. If not, variable k is set to 1, and variable i is incremented by one before returning process control to operation 76. After i equals N, operation 84 calculates the estimated phase using Q1, Q2, Q3, and Q4, and outputs the estimated phase. Next, operation 86 determines whether to continue process 68. If to continue, variable k is reset to 1, and process control returns to operation 76. If process 68 is not to continue, the device is powered off in operation 90, and process 68 ends in operation 92.

[0057] Figure 4 This is a schematic diagram illustrating an example embodiment using two modulation frequencies, for example, f1 = 10 MHz and f2 = 1 MHz. For the lower frequency modulation f2, the phase position can be calculated first, and an estimate of the Time-of-Flight (TOF) can be obtained. From the higher frequency modulation f1, a more accurate TOF estimate can be obtained within a specific range determined by the first frequency. It can be seen that at the higher frequency f1, the TOF[X] angle... Repeated every 2π, but with good resolution, while for lower frequencies f2, the TOF[X] angle... It is not repeated until it is aligned with the second TOF[X]. In this example, “X” represents the TOF point, and the width of the brackets “[]” indicates the range of measurement accuracy. In other words, low-frequency modulation f2 can be used to eliminate ambiguity caused by phase entanglement, and higher-frequency modulation f1 can be used to improve the accuracy of the estimate.

[0058] Figure 5This is a block diagram of a TOF sensor 10” illustrated by way of example and not limitation, which can be used to resolve phase entanglement ambiguities while maintaining high resolution. The TOF sensor 10” includes Figure 3 The TOF sensor 10' can be supplemented with a DEMUX 44', a counter 46', and a phase estimator 66'. Therefore, in this example embodiment, the TOF sensor 10' has two phase estimators, namely, the phase estimator 66'. and the development of phase estimator 66' Phase estimator 66 and phase estimator 66' are constructed and operate essentially the same. Therefore, it will be appreciated that in this example embodiment, a harmonic-correlated correlator without multipliers is used to detect two simultaneously modulated frequencies.

[0059] In this non-limiting example, DEMUX 44' is an n-bit (5-bit) controlled demultiplexer, where n = 5, whose signal input is coupled to ADC 16 via line 94. Therefore, DEMUX 44' is a 1:2 n (1:32) Demultiplexer. Therefore, DEMUX44' has 32 outputs, of which only four are used. That is, the first DEMUX output Y0, the second DEMUX output Y8, the third DEMUX output Y16, and the fourth DEMUX output Y24. In this example, counter 46 is a 5-bit counter with a clock input CLK and outputs coupled to five control ports of DEMUX 44'. Since counter 46 and counter 46' have the same clock rate, it will be recognized that the signal frequency of the input phase estimator 66 is eight times that of the input phase estimator 66'. This is because only one of the seven outputs of DEMUX 44' is used.

[0060] It should be noted that a TOF sensor 10” can be used to provide optimal spatial resolution performance. An object in the scene is detected by a reference or ranging modulation frequency, and the modulation frequency of a second signal is automatically or programmatically adjusted to provide the highest spatial resolution without phase entanglement.

[0061] In an example embodiment, a sinusoidal waveform is modulated by a binary code comprising multiple chips (“bits”) and having a pulsed cyclic autocorrelation. Those skilled in the art will recognize that such binary code can be of various types, including maximum length sequences (MLS), Kasami codes, and Gold codes. As used herein, “chips” can be used interchangeably with “bits,” and “pulsed” is defined and indicates that the peak autocorrelation sidelobes (PSL) of the signal are substantially suppressed, for example, not exceeding 1 / 16 of the sequence length. More specifically, as used herein, “pulsed” means that the PSL of the signal can range from zero (pure pulse) to a small proportion of the sequence length, for example, 1 / 32, 1 / 16, or 1 / 8 of the sequence length, depending on the application. Those skilled in the art will recognize that this definition of “pulsed” relates to binary code having a periodic autocorrelation function that closely approximates the Dirac delta function. As a non-limiting example, an MLS binary code comprising multiple chips and having a pulsed cyclic autocorrelation will be discussed in more detail.

[0062] exist Figure 6 In this example, the Maximum Length Sequence (MLS) generator 96 includes a maximum linear feedback shift register 98 and a modulo-2 adder 100. In this example, the shift register 98 has a length of 4. For this example embodiment, the next value in register a3 is determined by the modulo-2 sum of a0 and a1. The MLS generator 96 recursively generates pseudo-random binary (MLS) sequences, which can be used for disambiguation purposes.

[0063] The circular autocorrelation of MLS is a Kronecher delta function, and the DC offset and time delay depend on its implementation. For ±1, the conventional approach is as follows:

[0064]

[0065] Where s* represents complex conjugation, and [m+n] N Represents a circular offset.

[0066] Figure 7This is a flowchart of method 102, illustrated by way of example and not limitation, for distinguishing a phase-shifted reflected waveform of a transmitted waveform from other waveforms. In operation 104, binary code with pulse-like cyclic autocorrelation properties is obtained, for example, from an MLS generator 96, a lookup table, etc. Next, in operation 106, a sine (e.g., cosine) wave is modulated by the binary code to form a series of bits (“chips”) on a sine carrier. Optionally, gaps are provided between adjacent chips in this operation 106, as will be described in detail below. Then, in operation 108, the modulated waveform is transmitted as the transmitted waveform from, for example, an LED or laser diode. Next, in operation 110, the reflected waveform is received, for example, by a photodetector and demodulated in operation 112 using the same binary code used to generate the transmitted waveform. This operation 112 may optionally provide gaps between adjacent chips (replacing or combining operation 106), after which the initial signal of the transmitted waveform is recovered. For example, signal gaps are provided between adjacent chips in a related signal (e.g., binary code used to modulate the modulated waveform) to accommodate symbol variations with timing uncertainties in the reflected waveform.

[0067] Reference Figure 8A , 8B 9A and 9B are discussed in more detail. Figure 7 The process of operation 106. Figure 8A The diagram illustrates the use of cosine wave MLS modulation into several bits or "chips" C, such as C0, C1...C5, of the MLS signal. Figure 8B As shown. In another example embodiment, the sequence contains 255 chips. Chip C can have a value of +1 or -1. That is, for each positive chip, there is a positive burst sequence of cosine waves, and for each negative chip, there is a negative burst sequence of cosine waves. These cosine burst sequences can be demodulated into positive and negative pulses 118 by a phase detector. In this example, the pulses constitute the sequence +1, +1, -1, -1, +1, -1, +1.

[0068] like Figure 8BAs shown, optionally, a gap is provided between each chip C, where the value is zero, for example, a brief pause in the signal to accommodate symbol changes with timing uncertainties. That is, if a receiver of the modulated cosine wave attempts to read the signal when the symbol changes between two chips, a decoding error may occur. By providing a gap between each chip, this embodiment prevents bit errors between chips that are changing symbols by providing a buffer between adjacent chips. Although this technique may reduce the signal strength of the modulated cosine wave 116 to some extent, this is far outweighed by the reduction in errors that occur during symbol changes between chips. It should be noted that the gap between chips C can be provided before transmission, for example, in the transmitted waveform, or after the reflected waveform is received. In either case, a related waveform (e.g., as shown) is provided. Figure 8B As shown, the binary code (as one input to the correlator) includes a pulse-like cyclic autocorrelation property and an optional gap between adjacent chips.

[0069] Figure 9A This is a schematic diagram of an MLS-modulated cosine wave 120. Figure 9B This is an enlarged version of the MLS-modulated cosine wave 120, showing the gap "G" at the waveform apex. The width "W" of the gap is preferably sufficient to overcome any timing uncertainties during the sign change and represents only phase information, not amplitude.

[0070] It should be noted that the benefits of gaps can be achieved either by transmitting a signal that already includes a gap or by transmitting an uninterrupted modulated transmission waveform and inserting an equivalent gap into the received reflected waveform. In either case, the relevant waveform (e.g., as...) Figure 8B (As shown) forms the input to the correlator. Since the correlation is essentially the integral of the product of two signals, it doesn't matter whether the multiplier or the multiplicand is zero, because the result will be zero in either case.

[0071] Although specific terms and devices have been used to describe various embodiments, such description is for illustrative purposes only. The language used is descriptive and not limiting. It should be understood that changes and variations can be made by those skilled in the art without departing from the spirit and scope of the invention as supported by the written disclosure and drawings. Furthermore, it should be understood that aspects of the other embodiments can be interchanged, in whole or in part. Therefore, the claims are intended to be interpreted in accordance with the true spirit and scope of the invention, without limitation or estoppel.

Claims

1. A phase-shift sensor for reflected signals, comprising: A digital processor having a digital processor port, a digital input terminal, and a digital output terminal; A driver having a digital driver port coupled to the digital processor port and the driver output port; A waveform transmitter coupled to the driver output port and used to transmit a modulated waveform provided at the driver output port, the modulated waveform comprising a binary code modulated sine wave, the binary code modulated sine wave comprising a plurality of chips, wherein the binary code has a pulsed cyclic autocorrelation property. A waveform receiver, the waveform receiver being used to receive a reflected waveform that has undergone a phase shift relative to the modulated waveform; as well as An analog-to-digital converter having an input terminal coupled to the waveform receiver and an output terminal coupled to the digital input terminal of the digital processor. Signal gaps are provided between adjacent chips of the binary code modulated sine wave or between adjacent chips of the reflected waveform to accommodate symbol changes with timing uncertainties.

2. The reflected signal phase shift sensor according to claim 1, wherein the waveform transmitter comprises at least one of a light-emitting diode and a laser, and wherein the waveform receiver is a photoelectric detection device, such that the modulated waveform is a modulated electromagnetic waveform and the reflected waveform is a reflected electromagnetic waveform that has undergone a phase shift relative to the modulated electromagnetic waveform.

3. The reflected signal phase shift sensor according to claim 2, wherein the digital processor further includes a digital code generator that generates digital codes having pulse-like cyclic autocorrelation properties.

4. The reflected signal phase shift sensor according to claim 3, wherein the digital processor is implemented as an application-specific integrated circuit.

5. A method for distinguishing a phase-shifted reflected waveform of a transmitted waveform from other waveforms, comprising: Obtain binary code with a sequence of pulsed cyclic autocorrelation properties arranged as multiple chips; The binary code is used to modulate a sine wave to generate a modulated waveform; The modulated waveform is transmitted using a waveform transmitter; Receive the reflected waveform that has undergone a phase shift relative to the modulated waveform; as well as Demodulate the related waveform, including the reflected waveform, using the same binary code used to generate the modulated waveform. The method further includes: providing gaps between the chips of the modulated waveform before transmitting the modulated waveform using the waveform transmitter, or providing gaps between the chips of the reflected waveform before demodulating a related waveform including the reflected waveform.

Images