# Adaptive High-Speed Echo Data Acquisition Method for Bathymetric LiDAR

## Abstract

The real-time data acquisition system (RTDAQS) in the bathymetric light detection and ranging (LiDAR) instrument, named “GQ-Cormorant 19,” previously developed by our group cannot completely acquire the echo data from the water surface and water bottom owing to the inaccurate determination of the instant of echo signal acquisition, resulting in echo data loss. Therefore, this study developed an adaptive echo signal acquisition method based on a field-programmable gate array (FPGA) chip. The proposed method utilizes the flying height for roughness adjustment, and the peak of echo signal detection and the fine adjustment are combined to accurately determine the instant of echo signal acquisition. The improved RTDAQS onboard the GQ-Cormorant 19 was mounted on unmanned aerial vehicle (UAV) and unmanned surface vessel (USV) platforms and the performance was verified through an indoor corridor and several outdoor water fields. The experimental results demonstrated that the improved RTDAQS can improve the effective echo data rate by 11.5% when compared with the previously developed RTDAQS. Therefore, it can be concluded that the improved RTDAQS can implement the adaptive high-speed acquisition of echo data and obtain high-quality echo data acquisition (DAQ) on various platforms, such as UAV and USV.

## Authors

Guoqing Zhou *College of Mechanical and Control Engineering, Guangxi Key Laboratory of Spatial Information and Geomatics, College of Earth Sciences, College of Geomatics and Geoinformation, Guilin University of Technology, Guilin, China; Center for Remote Sensing, Tianjin University, Tianjin, China* [ORCID: 0000-0001-8295-0496](https://orcid.org/0000-0001-8295-0496)

Guoshuai Jia *College of Mechanical and Control Engineering and Guangxi Key Laboratory of Spatial Information and Geomatics, Guilin University of Technology, Guilin, China*

Xiang Zhou *College of Mechanical and Control Engineering and Guangxi Key Laboratory of Spatial Information and Geomatics, Guilin University of Technology, Guilin, China; School of Microelectronics, Tianjin University, Tianjin, China*

Naihui Song *College of Mechanical and Control Engineering and Guangxi Key Laboratory of Spatial Information and Geomatics, Guilin University of Technology, Guilin, China* [ORCID: 0009-0003-8401-5495](https://orcid.org/0009-0003-8401-5495)

Jinhuang Wu *College of Mechanical and Control Engineering and Guangxi Key Laboratory of Spatial Information and Geomatics, Guilin University of Technology, Guilin, China* [ORCID: 0000-0003-0306-5742](https://orcid.org/0000-0003-0306-5742)

Ke Gao *College of Mechanical and Control Engineering and Guangxi Key Laboratory of Spatial Information and Geomatics, Guilin University of Technology, Guilin, China*

Jing Huang *College of Physics and Electronic Information Engineering, Guilin University of Technology, Guilin, China*

Jiasheng Xu *College of Geomatics and Geoinformation, Guilin University of Technology, Guilin, China* [ORCID: 0000-0002-2055-8947](https://orcid.org/0000-0002-2055-8947)

Qiang Zhu *College of Earth Sciences, Guilin University of Technology, Guilin, China*

## Publication Information

**Journal:** IEEE Transactions on Geoscience and Remote Sensing **Year:** 2024 **Volume:** 62 **Pages:** 1-17 **DOI:** [10.1109/TGRS.2024.3386687](https://doi.org/10.1109/TGRS.2024.3386687) **Article Number:** 10495068 **ISSN:** Print ISSN: 0196-2892, Electronic ISSN: 1558-0644

## Metrics

**Paper Citations:** 6 **Total Downloads:** 347

## Funding

- Guangxi Science and Technology Base and Talent Project (Grant: Guike AD19254002 and Guike AD23023012)
- National Natural Science of China (Grant: 41961065)
- Guangxi Innovative Development Grand Program (Grant: GuikeAA18118038 and GuikeAA18242048)
- Guangxi Natural Science Foundation for Innovation Research Team (Grant: 2019GXNSFGA245001)
- Guilin Research and Development Plan Program (Grant: 201902102)
- National Key Research and Development Program of China (Grant: 2016YFB0502501)
- BaGuiScholars Program of Guangxi

---

## Keywords

**IEEE Keywords:** Laser radar, Data acquisition, Surface emitting lasers, Field programmable gate arrays, Adaptive systems, Computer architecture, Autonomous aerial vehicles

**Index Terms:** Adaptive Method, Light Detection And Ranging, High-speed Data, Echo Data, High-speed Acquisition, High-speed Data Acquisition, Unmanned Aerial Vehicles, Data Acquisition System, Echo Signal, Real-time Acquisition, Autonomous Surface Vehicles, Flying Height, Data Quality, Sampling Time, Triggering, Time Delay, Experimental Process, Default Mode, Square Wave, Manual Mode, Square Wave Signal, Solid-state Drives, External Trigger, Unmanned Aerial Vehicle Platform, Wave Signal, Photomultiplier Tube, Submodule, Narrow Signal, Bathymetric Map

**Author Keywords:** Adaptive acquisition, bathymetry, echo signal, field-programmable gate array (FPGA), light detection and ranging (LiDAR)

undefined
## SECTION I. Introduction

The research team from the Guilin University of Technology has developed a real-time data acquisition system (RTDAQS) for a single-band bathymetric light detection and ranging (LiDAR) system [^1], [^2], [^3], known as GQ-Cormorant 19, which operates at a wavelength of 532 nm with a photomultiplier tube (PMT) detector for capturing the echo signals from the water surface and water bottom [^4], [^5] and stores the detected echoes on a solid-state drive. This device has been specifically developed for bathymetric measurements in coastal shallow waters [^6], [^7], [^8], [^9], [^10], [^11], [^12], [^13], [^14].

The previously developed RTDAQS was based on analog-to-digital converter (ADC) + field-programmable gate array (FPGA) + Zynq-7000 All Programmable SoC (ZYNQ). This system features a sampling rate of up to 2 GS/s, a sampling resolution of 14 bits, and a sampling accuracy of six LSB [^15]. However, this RTDAQS has some shortcomings. When the flying height of the platform changes, the data acquisition (DAQ) is initialized only by external triggering. The impact caused by the time difference between the time of trigger acquisition and the receiving time of the first echo results in an incomplete acquisition of the echo data, even completely missing the echo data. In this case, the effective echo data rate, which is a ratio of the number of effective echo data to the number of total data, where the effective echo data are defined as the water depth that can be calculated by the waveform decomposition of echo data, is reduced. Therefore, high-quality bathymetry data cannot be ensured. For this reason, this study presents an adaptive sampling method using external trigger signals in combination with both the peak moment of the echo signal and the flying height of the LiDAR instrument to improve the quality of echo data received by the RTDAQS.

## SECTION II. Related Work

The echo DAQ system is a critical part of the bathymetric LiDAR [^16], [^17], [^18], [^19], [^20], [^21], and many scholars have attempted to develop a high-speed DAQ system [^22], [^23]. For example, Xingguang et al. [^24] developed a 300 MS/s sampling rate DAQ system. However, this system has a low sampling rate, and it failed to obtain high-accuracy echo data with LiDAR for bathymetry measurement. Schiffer et al. [^25] created a software program on the PC side that allowed the user to specify the trigger parameters and the threshold for radiation DAQ. However, the parameter cannot be changed adaptively according to other parameters such as the flying height; therefore, it cannot be applied to bathymetric LiDAR. Kelley [^26] designed a data acquisition system to detect cosmic rays by triggering the data acquisition through self-triggering. However, it is difficult to ensure that all water surface and water bottom echo data are acquired and stored in a solid-state drive when the sampling time is fixed. Khedkar and Khade [^27] proposed a high-speed FPGA-based data acquisition system with an external trigger. This triggering method cannot guarantee the integrity of the echo data under the condition that the sampling time is determined; therefore, it cannot be applied to bathymetric LiDAR for measuring water depth. Veiga and Grunfeld [^28] designed a digital signal processor for gamma-ray applications with free-running ADC. Since it collects a large amount of data, it cannot be applied to the RTDAQS, which features a sampling rate of up to 2 GS/s for bathymetric LiDAR. Shu et al. [^29] designed a tokamak data acquisition system by triggering the signal acquisition with an external trigger. However, only using an external trigger to trigger the echo signal acquisition cannot determine the acquisition instant; therefore, it cannot perform adaptive acquisition for the bathymetric LiDAR. Adamczewski-Musch et al. [^30] used PEXORNET driver software on a Linux system to realize trigger data acquisition. However, using this method to trigger the echo signal acquisition for LiDAR is complicated and requires manual operation. Tian et al. [^31] designed a real-time multibeam sonar system to control the ADC acquisition of the echo signal by generating a continuous wave (CW) pulse control signal through FPGA. However, the sampling rate is only 5 MS/s, which is not suitable for high-precision bathymetry readout drivers (LiDAR). Sahoo et al. [^32] designed a data acquisition system for superheated emulsion detectors, which can adjust prepost triggers. In this system, the pretrigger acquisition length must be set manually; therefore, it cannot be applied to adaptive echo data acquisition for LiDAR. Li et al. [^33] designed a data acquisition system containing RODS for the STAR small-strip Thin Gap Chamber (sTGC) detector by threshold triggering, which can read the charged signals of the sTGC detector and implement data filtering. Cheng and Xie [^34] designed a high-repetition-rate LiDAR integrated off-line echo signal acquisition system using external triggering. This system enabled atmospheric composition analysis and environmental pollution detection. However, it does not take into account the movement of echo signal, and thus cannot be applied to LiDAR for variable flying height unmanned aerial vehicle (UAV) platforms.

As overviewed above, self-triggering [^35], [^36], external triggering [^37], [^38], [^39], software triggering [^40], [^41], and other triggering methods cannot ensure the integrity of the water surface and water bottom echo data for variable flying height LiDAR platform. To achieve an adaptive high-speed echo data acquisition, the acquisition instant must be accurately determined. This study proposes an adaptive sampling method using external trigger signals combined with the peak moment of the echo signal and the height of the LiDAR system for the real-time acquisition of surface and bottom water data. The organization of the article is as follows. The development of both hardware and software for the improved RTDAQS is described in Section III. The experiments and discussions are presented in Section IV, and the conclusions are given in Section V.

## SECTION III. Adaptive High-Speed Echo Data Acquisition Method

### A. Architecture of GQ-Cormorant 19 With the RTDAQS

The architecture of the original RTDAQS embedded in the GQ-Cormorant 19 LiDAR system is shown in Fig. 1(a). It consists of the power module, laser module, control module, 4G Wi-Fi module, and detection module. The power module provides different amplitude power supplies to other modules. The laser module generates and emits a 532 nm laser, it uses a correlation probe and circuitry to generate the periodic narrow pulse signal and the square wave signal. The detection module detects weak light signals reflected from the water surface and bottom and converts them into amplified voltage signals. The control module handles various tasks including receiving and converting position and orientation system (POS) data from the GQ-Cormorant 19, sending and receiving the initial flying height data, controlling laser parameter settings, and the starting and stopping operation of the GQ-Cormorant 19. The 4G Wi-Fi module receives the relevant commands and the initial flying height data from the remote ground station and sends them to the control module. The RTDAQS is responsible for data acquiring and storing the detected echo signals from the water surface and bottom into a solid-state drive.

![Figure 1](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou1abc-3386687-large.gif)

*Fig. 1. (a) Architecture of GQ-Cormorant 19. (b) Architecture of the original RTDAQS embedded in GQ-Cormorant 19. (c) Proposed RTDAQS.*

### B. Design and Implementation of Adaptive Echo Data Acquisition

The architecture of RTDAQS is shown in Fig. 1. The improved RTDAQS [Fig. 1(c)] consists of three modes: “default mode,” “manual mode,” and “adaptive mode,” designed in the FPGA. In this study, the “default mode” is defined as the original RTDAQS utilizing an external trigger for data acquisition. An “adaptive mode” is defined as the acquisition of the complete echo signal by adaptively determining the instant at which the acquisition is triggered. A detailed description is given as follows:

#### 1) Design of the Adaptive Echo Data Acquisition:

The timing diagram of the main signals for RTDAQS is shown in Fig. 2. The pulsewidth of the periodic narrow pulse signal is 2.0 ± 0.5 ns, with a wavelength of 532 nm and a frequency of 2 kHz. The square wave signal has the same frequency and an amplitude of 3.3 V.

![Figure 2](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou2-3386687-large.gif)

*Fig. 2. Overall timing diagram for adaptive echo data acquisition. $T_{1} ^{n}$ is the time difference between the square wave signal and the periodic narrow pulse signal output from the laser adjusted after the $n \text { th}$ comparison; $T_{2} ^{n}$ is the delay time before sampling adjusted after the nth comparison, $n=1,2,\ldots,d$ ; $t_{0}$ is the sampling time of the RTDAQS; $t_{1}$ is the time required for the PMT of the GQ-Cormorant 19 to detect the water surface echo signal at the initial flying height data without considering any other factors; $t_{2}$ is the fixed delay of the GQ-Cormorant 19 system (i.e., the delay from emitting 532 nm laser to the RTDAQS adaptive acquisition and storage to the data when the trigger signal and the periodic narrow pulse signal of the laser are engraved at the same time and the delay time before sampling is 0); $t_{\max }$ is the peak position of the first echo signal in the sampling time.*

To realize the adaptive mode of operation of RTDAQS, an adaptive module, with submodules of laser delay adjustment, flying height roughness adjustment, peak detection, and fine adjustment, was designed [Fig. 1(c)]. The design of each submodule is described as follows:

a) Flying Height Roughness Adjustment Submodule:

It is designed to determine the instant at which the echo data acquisition is triggered by combining the flight height of the GQ-Cormorant 19 with the external trigger to roughly ensure that the echo data can be collected by the RTDAQS. Mathematically, it can be represented using the following equations. Assuming that $t_{3}$ is defined as the time required from the start of the laser emission until the RTDAQS collects the water surface echo signal detected by the PMT

$$
\begin{align*} t_{1} &=2h_{0} /c \tag{1}\\ t_{3} &=t_{1} +t_{2} \tag{2}\end{align*}
$$

where *c* is defined as the speed of propagation of laser in air, $c=3\times 10^{8} \textrm {m}/\textrm {s}$. $h_{0}$ is the initial flight height of the GQ-Cormorant 19. After backtracking from experiments using GQ-Cormorant 19, $t_{2} =10 \textrm {}{\textrm {ns}}$. To verify whether the RTDAQS of the GQ-Cormorant 19 system can acquire the echo data at this flying height, the magnitudes of $t_{0}$ and $t_{3}$ can be compared. If $t_{3} \ge t_{0}$, the RTDAQS cannot collect any water surface echo signal, and it is necessary to adjust the delay time before sampling to roughly adjust this echo data in the middle of the sampling time. The delay time before sampling that must be adjusted at this time is given by $T_{2}^{0} =t_{3} -t_{0} /2$.

If $t_{3} < t_{0}$, it means that the RTDAQS can collect water surface echo signal, and no adjustment is required, that is $T_{2} ^{0}=0$.

At this time, it can be guaranteed that the data acquired by the RTDAQS contain water surface echo signals.

b) Peak Detection Submodule:

Through the previous module, the quality of the data cannot be guaranteed, and there are incomplete echo data. Therefore, this module finds the cache through the traversal method to get the maximum value of the echo data location.

c) Fine Adjustment Submodule:

Based on the previous combination of modules, the instant of triggered acquisition is finely adjusted according to the position of the echo peak to ensure data integrity and improve data quality. Mathematically, it can be described as follows. The maximum time difference in the calculation of the laser water surface and water bottom echo signal $t_{4}$ can be calculated using the following equation:

$$
\begin{equation*} t_{4} =2h_{\max } /c_{1} \tag{3}\end{equation*}
$$

where $c_{1}$ is the speed of propagation of light in water, $c_{1} =3/4\times 3\times 10^{8} \textrm {m}/\textrm {s}$; $h_{\max }$ is the maximum bathymetry of the GQ-Cormorant 19 LiDAR, $h_{\max } =20 \textrm {}\text {m}$. The maximum value of $t_{4}$ was calculated to be $\approx 178 \textrm {}{\textrm {ns}}$. Since GQ-Cormorant 19 is required to measure the water depth, $t_{0}$ was set to $0.4 \textrm {}\textrm {ns}$. At this point, the magnitudes of $t_{\max }$, $t_{0} /4$, and $t_{0} /2$ can be compared. If $t_{\max } < t_{0} /4$, $\Delta T_{1} =t_{0} /4-t_{\max }$, adjust the time difference between the laser output square wave signal and periodic narrow pulse signal $T_{1} ^{n}=T_{1}^{n-1}+\Delta T_{1}$ and the delay time before sampling $T_{2} ^{n}=T_{2}^{n-1}$. $T_{1} ^{0}$ is the initial time difference between the square wave signal output from the laser and the periodic narrow pulse signal, and the default is $0.~T_{2} ^{0}$ is obtained from the flying height roughness adjustment submodule.

If $t_{0} /4\le t_{\max } \le t_{0} /2$, adjust the time difference between the laser output square wave signal and periodic narrow pulse signal $T_{1} ^{n}=T_{1} ^{n-1}$ and the delay time before sampling $T_{2} ^{n}=T_{2}^{n-1}$.

If $t_{0} /2 < t_{\max }$, $\Delta T_{2} =t_{\max } -t_{0} /4$, adjust the time difference between the laser output square wave signal and periodic narrow pulse signal $T_{1} ^{n}=T_{1} ^{n-1}$ and the delay time before sampling $T_{2} ^{n}=T_{2}^{n-1}+\Delta T_{2}$.

d) Laser Delay Adjustment Submodule:

After the FPGA improved RTDAQS determines $T_{1} ^{n}$, it uses the J30J connector to implement the RS232 protocol to send the relevant commands to the control module. The corresponding commands are then sent through the control module to make adjustments to the laser module.

The above steps can ensure the complete acquisition of the water surface and water bottom echo signals. At this time, when the bathymetry for the case of the maximum bathymetry, due to the echo peak distance from the beginning of the acquisition and the end of the acquisition time are approximately $t_{0} /4$, from (1), which can be calculated to cope with the change in the flying height of UAV carried GQ-Cormorant 19 is approximately $\Delta h=15 \textrm {}\text {m}$. Because of the UAV climbing or descending with a maximum speed of 4 m/s, the detection of the peak of the FPGA adaptive condition of the frequency of 500 Hz is sufficient to cope with the change in the flying height brought about by the drone flying height changes.

#### 2) Implementation of Adaptive Echo Data Acquisition:

According to the design, the parameters $t_{0}$, $T_{2}^{n}$, and $T_{1} ^{n}$ must be varied, and $t_{\max }$ must be detected (Fig. 2), which is implemented as follows:

a) Adjustment of $t_{0}$ and $T_{2} ^{n}$:

The two parameters are adjusted by the FPGA in the RTDAQS. The timing diagram for sampling time adjustment is shown in Fig. 3, where “pw_buf” represents the sampling time signal. The proposed finite-state machine (FSM) for the delay time control before sampling is shown in Fig. 4, where “delay” represents the delay time before sampling. Consider the acquisition module channel 1 as an example; when the reset signal “rst = 1,” the 16-bit storage counter will be “0” in each bit, and the state will be set to “00.” When there is a rising edge of the clock, if “store_en1_cnt &lt; delay,” the counter will be increased by one; if “store_en1_cnt = delay,” it will enter into the state “01,” and the counter will be set to “0,” and at this time, it will provide the time for acquisition. Next, set the value “pw_buf.” If “store_en1_cnt &lt; pw_buf,” then store enable signal “store_en_1_buf” will be set to “1” for storage until the counter reaches the sampling time setting value.

![Figure 3](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou3-3386687-large.gif)

*Fig. 3. Timing diagram for sampling time adjustment. $N_{1}$ is the value of “pw_buf” signal. $N_{2}$ is the value of the “delay” signal.*

![Figure 4](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou4-3386687-large.gif)

*Fig. 4. Proposed FSM for the delay time control before sampling.*

b) Detection of $t_{\max}$:

The block diagram of the peak detection submodule is shown in Fig. 5. The frequency of the clock signal is 250 MHz. The data_chmax submodule will judge if condition1 of “max_en = 0” is satisfied. If condition1 is satisfied, then make each bit of the 24-bit counter to be “0,” and make the valid signal to be “0.” If condition1 is not satisfied, it will judge condition2 of “cnt $< 499~900$.” If condition2 is satisfied, then make the counter to increase by “1,” make the trig signal to be “1,” and make the valid signal to be “0.” At this point, it is in the phase of obtaining the maximum value and the position of the maximum value. If condition2 is not satisfied, it will judge condition3 of “499$900\le$ cnt &lt; 500000.” If condition3 is satisfied, then make the counter to increase by “1,” make the trig signal to be “0,” and make the valid signal to be “1.” At this point, this is the maximum value holding phase, which ensures the transmission of the data_max signal. In the phase of obtaining the maximum value and the position of the maximum value, consider the acquisition module channel 1 as an example. The 128-bit wide data_1 signal, input to the submodule, is sequentially split into eight 16-bit wide data. Because the sampling rate of the RTDAQS is 2 GS/s and the sampling time is 400 ns, each dataset has 800 points. Take the absolute values of these data and compare them to get the maximum value. After deriving the maximum value, take the position of the first maximum value in the sampling time $t_{\max }$. Finally, the 16-bit maximum value and 10-bit maximum value positions of the four channels are combined to form the 104-bit data_max output signal that transmits data to the following submodules when the condition of “valid = 1” is satisfied.

![Figure 5](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou5-3386687-large.gif)

*Fig. 5. Block diagram of peak detection submodule.*

In contrast to the original RTDAQS, echo data acquisition using improved RTDAQS is as follows: first, when the UAV arrives at the designated location, the GQ-Cormorant 19 LiDAR is operated by clicking the “run” button on the cell phone. The flying height data acquired by the UAV’s host computer is transmitted to the control module via the 4G Wi-Fi module. At this time, the laser module generates and emits a 532 nm laser. The main wave detector detects the laser periodic narrow pulse signal and will be accessed to channel 1 of the RTDAQS for acquisition and simultaneously generates a square wave signal with the same frequency, which serves as an external trigger signal combined with initial flying height. Next, the normally ON-type (NN-type) and normally OFF-type (NF-type) PMTs in the detection module convert the received laser pulse signals into electrical signals, which are converted into digital signals by the two AD9208 chips using a multichannel synchronous approach. The peak moment analysis is then conducted based on the collected data, and the corresponding parameter adjustments are conducted to realize the adaptive and complete collection of echo data.

The above part described the adaptive high-speed echo data acquisition function, by introducing the principle of adaptive acquisition and the implementation of the improved RTDAQS. Section IV describes the experimental comparison with the default mode to verify the feasibility and superiority of the improved RTDAQS.

### C. Hardware Implementation for the Improved RTDAQS

The hardware modules of the improved RTDAQS in the GQ-Cormorant 19 are shown in Fig. 6. It contains two AD9208 chips [^42], [^43], the XC7K480T-2FFG1 in the FPGA Kintex7series [^44], [^45], the XCZU4CG in the Zynq UltraScale+ MPSoC, the SSD for Nvme.M.2, and the heat dissipation module. Compared to the previous generation, the hardware part contains a heat dissipation module, which consists of fans and thermal conductive silicone films. The solid-state drive has been upgraded from 970 EVO Plus to 970 Pro. The new generation solid-state drive offers better heat dissipation performance. The improved RTDAQS is powered by a separate 12-V power supply. It solves the hardware problem that when this system is onboard the GQ-Cormorant 19, the temperature of the internal chip increases, which leads to its inoperability and the possible loss of data transmission.

![Figure 6](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou6-3386687-large.gif)

*Fig. 6. Hardware implementation of the improved RTDAQS.*

### D. Case Analysis With a Fixed Flying Height

In a special case when the height of the platform carried with GQ-Cormorant 19 from the water surface is fixed, the height and peak moment used for the adaptive acquisition trigger designed in this study are unchanged. This method of repeated detection would waste resources and generate excess energy consumption. In this study, an acquisition method was adopted by the RTDAQS to solve this special case, referred to as “manual mode.” The “manual mode” is discussed below as an example with the GQ-Cormorant 19 mounted on the USV.

In the “manual mode” of RTDAQS, it is not required to repeatedly modify the magnitudes of $T_{1} ^{n}$ and $T_{2}^{n}$; $T_{1}$ will be used accordingly to represent $T_{1} ^{n}$ and $T_{2}$ to represent $T_{2} ^{n}$.

$h_{0}$ in (1) represents the height of the GQ-Cormorant 19 mounted on the USV from the water surface.

From (1) and (2), $t_{3}$, the time required for the GQ-Cormorant 19 to collect the echo signal detected by the PMT from the time it emits the laser light, can be derived after fixing the height. Next, $t_{3}$ and $t_{0} /2$ were compared.

If $t_{3} < t_{0} /2$, $T_{1}$ must be adjusted, which can be calculated using the following equation:

$$
\begin{equation*} T_{1} =t_{0} /2-t_{3}. \tag{4}\end{equation*}
$$

If $t_{3} =t_{0} /2$, no parameters require adjustments.

If $t_{3} >t_{0} /2$, $T_{2}$ must be adjusted, which can be calculated using the following equation:

$$
\begin{equation*} T_{2} =t_{_{3}} -t_{0} /2. \tag{5}\end{equation*}
$$

After the above steps through the “manual mode” realize the complete collection of echo data by adjusting the echo data to the middle of the stored results with the platform height fixed. The above part describes the realization of the RTDAQS of GQ-Cormorant 19 adaptive echo data acquisition function, and the flowchart for the realization of “manual mode” and “adaptive mode” is shown in Fig. 7.

![Figure 7](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou7-3386687-large.gif)

*Fig. 7. Flowchart of adaptive echo data acquisition.*

This section has described the design and implementation of the adaptive high-speed echo data acquisition function. Section IV describes the experiments to verify the feasibility and advantages of the proposed RTDAQS.

## SECTION IV. Experiment Validation and Discussion

### A. Validation in the Laboratory Corridor

The experiment was set up as follows: the GQ-Cormorant 19 LiDAR was placed on one side of the laboratory corridor, and a black baffle was placed on the other side, with a distance of 20 m (Fig. 8). The specifications of the GQ-Cormorant 19 LiDAR are shown in Table I. The experimental process is as follows:

1. Adjust the laser emission position and the flying height of the baffle so that the laser can return to the GQ-Cormorant 19 optical system before the GQ-Cormorant 19 system is activated.
2. Click run on the touch screen to emit the laser onto the black baffle, which is then diffusely reflected, and the light signal is detected by the PMT in the GQ-Cormorant 19 detection module.
3. Change the distance between the black baffle and the GQ-Cormorant 19: 14, 18, and 22 m, and the experimental results of the echo signal at 1 min are shown in Fig. 9(a1)–(a3).
4. Turn on the manual mode, repeat steps 1) and 2), manually input the flying height of the GQ-Cormorant 19 as 18, 20, and 22 m, and change the distance between the black baffle and the GQ-Cormorant 19 accordingly: 14, 18, and 22 m. The experimental results of the echo signal at 1 min are shown in Fig. 9(b1)–(b3).

Fig. 9(a1)–(a3)

Fig. 9(b1)–(b3)

![Figure 8](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou.t1-3386687-large.gif)

*TABLE I*

![Figure 9](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou8-3386687-large.gif)

*Fig. 8. Experiment in the laboratory corridor.*

![Figure 10](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou9ab-3386687-large.gif)

*Fig. 9. Experimental results for the laboratory corridors. (a1)–(a3) Collected echo signals at different distances, with the RTDAQS in the default mode. (b1)–(b3) Collected echo signals in different distances, with the RTDAQS in the manual mode, where only the portion of the box is zoomed in horizontally.*

### B. Validation in the Outdoor Pond

The experiment was set up as follows: the GQ-Cormorant 19 LiDAR was placed on one side of an outdoor circular pond, and a reflector was placed on top of the pond to reflect the laser into the circular pond, with a distance of 20 m (Fig. 10). The distance between the reflector and the water surface is 1.05 m. The experimental process is as follows:

1. Adjust the laser emission position and mirror angle so that the laser can return to the GQ-Cormorant 19 optical system before the GQ-Cormorant 19 bathymetric system is activated.
2. Click run on the touch screen to emit the laser into the reflector using the RTDAQS default mode. Next, reflect the laser into the pond, which is then detected by the PMT in the GQ-Cormorant 19 detection module.
3. Change the distance between the GQ-Cormorant 19 and the reflector to 14, 20, 26, and 38 m. The experimental results of the echo signal at 1 min are shown in Fig. 11(a1)–(a4).
4. Repeat step 3) by setting the manual mode flying height to 14, 20, 26, and 38 m. The echo signal results at 1 min are shown in Fig. 11(b1)–(b4).
5. Repeat step 3) by setting the initial value of the adaptive mode flying height to 14 m. The echo signal results at 1 min are shown in Fig. 11(c1)–(c4). The comparing analysis is shown in Table II.

Fig. 11(a1)–(a4)

Fig. 11(b1)–(b4)

Fig. 11(c1)–(c4)

Fig. 11(a1)–11(a3)–11(b1)–11(b3)–11(c1)–11(c3)

![Figure 11](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou.t2-3386687-large.gif)

*TABLE II*

![Figure 12](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou10-3386687-large.gif)

*Fig. 10. Experiment in the outdoor pond. d is the distance between the GQ-Cormorant 19 LiDAR and the reflector; $d_{0}$ is the distance between the reflector and the water surface.*

![Figure 13](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou11abc-3386687-large.gif)

*Fig. 11. Experimental results for the outdoor pond. (a1)–(a4) Collected echo signals at different distances, with the RTDAQS in the default mode. (b1)–(b4) Collected echo signals at different distances, with the RTDAQS in manual mode. (c1)–(c4) Collected echo signals at different distances, with the RTDAQS in the adaptive mode.*

Table II shows that as the distance between the GQ-Cormorant 19 and the reflector increases, the time of maximum peak increases in the default mode; it remains around 200 ns in the manual mode; it remains between 100 and 200 ns in the adaptive mode.

### C. Validation in the Beihai Bay of the Pacific Ocean

The experiment was set up as follows: the GQ-Cormorant 19 LiDAR, POS, and power supply were mounted on the GQ-S20 unmanned surface vessel (USV) (Fig. 12). The GQ-S20 USV is a lightweight dual-propulsion platform developed by our group, which can be operated through automatic path tracking, navigation, and running. Its detailed parameters are shown in Table III. This experiment was conducted in the near-coastal area of Beihai Bay of the Pacific Ocean, Guangxi. The experimental process is as follows:

1. Click run on the touch screen using the RTDAQS default mode of the GQ-Cormorant 19 system. Control the GQ-S20 to scan according to the preset trajectory, and then return to the starting position.
2. Stop the system running and export the data. The echosignal results at 0.5, 1.0, and 1.5 min are shown in Fig. 13(a1)–(a3).
3. Adopt the manual mode with improved RTDAQS, and adjust the height to 0.45 m. Click run on the touch screen of the GQ-Cormorant 19 system to control the GQ-S20 to scan according to the same trajectory, and then return to the starting position.
4. Stop the system running and export the data. The echo signal results at 0.5, 1.0, and 1.5 min are shown in Fig. 13(a1)–(a3).
5. Adopt the manual mode with improved RTDAQS, and adjust the height to 0.45 m. Click run on the touch screen of the GQ-Cormorant 19 system to control the GQ-S20 to scan according to the same trajectory, and then return to the starting position.
6. Stop the system running and export the data. The echo signal results at 0.5, 1.0, and 1.5 min are shown in Fig. 13(b1)–(b3).

![Figure 14](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou.t3-3386687-large.gif)

*TABLE III*

![Figure 15](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou12-3386687-large.gif)

*Fig. 12. Experiment in the Beihai Bay of the Pacific Ocean.*

![Figure 16](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou13ab-3386687-large.gif)

*Fig. 13. Experimental results for the Beihai Bay of the Pacific Ocean. (a1)–(a3) Collected echo signals at different times, with the RTDAQS in the default mode. (b1)–(b3) Collected echo signals at different distances, with the RTDAQS in manual mode.*

Fig. 13(a1)–(a3) shows that when the GQ-S20 follows a predetermined trajectory, the peak echo positions of the shallow-water channel detected by the GQ-Cormorant 19 are 15, 17, and 16.5 ns, correspondingly. Fig. 13(b1)–(b3) shows that when the GQ-S20 follows the same trajectory, the peak echo positions of the shallow-water channel are 200.5, 202.5, and 203ns. According to statistics, 5000 sets of data are randomly sampled from the result data of the two experiments respectively. In the default mode, the number of valid datasets is 3058, and the valid data rate is 61.2%. When the manual mode is adopted, the number of valid datasets is increased to 3486, and the valid data rate is increased to 69.7%. After comparing the two modes, it can be concluded that when the improved RTDAQS is mounted on the USV platform, adopting the manual mode, it can adjust the position of the peak echo of the shallow-water channel in the middle of the whole set of sampling data, which can effectively ensure the quality of the collected surface and underwater echo signal data.

### D. Validation in the Wujiu Beach by the Lijiang River

The LiDAR experiment was set up as follows: the GQ-Cormorant 19 LiDAR and POS were assembled on the UAV, and the UAV flying height was 50 m (Fig. 14). The UAV in this experiment is an eight-rotor large-scale UAV platform named KWT-X8L-25, capable of carrying electromagnetic sensitive devices, and can realize automatic hovering of the UAV. The detailed parameters of this UAV are shown in Table IV.

![Figure 17](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou.t4-3386687-large.gif)

*TABLE IV*

![Figure 18](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou14-3386687-large.gif)

*Fig. 14. Experiment conducted in Wujiu Beach in the Lijiang River using the GQ-Cormorant 19.*

This experiment was conducted in the Lijiang River next to Wujiu Beach, Guilin City, Guangxi. The environmental parameters on the day of the experiment are shown in Table V. The experimental process is as follows:

1. Click run on the phone screen at the remote ground station when the UAV reaches its designated position using the RTDAQS default mode. Then, the UAV flies according to the specified route path. After the scanning is completed, stop the GQ-Cormorant 19 running and control the UAV to land, and export the data. The echo signal results at 0.5, 1, 1.5, and 2 min are shown in Fig. 15(a1)–(a4).
2. Turn on the GQ-Cormorant 19 through the phone when the UAV reaches the designated position. The UAV again follows the same path. At this time, adopt the adaptive mode of the GQ-Cormorant 19.
3. Stop running and control the UAV to land, and export the data after the scanning is completed. The echo signal results at 0.5, 1, 1.5, and 2 min are shown in Fig. 15(b1)–(b4).

Fig. 15(a1)–(a3)

Fig. 15(b1)–(b4)

Fig. 15(a4)

$1/4\sim 1/2$ ![Figure 19](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou.t5-3386687-large.gif)

*TABLE V*

![Figure 20](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou15ab-3386687-large.gif)

*Fig. 15. Experimental results for the Wujiu Beach by the Lijiang River. (a1)–(a4) Collected echo signals at different times, with the RTDAQS in the default mode. (b1)–(b4) Collected echo signals at different distances, with the RTDAQS in the adaptive mode.*

Table VI shows that when the RTDAQS adopts default mode with an initial flying height of the UAV platform of 50 m, complete echo data were not collected in some cases. The effective echo data containing water depth information was fewer. According to statistics, in 5000 randomly sampled datasets, the effective echo datasets are 3070, and the effective echo data rate is 61.4%. When RTDAQS adopts the adaptive mode, the number of effective datasets increases to 3647, and the effective echo data rate increases to 72.9%. Thus, it can be concluded that the GQ-Cormorant mounted on the UAV platform adopts the adaptive acquisition mode, which can increase the effective echo data rate, thus ensuring the echo data quality.

![Figure 21](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou.t6-3386687-large.gif)

*TABLE VI*

The echo data in the Wujiu Beach acquired using the improved RTDAQS were subsequently processed to generate a point cloud data map, which is used to demonstrate the advantages of the proposed method [Fig. 16(a)]. The subsequent processes are: first, the waveform decomposition is carried out through the steps of: 1) Gaussian filtering; 2) parameter estimation and noise elimination using the Gaussian inflection point selection method; and 3) parameter fitting using a particle swarm algorithm developed by our research group [^46]. Finally, 3-D coordinates are calculated using software developed by our research group [^46] and bathymetric map is generated and depicted in Fig. 16(b).

![Figure 22](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou16abc-3386687-large.gif)

*Fig. 16. (a) Depicts the point cloud data collected in the Wujiu Beach in the Lijiang River using both the GQ-Cormorant 19 LiDAR and the shallow-water multibeam sounder. (b) Bathymetric map collected using GQ-Cormorant 19 LiDAR. (c) Bathymetric map using the shallow-water multibeam sounder.*

In order to confirm the 3-D coordinate accuracy achievable, a shallow-water multibeam sounder mounted on the unmanned ship vehicle (USV) is used to acquire bathymetric data in the same water zone. The experiment was set up as follows: the sounder was assembled on the USV (Fig. 17). The specifications of the shallow-water multibeam sounder are shown in Table VII. The experimental process is as follows:

1. Launch the assembled USV into the water and make the USV navigate according to the same trajectory.
2. Collect the corresponding data and transfer them to the computer through the serial port.
3. Generate the trajectory map [Fig. 16(a)] and bathymetric map [Fig. 16(c)] by CloudCompare software.

Fig. 16

![Figure 23](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou.t7-3386687-large.gif)

*TABLE VII*

![Figure 24](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou17-3386687-large.gif)

*Fig. 17. Experiment conducted in the Wujiu Beach in the Lijiang River using the shallow-water multibeam sounder onboard USV.*

In addition, the eight checkpoints, acquired by GQ-Cormorant 19 LiDAR and the multibeam sounder, are selected to further quantitively evaluate the bathymetric accuracy. Under the same world geodetic system (WGS)-84, eight points with the same *X* and *Y* coordinates were randomly selected, and their *Z* coordinate was compared ($Z_{1}$ coordinate is acquired by GQ-Cormorant 19 LiDAR, and $Z_{2}$ coordinate is acquired by the multibeam sounder). The results are shown in Table VIII. As observed from Table VIII, the mean of $\Delta Z$ reaches −0.1 m, and the standard deviation is 0.3 m. This result again verified that the bathymetric map measured by GQ-Cormorant 19 LiDAR reaches the same level of accuracy as that measured by the shallow-water multibeam sounder, i.e., the quality of the echo data collected by the GQ-Cormorant 19 LiDAR with the improved RTDAQS system is improved.

![Figure 25](https://ieeexplore.ieee.org/mediastore/IEEE/content/media/36/10354519/10495068/zhou.t8-3386687-large.gif)

*TABLE VIII*

## SECTION V. Conclusion

This study presents an adaptive high-speed echo data acquisition method for bathymetric LiDAR, GQ-Cormorant 19, to overcome the existing problems in the previous generation RTDAQS system embedded in the bathymetric LiDAR. External triggering is used in combination with the flying height and the peak moment of the echo signal to accurately determine the instant of water surface and bottom echo data acquisition. The proposed method was practically verified in the laboratory corridor, outdoor pond, Beihai Bay of the Pacific Ocean, and Wujiu Beach by the Lijiang River. The advantages of this improved method can be summarized as follows: the improved RTDAQS system can adaptively acquire the echo data, and effectively ensure high-quality echo data. Experimental results showed that the effective echo data rate reached 72.9%, that is, an 11.5% increase when compared with the previous generation of the RTDAQS system (default mode). The effective echo data rate reached 69.7%, that is, increasing 8.5% when the flying height of the platform was fixed. The improved RTDAQS system still requires improvement in the selection of the peak position of the echo signal. Therefore, the future work will focus on the selection of peak positions.

## References

[^1]: G. Zhou, “Development of a lightweight single-band bathymetric LiDAR,” Remote Sens., vol. 14, no. 22, p. 5880, Nov. 2022, doi: 10.3390/rs14225880. [DOI](https://doi.org/10.3390/rs14225880) [Google Scholar](https://scholar.google.com/scholar?as_q=Development+of+a+lightweight+single-band+bathymetric+LiDAR&as_occt=title&hl=en&as_sdt=0%2C31)

[^2]: G. Zhou, C. Li, D. Zhang, D. Liu, X. Zhou, and J. Zhan, “Overview of underwater transmission characteristics of oceanic LiDAR,” IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 14, pp. 8144–8159, 2021, doi: 10.1109/JSTARS.2021.3100395. [IEEE](https://ieeexplore.ieee.org/document/9497670) [Google Scholar](https://scholar.google.com/scholar?as_q=Overview+of+underwater+transmission+characteristics+of+oceanic+LiDAR&as_occt=title&hl=en&as_sdt=0%2C31)

[^3]: G. Zhou, “Off-axis four-reflection optical structure for lightweight single-band bathymetric LiDAR,” IEEE Trans. Geosci. Remote Sens., vol. 61, 2023, Art. no. 1000917, doi: 10.1109/TGRS.2023.3298531. [IEEE](https://ieeexplore.ieee.org/document/10203027) [Google Scholar](https://scholar.google.com/scholar?as_q=Off-axis+four-reflection+optical+structure+for+lightweight+single-band+bathymetric+LiDAR&as_occt=title&hl=en&as_sdt=0%2C31)

[^4]: G. Zhou, “PMT gain self-adjustment system for high-accuracy echo signal detection,” Int. J. Remote Sens., vol. 43, nos. 19–24, pp. 7213–7235, Dec. 2022, doi: 10.1080/01431161.2022.2155089. [DOI](https://doi.org/10.1080/01431161.2022.2155089) [Google Scholar](https://scholar.google.com/scholar?as_q=PMT+gain+self-adjustment+system+for+high-accuracy+echo+signal+detection&as_occt=title&hl=en&as_sdt=0%2C31)

[^5]: G. Zhou, “An RF amplifier circuit for enhancement of echo signal detection in bathymetric LiDAR,” IEEE Sensors J., vol. 22, no. 21, pp. 20612–20625, Nov. 2022, doi: 10.1109/JSEN.2022.3206763. [IEEE](https://ieeexplore.ieee.org/document/9898899) [Google Scholar](https://scholar.google.com/scholar?as_q=An+RF+amplifier+circuit+for+enhancement+of+echo+signal+detection+in+bathymetric+LiDAR&as_occt=title&hl=en&as_sdt=0%2C31)

[^6]: G. Zhou, “Adaptive model for the water depth bias correction of bathymetric LiDAR point cloud data,” Int. J. Appl. Earth Observ. Geoinf., vol. 118, Apr. 2023, Art. no. 103253, doi: 10.1016/j.jag.2023.103253. [DOI](https://doi.org/10.1016/j.jag.2023.103253) [Google Scholar](https://scholar.google.com/scholar?as_q=Adaptive+model+for+the+water+depth+bias+correction+of+bathymetric+LiDAR+point+cloud+data&as_occt=title&hl=en&as_sdt=0%2C31)

[^7]: E. P. Baltsavias, “Airborne laser scanning: Existing systems and firms and other resources,” ISPRS J. Photogramm. Remote Sens., vol. 54, nos. 2–3, pp. 164–198, Jul. 1999, doi: 10.1016/s0924-2716(99)00016-7. [DOI](https://doi.org/10.1016/s0924-2716(99)00016-7) [Google Scholar](https://scholar.google.com/scholar?as_q=Airborne+laser+scanning%3A+Existing+systems+and+firms+and+other+resources&as_occt=title&hl=en&as_sdt=0%2C31)

[^8]: T. Kutser, J. Hedley, C. Giardino, C. Roelfsema, and V. E. Brando, “Remote sensing of shallow waters—A 50 year retrospective and future directions,” Remote Sens. Environ., vol. 240, Apr. 2020, Art. no. 111619, doi: 10.1016/j.rse.2019.111619. [DOI](https://doi.org/10.1016/j.rse.2019.111619) [Google Scholar](https://scholar.google.com/scholar?as_q=Remote+sensing+of+shallow+waters%E2%80%94A+50+year+retrospective+and+future+directions&as_occt=title&hl=en&as_sdt=0%2C31)

[^9]: J. Li, “Range difference between shallow and deep channels of airborne bathymetry LiDAR with segmented field-of-view receivers,” IEEE Trans. Geosci. Remote Sens., vol. 60, 2022, Art. no. 5703616, doi: 10.1109/TGRS.2022.3172351. [IEEE](https://ieeexplore.ieee.org/document/9766335) [Google Scholar](https://scholar.google.com/scholar?as_q=Range+difference+between+shallow+and+deep+channels+of+airborne+bathymetry+LiDAR+with+segmented+field-of-view+receivers&as_occt=title&hl=en&as_sdt=0%2C31)

[^10]: A. Wehr and U. Lohr, “Airborne laser scanning—An introduction and overview,” ISPRS J. Photogramm. Remote Sens., vol. 54, nos. 2–3, pp. 68–82, Jul. 1999, doi: 10.1016/s0924-2716(99)00011-8. [DOI](https://doi.org/10.1016/s0924-2716(99)00011-8) [Google Scholar](https://scholar.google.com/scholar?as_q=Airborne+laser+scanning%E2%80%94An+introduction+and+overview&as_occt=title&hl=en&as_sdt=0%2C31)

[^11]: T. K. Cossio, K. C. Slatton, W. E. Carter, K. Y. Shrestha, and D. Harding, “Predicting small target detection performance of low-SNR airborne LiDAR,” IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 3, no. 4, pp. 672–688, Dec. 2010, doi: 10.1109/JSTARS.2010.2053349. [IEEE](https://ieeexplore.ieee.org/document/5530334) [Google Scholar](https://scholar.google.com/scholar?as_q=Predicting+small+target+detection+performance+of+low-SNR+airborne+LiDAR&as_occt=title&hl=en&as_sdt=0%2C31)

[^12]: M. Amani, “Remote sensing systems for ocean: A review (Part 2: Active systems),” IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 15, pp. 1421–1453, 2022, doi: 10.1109/JSTARS.2022.3141980. [IEEE](https://ieeexplore.ieee.org/document/9677934) [Google Scholar](https://scholar.google.com/scholar?as_q=Remote+sensing+systems+for+ocean%3A+A+review+%28Part+2%3A+Active+systems%29&as_occt=title&hl=en&as_sdt=0%2C31)

[^13]: J. Irish and W. Lillycrop, “Scanning laser mapping of the coastal zone: The SHOALS system,” ISPRS J. Photogramm. Remote Sens., vol. 54, nos. 2–3, pp. 123–129, Jul. 1999, doi: 10.1016/S0924-2716(99)00003-9. [DOI](https://doi.org/10.1016/s0924-2716(99)00003-9) [Google Scholar](https://scholar.google.com/scholar?as_q=Scanning+laser+mapping+of+the+coastal+zone%3A+The+SHOALS+system&as_occt=title&hl=en&as_sdt=0%2C31)

[^14]: T. Kogut and A. Slowik, “Classification of airborne laser bathymetry data using artificial neural networks,” IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 14, pp. 1959–1966, 2021, doi: 10.1109/JSTARS.2021.3050799. [IEEE](https://ieeexplore.ieee.org/document/9320614) [Google Scholar](https://scholar.google.com/scholar?as_q=Classification+of+airborne+laser+bathymetry+data+using+artificial+neural+networks&as_occt=title&hl=en&as_sdt=0%2C31)

[^15]: G. Zhou, “A real-time data acquisition system for single-band bathymetric LiDAR,” IEEE Trans. Geosci. Remote Sens., vol. 61, 2023, Art. no. 5703616, doi: 10.1109/TGRS.2023.3282624. [IEEE](https://ieeexplore.ieee.org/document/10144340) [Google Scholar](https://scholar.google.com/scholar?as_q=A+real-time+data+acquisition+system+for+single-band+bathymetric+LiDAR&as_occt=title&hl=en&as_sdt=0%2C31)

[^16]: K. Guo, “Development of a single-wavelength airborne bathymetric LiDAR: System design and data processing,” ISPRS J. Photogramm. Remote Sens., vol. 185, pp. 62–84, Mar. 2022, doi: 10.1016/j.isprsjprs.2022.01.011. [DOI](https://doi.org/10.1016/j.isprsjprs.2022.01.011) [Google Scholar](https://scholar.google.com/scholar?as_q=Development+of+a+single-wavelength+airborne+bathymetric+LiDAR%3A+System+design+and+data+processing&as_occt=title&hl=en&as_sdt=0%2C31)

[^17]: J. Lillycrop and J. R. Banic, “Advancements in the U.S. army corps of engineers hydrographic survey capabilities: The SHOALS system,” Mar. Geodesy, vol. 15, nos. 2–3, pp. 177–185, Jan. 1992, doi: 10.1080/01490419209388053. [DOI](https://doi.org/10.1080/01490419209388053) [Google Scholar](https://scholar.google.com/scholar?as_q=Advancements+in+the+U.S.+army+corps+of+engineers+hydrographic+survey+capabilities%3A+The+SHOALS+system&as_occt=title&hl=en&as_sdt=0%2C31)

[^18]: X. Ji, B. Yang, Y. Wang, Q. Tang, and W. Xu, “Full-waveform classification and segmentation-based signal detection of single-wavelength bathymetric LiDAR,” IEEE Trans. Geosci. Remote Sens., vol. 60, 2022, Art. no. 4208714, doi: 10.1109/TGRS.2022.3198168. [IEEE](https://ieeexplore.ieee.org/document/9854901) [Google Scholar](https://scholar.google.com/scholar?as_q=Full-waveform+classification+and+segmentation-based+signal+detection+of+single-wavelength+bathymetric+LiDAR&as_occt=title&hl=en&as_sdt=0%2C31)

[^19]: S. Matteoli, L. Zotta, M. Diani, and G. Corsini, “POSEIDON: An analytical end-to-end performance prediction model for submerged object detection and recognition by LiDAR fluorosensors in the marine environment,” IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 10, no. 11, pp. 5110–5133, Nov. 2017, doi: 10.1109/JSTARS.2017.2737645. [IEEE](https://ieeexplore.ieee.org/document/8022939) [Google Scholar](https://scholar.google.com/scholar?as_q=POSEIDON%3A+An+analytical+end-to-end+performance+prediction+model+for+submerged+object+detection+and+recognition+by+LiDAR+fluorosensors+in+the+marine+environment&as_occt=title&hl=en&as_sdt=0%2C31)

[^20]: H. Abdallah, J.-S. Bailly, N. N. Baghdadi, N. Saint-Geours, and F. Fabre, “Potential of space-borne LiDAR sensors for global bathymetry in coastal and inland waters,” IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 6, no. 1, pp. 202–216, Feb. 2013. [IEEE](https://ieeexplore.ieee.org/document/6327381) [Google Scholar](https://scholar.google.com/scholar?as_q=Potential+of+space-borne+LiDAR+sensors+for+global+bathymetry+in+coastal+and+inland+waters&as_occt=title&hl=en&as_sdt=0%2C31)

[^21]: S. Peeri, J. V. Gardner, L. G. Ward, and J. R. Morrison, “The seafloor: A key factor in LiDAR bottom detection,” IEEE Trans. Geosci. Remote Sens., vol. 49, no. 3, pp. 1150–1157, Mar. 2011, doi: 10.1109/TGRS.2010.2070875. [IEEE](https://ieeexplore.ieee.org/document/5599295) [Google Scholar](https://scholar.google.com/scholar?as_q=The+seafloor%3A+A+key+factor+in+LiDAR+bottom+detection&as_occt=title&hl=en&as_sdt=0%2C31)

[^22]: C. K. Chow, S. S. M. Wang, T. Kaneko, and T. R. Perry, “An experimental data acquisition system for ultrasound imaging,” IEEE Trans. Instrum. Meas., vol. IM-28, no. 1, pp. 79–83, Mar. 1979, doi: 10.1109/TIM.1979.4314765. [IEEE](https://ieeexplore.ieee.org/document/4314765) [Google Scholar](https://scholar.google.com/scholar?as_q=An+experimental+data+acquisition+system+for+ultrasound+imaging&as_occt=title&hl=en&as_sdt=0%2C31)

[^23]: D. W. Braudaway, “Uncertainty specification for data acquisition (DAQ) devices,” IEEE Trans. Instrum. Meas., vol. 55, no. 1, pp. 74–78, Feb. 2006, doi: 10.1109/TIM.2005.862014. [IEEE](https://ieeexplore.ieee.org/document/1583865) [Google Scholar](https://scholar.google.com/scholar?as_q=Uncertainty+specification+for+data+acquisition+%28DAQ%29+devices&as_occt=title&hl=en&as_sdt=0%2C31)

[^24]: L. Xingguang, C. Dianren, and C. Lei, “A high-speed data acquisition system based on FPGA,” in Proc. Int. Conf. Test Meas., Dec. 2009, pp. 290–293, doi: 10.1109/ictm.2009.5412935. [IEEE](https://ieeexplore.ieee.org/document/5412935) [Google Scholar](https://scholar.google.com/scholar?as_q=A+high-speed+data+acquisition+system+based+on+FPGA&as_occt=title&hl=en&as_sdt=0%2C31)

[^25]: R. T. Schiffer, M. Flaska, S. A. Pozzi, S. Carney, and D. D. Wentzloff, “A scalable FPGA-based digitizing platform for radiation data acquisition,” Nucl. Instrum. Methods Phys. Res. A, Accel. Spectrom. Detect. Assoc. Equip., vol. 652, no. 1, pp. 491–493, Oct. 2011, doi: 10.1016/j.nima.2010.10.039. [DOI](https://doi.org/10.1016/j.nima.2010.10.039) [Google Scholar](https://scholar.google.com/scholar?as_q=A+scalable+FPGA-based+digitizing+platform+for+radiation+data+acquisition&as_occt=title&hl=en&as_sdt=0%2C31)

[^26]: J. L. Kelley, “Data acquisition, triggering, and filtering at the Auger engineering radio array,” Nucl. Instrum. Methods Phys. Res. A, Accel. Spectrom. Detect. Assoc. Equip., vol. 725, pp. 133–136, Oct. 2013, doi: 10.1016/j.nima.2012.11.153. [DOI](https://doi.org/10.1016/j.nima.2012.11.153) [Google Scholar](https://scholar.google.com/scholar?as_q=Data+acquisition%2C+triggering%2C+and+filtering+at+the+Auger+engineering+radio+array&as_occt=title&hl=en&as_sdt=0%2C31)

[^27]: A. A. Khedkar and R. H. Khade, “High speed FPGA-based data acquisition system,” Microprocessors Microsyst., vol. 49, pp. 87–94, Mar. 2017, doi: 10.1016/j.micpro.2016.11.006. [DOI](https://doi.org/10.1016/j.micpro.2016.11.006) [Google Scholar](https://scholar.google.com/scholar?as_q=High+speed+FPGA-based+data+acquisition+system&as_occt=title&hl=en&as_sdt=0%2C31)

[^28]: A. Veiga and C. Grunfeld, “A modular pipelined processor for high resolution gamma-ray spectroscopy,” IEEE Trans. Nucl. Sci., vol. 63, no. 1, pp. 297–303, Feb. 2016, doi: 10.1109/TNS.2016.2515851. [IEEE](https://ieeexplore.ieee.org/document/7398193) [Google Scholar](https://scholar.google.com/scholar?as_q=A+modular+pipelined+processor+for+high+resolution+gamma-ray+spectroscopy&as_occt=title&hl=en&as_sdt=0%2C31)

[^29]: S. Shu, “A high-speed data acquisition system based on FPGA for tokamak,” Rev. Sci. Instrum., vol. 89, no. 10, Oct. 2018, Art. no. 10K120, doi: 10.1063/1.5035364. [DOI](https://doi.org/10.1063/1.5035364) [Google Scholar](https://scholar.google.com/scholar?as_q=A+high-speed+data+acquisition+system+based+on+FPGA+for+tokamak&as_occt=title&hl=en&as_sdt=0%2C31)

[^30]: J. Adamczewski-Musch, N. Kurz, and S. Linev, “MBSPEX and PEXORNET—Linux device drivers for PCIe optical receiver DAQ and control,” IEEE Trans. Nucl. Sci., vol. 65, no. 2, pp. 788–795, Feb. 2018, doi: 10.1109/TNS.2017.2783043. [IEEE](https://ieeexplore.ieee.org/document/8194862) [Google Scholar](https://scholar.google.com/scholar?as_q=MBSPEX+and+PEXORNET%E2%80%94Linux+device+drivers+for+PCIe+optical+receiver+DAQ+and+control&as_occt=title&hl=en&as_sdt=0%2C31)

### Additional References