1.

Introduction

The most important applications of lasers are in health sciences, engineering, security, and military systems. An increase in the maturity of compact optical and laser devices has improved their abilities for military purposes. Therefore, today’s modern battlefield consists of many laser weapon systems: laser-assisted weapons, which are missiles or projectiles deployed to a target via a laser beam, laser rangefinders (LRFs) for detecting the distance and direction of a potential target,¹ and directed energy (DE) weapons, which have the potential to leave devastating effects on a large area.² One of the advantages of using lasers in these systems is the transmission at the speed of light, which allows laser-dependent weapons to be engaged immediately after target detection. Besides, the DE of the laser provides less collateral damage and can operate in secret.³

Table 1 shows some of the ongoing military laser projects.

Table 1

Projects for demonstrating laser technology in military operations.

Kaushal and Kaddom⁸ scrutinized the use of lasers in military tactical operations.

Reliable laser source detection can minimize these damages and provide accurate information to take action.⁹

The principle of laser threat performance is using a laser beam to increase the chance of a weapon hitting the target. To this end, these threats fall into four classes:

As discussed above, a laser threat from an enemy can come across any instance from any direction and will remain detectable for only 5 to 10 ns.¹² Hence, defense forces provide rapid threat detection and proper response to prevent damage using laser warning system (LWS). This system consists of optical, detection, and processing subsystems and usually displays several parameters of incident radiation, such as the angle of arrival (AoA), wavelength, and power. It must detect and store all the information regarding the laser threat within this minor interval. Many research centers and defense agencies worldwide are developing this system to protect military platforms from laser weapons. Our group in the Nanoptronics Research Center (NRC) at Iran University of Science and Technology (IUST) proposed a new optical subsystem for optimizing the optical array using six lenses for gathering the incident radiation.¹³

The most commonly used lasers in today’s modern battlefield are in the infrared (IR) region, especially near-infrared (NIR). Lasers such as Nd: YAG with a wavelength of 1064 nm, Er: YAG laser, and a Raman shifted Nd: YAG laser with a 1530 to 1550 nm wavelength. Also, ruby laser at 694 nm and carbon dioxide (${\mathrm{CO}}_2$) lasers with a wavelength of $10.6\mu \mathrm{m}$ was reported.¹⁴

Detectors used in LWS are usually based on a semiconductor photodetector (PD) array, typically cryogenically or thermal-electric cooled. Sometimes avalanche photodiodes (APD), photoconductivity (PC), photoelectromagnetic (PEM), or photodiffusion (Dember effect) devices are often use without cooling.¹⁵ APDs have been widely used in systems with high-frequency applications and high-speed detection. With the advancement of quantum telecommunications, three-dimension (3D) radar imaging, weak signal detection, and modern military applications, the need for low-noise and high-sensitivity PDs has become more pronounced. Therefore, Si and AlGaAs-based APDs have been widely used in technical fields, especially in NIR and visible detection systems. Si APDs cover the spectral range of 400 to 1100 nm, and the InGaAs APDs cover 950 to 1550 nm. An APD provides higher sensitivity than a standard photodiode and is for extreme low-level light (LLL) detection and photon counting.¹⁶ Yttrium aluminum garnet (YAG) series detectors are optimized for fast response at 1060 nm, YAG laser light wavelength, and low capacitance for high operating speed and low noise. These detectors are suitable for sensing low-intensity light reflected from laser beams from objects (in ranging applications). The SPOT series of quadrant detectors will be used in targeting applications. These are all n on P devices. These detectors will require reverse bias voltage and work in the PC mode for high-speed applications.¹⁷

Today, AlGaAS and HgCdTe-based PDs are suitable for detecting short-wavelength infrared ($\mathrm{SWIR}\approx 1$ to $3\mu \mathrm{m}$). Besides the standard InGaAs detectors with a cut-off wavelength of $1.67\mu \mathrm{m}$, extended-wavelength (EW) InGaAs detectors have been in high demand in gas sensing, spectroscopy, and food safety. These detectors have been researched and developed for three decades. One of these critical and leading sources of these detectors in the market of aerospace and defense applications in the manufacture and commercialization is Teledyne Judson Technologies. These products are the essential discrete detectors with a cut-off wavelength of 1.9 to $2.6\mu \mathrm{m}$ with a P on n planar structures and in either single elements or linear arrays capable of operating from room temperature to thermoelectrically cooled temperatures.¹⁸

One of the detection structures that has attracted a lot of attention due to the simultaneous advantage of operating at room temperature and fast response time is IR PEM detectors. A variable gap semiconductor, mercury–cadmium–zinc–telluride (Hg-Cd-Zn-Te), with graded composition and doping level profiles helps the manufacturer to achieve the highest possible sensitivity with its IR PEM detectors and adjust the absorption wavelength in a wide range of the IR spectrum (2 to $12\mu \mathrm{m}$). Other advantages of this type of PD include no need for bias, sub-nanosecond response time, lightweight, rugged, and reliability, which makes it a good choice for use in LWSs.¹⁹

Four-quadrant detectors (4-QD) are another suitable device with advantages, such as high light-sensitive areas, fast response, and low noise. The LWS based on this detection structure is highly accurate. Still, its circuit structure is very complex, and obtaining the source of the threat is dependent on the processing algorithm.^20–22

The new class of LWS will also use a multi-wavelength detector. This type of detector has several semiconductor layers subdivided into multi-dielectric screens and focus the laser light into one of the semiconductors (proportional to the wavelength).²³ An active pixel-imaging sensor, integrated readout circuit, and array of detectors could be used in LWS.

Various techniques have been used to detect the position of the incoming laser designator, with varying accuracy levels, from soldier-mounted course detection to multiple discrete detectors positioned strategically on artillery equipment, evolving ultimately to devices, such as Excelitas Technologies’ high angular resolution laser irradiance detector (HARLID). The HARLID combines, within a single TO-8 can, multi-element silicon and InGaAs detector array and encoding mechanism to help to determine the incident angle. Using two detectors enables the detection of the most commonly deployed laser designator wavelengths, from 500 to 1650 nm, on the battlefield.²⁴ We will not focus on incident angle detection methods because it is beyond the scope of this article, but you can refer to Refs. 2526.–27 for further studies. One of the most critical parameters in LWSs is the false alarm rate (FAR). The LWS must be able to detect the real laser threat radiation and not interfere with the background radiations and noises. To avoid this problem, the use of methods, such as optical interferometers^28,29 and photoacoustic design,³⁰ is suggested.

In this paper, we first provide a basic configuration of LWS and examine a model that considers all the influential aspects of laser threat beam to the detection subsystem of LWS. Then, we stated the general theory and evaluation parameters of PDs. Then, we went to the article’s main body, analyzing the sensors used in this system. Finally, in a table, we excavate and compare several previously built plans. It should be noted that the design of a TIA circuit with two stages of voltage amplifier to condition a current of $1\mu \mathrm{A}$ from the output of the PD was simulated and built in the NRC of the Iran University of Science and Technology.

2.

Basic Configuration of LWS and Characteristics

The warning laser beam will diverge in the beam and weaken its original power when passing through the atmosphere and facing atmospheric disturbances. This divergence is an entirely destructive process, making it difficult for the detection subsystem to receive the signal. Therefore, optical signal focusing elements such as focusing lenses are this subsystem’s most crucial element. One of the filtration or threat identification mechanisms is also necessary. It may be an adjustable threshold mechanism, which filters receive noise signals by dynamically adjusting a noise threshold level based on detected ambient light conditions. A correlation filtering mechanism filters received noise signals based on correlation criteria of the timing of received laser radiation by spatially separated laser detectors—the friendly fire elimination mechanism filters detected laser radiation transmitted by a non-threat source. The threat identification mechanism may also be a time mapping mechanism that identifies and characterizes the threat based on pulse mappings of the received laser radiation or a first arrival mechanism that determines the threat direction based on where the received laser radiation arrived first from among multiple laser detectors.¹

After focusing and filtering, the laser beam will enter the detection subsystem to convert the existing optical signal into an electrical signal. Then, the signal will process by the digital signal processor system. In this subsystem, the laser beam’s characterizations, such as laser detection, AoA, wavelength discrimination, and temporal characterization, have been obtained to perform specific countermeasures (Fig. 1).³¹

Figure 2 shows an LWS model that can simulate all aspects of the LBR threat and the laser warning receiver (LWR) scenario. It simulates components of the factors affecting the propagation of laser beams in the atmosphere until they reach the target.

Fig. 1

LWS blocks.¹³

Fig. 2

Configuration of LWS.³² $P_{\text{out}}$, laser irradiator output power; ${\tau }_l$, pulse length; $\theta$, laser irradiator divergence angle; ${\lambda }_0$, radiation wavelength; $a$, transmitter aperture diameter; $R$, distance from transmitter to receiver; $P_{\text{in}}$, laser radiation power at the receiver input; $x$, the size of the laser beam on the target receiver; $D$, receive aperture diameter; $l$, size of PD sensor area; $f$, focal length of receiving the target; $\omega$, receiver FOV; $P_{\mathrm{thr}}$, threshold power; $k_{\mathrm{opt}}$, loss coefficient in optical elements; $T_{\mathrm{abs}}$, attenuation of atmospheric absorption; $T_{\mathrm{sct}}$: attenuation of atmospheric scattering; $\mathrm{\Delta }f$, bandwidth; $T$, temperatures; $\mathrm{\Delta }\lambda$, medium bandwidth filter.

Figure 3 shows the performance of noise filtering, the mechanisms of threat beam detection from non-threat, and its characterization.

Fig. 3

Flow diagram of LWS performance.¹

To evaluate the performance of an LWS, the following parameters are critical:

3.

General Theory of Photodetector

A PD is a critical device in the front end of an optical receiver that converts the incoming optical signal into an electrical signal. Semiconductor PDs are the main types of PDs used in optical communication systems because of their small size, fast detection speed, and high detection efficiency. It is a thick flat piece of homogeneous semiconductor with a factual electrical area $A_e$ and can be coupled to IR radiation by the optical region $A_o$. Optical concentrators increase the rate $A_o/A_e$ (Fig. 4).

Since being developed in the 1960s, it has been used in imaging, sensing, and communication applications.³⁸

To select a suitable PD in an LWS, characterizations such as responsivity ($A/W$), spectral response range and peak (nm), photosensitivity (A/W), detectivity $(\mathrm{cm}.{\mathrm{Hz}}^{1/2}/\mathrm{W})$, response time (ns), dark current (nA), bandwidth (MHz), effective photosensitive size (mm), quantum efficiency (%), reverse breakdown voltage (V), terminal capacity (pF), excess noise index, breakdown voltage temperature coefficient ($\mathrm{V}/°\mathrm{C}$), operating temperature range $(°\mathrm{C})$ and, the noise equivalent power (NEP) $(\mathrm{W}/{\mathrm{Hz}}^{1/2})$ must be considered.

Responsivity ($R$) is an essential characteristic in PDs that quantum efficiency and photoelectric gain play a key role in determining. The photoelectric gain ($g$) is the number of carriers passing through the contact per generated pair, and it is defined simply as the ratio of photoelectron lifetime to transit time. This characteristic indicates how the electron–hole pairs can generate a response current in PDs.

The spectral response of a PD is the range of optical wavelengths or frequencies in which the detector has a significant responsivity. The spectral response obtained from Eq. (2):

Photosensitivity ($S$) quantifies the minimum amount of light needed to produce an electronic signal and is typically wavelength-dependent.

Detectivity ($D^{*}$) of a PD is a figure of merit, defined as the inverse of the NEP. The larger the detectivity of a PD, the more it is suitable for detecting weak signals, which compete with the detector noise. It defines by Eq. (3):

Response time ($t_r,t_f$) of PD is when it takes the detector output to change in response to changes in the input light intensity. It depends mainly on the transit time of the photocarriers in the depletion region, the diffusion time of photocarriers outside the depletion region, and the circuit’s RC time constant.

Dark current ($I_d$) is the current from a PD, which occurs even without light input. It includes the photocurrent produced by the background radiation and the saturated current of the PDs.

Bandwidth (BW) is the difference between the high and low cut-off frequencies measured in Hertz. The bandwidth of the PD is approximately related to the rise time $(0.35/t_r)$.

The effective photosensitive area corresponds to the size of the actual light-sensitive region, independent of the package size.

Quantum efficiency ($\eta$) usually indicates how a detector can couple with electromagnetic radiation and will be defined as the number of electron–hole pairs generated per incident photon.

Reverse breakdown voltage ($V_{\mathrm{BR}}$) is the level of reverse voltage cause the breakdown and deterioration of the detector.

The terminal capacity ($C_t$) is the capacitance between the cathode and anode terminal of PD.

The excess noise factor ($x$) is a function of the carrier ionization ratio, $k$, where $k$ is usually defined as the ratio of the hole to electron ionization probabilities ($k<1$).

$V_{\mathrm{BR}}$ temperature coefficient ($\mathrm{\Gamma }$) describes the relative change of a physical property associated with a given change in temperature.

The operating temperature ($T$) is the ambient temperature range in a PD operation.

NEP is the amount of incident photon energy equivalent to the intrinsic noise level of the device, providing a signal-to-noise ratio (SNR) of one $\left(\frac{\text{noise current}(\mathrm{A}/\sqrt{\mathrm{Hz}})}{\text{responsivity at}{\lambda }_p(\mathrm{A}/\mathrm{W})}\right)$.

An optimized detector should include the following:

The above conditions may apply in heterojunction such as $P^{+}\text{-}n\text{-}{n}^{+}$ and $N^{+}\text{-}p\text{-}{p}^{+}$ with heavily doped contact regions. Homojunction devices such as $n\text{-}p,n^{+}\text{-}p,p^{+}\text{-}n$ suffer from the excess thermal generation problem, which increases the dark current and recombination, reduces the photocurrent. The total generation rate is obtained from thermal and optical generation rates:

4.

Laser Beam Detection Based on PN Photodiodes

PN photodiodes are semiconductor devices that include a p-n junction and, in some cases, an intrinsic (undoped) layer between the p-n layers called p-i-n. Light is absorbed in the depletion or intrinsic region and generates an electron holes pair, most of which will participate in the photocurrent. The photocurrent can be proportional to the intensity of the incident light. Due to the small depletion region width in PN photodiodes, fewer photons will be absorbed, so the diode current is low, and receiver sensitivity is not optimal. On the other hand, because the width of the junction’s depletion region depends on the reverse voltage, with the change in voltage, the width of the depletion region, and consequently the amount of photon absorption changes, this is not our desired dependence. In p-i-n photodiode, the depletion region is independent of bias voltage and is a low concentration doped region. PIN PDs are widely used in optical communications due to their low noise and high speed.

Some p-i-n photodiodes are silicon-based, and their sensitivity extends from visible to NIR (up to $1\mu \mathrm{m}$). Their absorption and photoresponse efficiencies decrease sharply at longer wavelengths; nevertheless, their cut-off parameters depend on the thickness of the intrinsic region. InGaAs p-i-n photodiodes are available for wavelengths up to 1.7 or $2.6\mu \mathrm{m}$ in the extended spectral response. However, they are much more expensive especially for PDs with a larger active area. Germanium-based photodiodes can be a good alternative, but they do not respond quickly due to large parasitic capacity and high dark current. Table 2 briefly compares the Thorlab Company’s p-i-n photodiodes based on silicon, InGaAs, and germanium.

Table 2

Brief comparison of Si, InGaAs, Ge based photodiode made by Thorlab company.40

From the comparison in Table 2, it is clear that Ge photodiodes are not suitable for detecting laser threats (especially LBR threats) due to their slow rise/fall time and high dark current.

In 2013, Ghuson H. Mohammed et al.⁴¹ reported developing a p-i-n photodiode for ${\mathrm{Nd}}^{+3}$-YAG laser detection. In their proposed structure, the detector consisted of three layers $(n^{+}\text{-}\pi \text{-}{p}^{+})$ with a deep junction fabricated on a wafer with a thickness of 1 mm and a large diameter. It was observed that the peak spectral responsivity of the detector was about $1.5\mathrm{A}/\mathrm{W}$ at a wavelength of $1.06\mu \mathrm{m}$. One of the most significant features of their device is working on 170 to 180 operating reverse voltage. It should be noted that the response time of this detector is about 85 ns, which can be suitable for laser threat detection applications (except LBR threat).

In 2015, Urchuk et al. reported simulation results from the spectral sensitivity characteristics of silicon p-i-n photodiodes. It was carried out the various factors influence such as the electrons and holes lifetime and mobility, surface recombination velocity on the spectral sensitivity characteristics of p-i-n-structures. To optimize the PD structure parameters, it was calculated when changing the top layer depth of p-doping levels and changing doping areas of the structure. To expand the spectral sensitivity characteristics of silicon PD structures toward shorter wavelengths, the maximum achieving space charge region is required to the surface. At the same time, action is needed to ensure the improvement of the surface properties to reduce the surface recombination velocity and the simultaneous use of anti-reflection coatings.⁴²

In 2019, Emre Doganci et al.⁴³ Examined the characteristics of silicon ${\mathrm{p}}^{+}-\mathrm{i}-{\mathrm{n}}^{-}$ photodiodes with $3.5\times 3.5$, $5.0\times 5.0$, or $7.0\times 7.0{\mathrm{mm}}^2$ active area. Voltage–current (V-I) and capacitor–voltage (C-V) measurements were performed in the photoconductive (PC) mode to achieve the device characterizations. Both measures were carried out in a dark environment at room temperature. The measured values of the dark current ($I_{\mathrm{dc}}$) and the capacitance of photodiodes were $-6.97$ to $-19.10\mathrm{nA}$ and 23 to 61 pF at $-5\mathrm{V}$, respectively. The quantum efficiency measurements of the devices increased up to 66%. Peak responsivity was found to be $0.436\pm 1\mathrm{mA}/\mathrm{W}$ at 820 nm. The results show that $I_{\mathrm{dc}}$ and device performance are improved, and therefore, this device can be used for optical electronics applications and commercial use. Experimental results showed that breakdown voltages are $-93$, $-84$, and $-77.5\mathrm{V}$ for $7.0\times 7.0$, $5.0\times 5.0$, and $3.5\times 3.5{\mathrm{mm}}^2$ active areas, respectively.

Of all the detectors used in the SWIR region, InGaAs has the best performance due to its high efficiency and low dark current at room temperature. This type of PD can be very suitable for imaging through foggy weather in this spectral range because it can have the maximum possible detectivity. Achieving maximum detector performance for many applications depends on the proximity of the detector bandgap energy to the light intensity energy. However, the PIN detector, which has the advantage of simplicity and the absence of geometric absorption constraints, will be matched to the typical substrate by the lattice. This often means that the bandgap energy is too low or not optimal for the desired wavelength, and the energy difference will usually cause excess noise.⁴⁴

As mentioned in the previous sections, one of the most typical sources of laser threat is Nd: YAG laser with a wavelength of 1064 nm and, in some cases, Yb: YAG laser with a wavelength of 1030 nm.¹⁴ However, there is a lack of high performance (low noise, high responsivity, high speed) PDs that can be operated at either of these wavelengths. Infrared-enhanced silicon PDs have a long wavelength cutoff at or below 1030 nm. The indirect bandgap of silicon necessitates very thick absorption layers to achieve even modest efficiencies, limiting these detectors’ operating speed, resulting in a low bandwidth-responsivity product. ${\mathrm{In}}_{0.53}{\mathrm{Ga}}_{0.47}\mathrm{As}$ PDs (compatible with InP substrates) can achieve very high quantum efficiencies and cutoff frequencies but require growth on InP substrates, which are smaller, more expensive, and more fragile, and require a less mature process than GaAs substrates.⁴⁵

Although Ge-based PDs can achieve very high speed due to the high mobility of the hole but suffer from high dark currents and low shunt resistors due to a very small bandgap. As a result, detectors based on this material will be very noisy at the desired wavelength. Dilute nitride semiconductors, such as InGaNAs and InGaNAsSb, are also suggested due to their ability to adjust the bandgap. Still, these compounds reduce the performance of PDs due to the relatively low quality of materials and are not suitable. ${\mathrm{In}}_x{\mathrm{Ga}}_{1-x}\mathrm{As}$ With low composition ($x<0.25$), metamorphic grading can ideally meet our needs due to the direct and adjustable bandgap.

In 2011, Swaminathan et al.⁴⁶ reported the growth and performance of a top-illuminated metamorphic ${\mathrm{In}}_{0.2}{\mathrm{Ga}}_{0.8}\mathrm{As}$ PIN PD on a GaAs substrate using a graded InGaAs buffer. This PD showed a very low dark current $(1.4\times {10}^{-7}\mathrm{A}/{\mathrm{cm}}^2)$ in $-2\mathrm{V}$ reverse bias at room temperature. They showed responsivity and special detectivity values of $0.72\mathrm{A}/\mathrm{W}$, $2.3\times {10}^{12}\mathrm{cm}·{\mathrm{Hz}}^{\frac{1}{2}}/\mathrm{W}$ and $0.69\mathrm{A}/\mathrm{W}$, $2.2\times {10}^{12}\mathrm{cm}·{\mathrm{Hz}}^{\frac{1}{2}}/\mathrm{W}$ are achieved for Yb:YAG (1030 nm) and Nd:YAG (1064 nm) laser wavelengths at $-2\mathrm{V}$, respectively. At a wavelength of 1064 nm, the product of bandwidth responsivity was obtained 0.21 GHz from theoretical analysis. Their work was a turning point for adjustable bandgap gallium detectors that could grow on the gallium substrate and be used in many practical applications.

In 2015, Michael Verdun et al.⁴⁷ studied the dark current components of thin-film InGaAs PIN photodiodes grown by metalorganic vapor-phase epitaxy for optical nano-resonators. Due to the electric field enhancement, the nano-resonators made it possible to significantly reduce the thickness of the active region up to 100 nm while maintaining the quantum efficiency of the PD. In this study, to cover a broad spectral band, they combined several resonance peaks induced by guided-mode resonances in a given spectral range. This type of geometry allowed them to place the InAlAs alloy at the edge of the thin InGaAs active region to reduce the generation/recombination and diffusion currents significantly. In this design, it is observed that the components of tunnel dark current increase with decreasing the thickness of the active layer and dominate the reverse dark current. By optimizing the epitaxial stack, while keeping its total thickness constant (the optical properties of the nano-resonator remained unchanged), they showed that they could achieve a specific detectivity of up to $1\times {10}^{13}\mathrm{cm}\sqrt{\mathrm{Hz}}{\mathrm{W}}^{-1}$ for $\lambda =1.55\mu \mathrm{m}$.

In 2019, Rutz et al.⁴⁸ reported using a planar process technology from InGaAs/InP-based PIN photodiodes to fabricate SWIR cameras with $640\times 512\text{pixels}$ capable of capturing LLL. One of the most arrogant features of this structure was the dark current densities below ${10}^{-7}\mathrm{A}/{\mathrm{cm}}^2$ at room temperature.

Optimizing the performance characteristics (response time, dark current, and responsivity) of III-V material-based PDs has always been one of the most popular topics for scientists and researchers. One of the suggested solutions is to use nanowires based on III-V group materials due to their one-dimensional morphology, direct and tunable bandgap, and unique optical and electrical properties. Li et al.⁴⁹ studied the latest status of this type of PD.

5.

Laser Beam Detection Based on Avalanche Photodiodes

There have been extensive studies, optimization, construction, and commercialization of APDs in educational, industrial, and military centers.⁵⁰ Compared to p-i-n photodiodes, APDs will have more considerable gain, higher sensitivity, and lower detection limit. For this reason, they show better performance in optical communication. The urgent need for advanced 3D imaging in laser detection and ranging (LADAR) applications have led to the use of this kind of PD structure in focal plane arrays (FPA).^51–53 Using these detectors in FPAs will have advantages, such as internal photoelectric gain, small size, low driving voltages, high efficiency, and fast response, making it possible to create 3D imaging methods and, subsequently, more detailed information to identify the object. Lincoln Laboratory at the Massachusetts Institute of Technology (MIT) is one of the most important centers that has made significant progress in this field. They improved the most advanced and state-of-the-art Si FPAs, the Giger-mode InGaAs, and applied them to some LADAR systems. Another center that has done extensive research in this area and achieved good results is Princeton Lightwave, which has succeeded in producing FPAs based on single-photon APDs and has used it in commercial cameras.

5.1.

Linear-Mode Si APDs

The operation of LADAR imaging systems is based on the spatial or temporal sampling of information from optical beams to an array of detectors. LADAR systems often desire linear-mode (applied bias slightly lower than the breakdown voltage) APDs because their slow time is usually much shorter than that of Geiger-mode (applied bias somewhat higher than the breakdown voltage) APDs can measure sequential pulse returns from closely spaced multiple objects. In extreme cases, linear-mode APDs can even detect a few photons or a single photon, adding an extra dimension to LADAR scene data. In general, in the NIR spectral band, especially at 905 nm, Si APDs might be applied for ultra-weak light detection and used in linear-mode at gains up to about 500 or greater. Therefore, linear-mode Si APD arrays were developed and involved in LADAR systems.

As shown in Fig. 5, a simple linear mode APD detector will consist of an APD element (convert the incident light into photo-generated carriers and photocurrent and then amplify the resulting photocurrent through the avalanche gain by impact ionization) and a readout integrated circuit (ROIC). The ROIC consists of a trans-impedance amplifier (TIA) (converts the amplified current of the APD to a voltage signal that is proportional to the total multiplied charge delivered by the APD), a stabilivolt source circuit, and a comparator.⁵³

Fig. 4

Schematic illustration of a general PD.³⁹

Fig. 5

Schematic of linear-mode Si APD.⁵³

It is important to note that in silicon; much more electrons can be ionized than holes. Therefore, instead of holes, electrons should be swept in regions with a larger electric field where multiplication occurs, which is done by the electric field.

Therefore, there should be a $\pi$-type absorption region with a suitable width for absorbing incident beams. The radiation should be able to enter this region without loss in any $n^{+}$ layer (Fig. 6).

6.

Laser Beam Detection Based on Photoelectromagnetic Detectors

One of the most critical sources of laser threats on modern battlefields is the use of lasers at a wavelength of $10.6\mu \mathrm{m}$. Undeniable LWIR detectors are needed to diagnose and characterize this type of threat. In addition to PC detectors and photodiodes, two types of contactless devices will be used for uncooled IR detectors, such as PEM and Dember effect detectors. These detectors will be used in high-speed response applications.¹⁹

Series of PEM detectors work on the effect of PEM on semiconductors. (The generation of an electric current in an intermetallic semiconductor in a magnetic field by the action of light.) These devices are usually optimized for the best performance at $10.6\mu \mathrm{m}$. The detector consists of an MCT-based active element with an engineered bandgap with a selective composition and selective doping profile and a small permanent magnet to generate a magnetic field. Exhibiting no flicker noise will allow these detectors to be used in CW detection and the low frequency modulated radiation in the range of 2 to $11\mu \mathrm{m}$ simultaneously. The PEM effect was used primarily for InSb room temperature detectors in the MWIR and LWIR bands for a long time. However, the uncooled InSb devices with the cutoff wavelength at $\approx 7\mu \mathrm{m}$ exhibits no response in the 8 to $14\mu \mathrm{m}$ atmospheric window relatively average performance in a 3 to $5\mu \mathrm{m}$ window. ${\mathrm{Hg}}_{1-x}{\mathrm{Cd}}_x\mathrm{Te}$, ${\mathrm{Hg}}_{1-x}{\mathrm{Zn}}_x\mathrm{Te}$, and ${\mathrm{Hg}}_{1-x}{\mathrm{Mg}}_x\mathrm{Te}$ alloys make it possible to use PEM detectors at any specific wavelength.⁸²

The diffusion of photogenerated carriers will create the effect of PEM due to the photo-induced carrier in depths concentration gradient and by the deflection of electron and holes paths in different directions by the magnetic field (Fig. 15). If the two ends are open-circuit in the $X$ direction, a space charge will be created to increase the electric field along the $X$-axis (open circuit voltage).

If the two ends are short-circuited in the $X$ direction, a current will flow across the short circuit (short circuit current). Unlike PC and PV devices, PEM photovoltage (photocurrent) generation does not require simple photogeneration but instead forms an in-depth gradient of photogenerated carriers. Usually, this will be done by producing nonhomogeneous optical generation due to the radiation absorption in the area near the part’s surface. The advantages of this detection structure can be:

One of the PEM structures used to detect $10.6\mu \mathrm{m}$ radiation is based on a single crystal with an $x$ composition of 0.16 to 0.17, and $Eg=0.1-0.11\mathrm{eV}$ at 300 K. In 2004, Gaziyev and Huseynov reported the results of an improvement and review of an MCT-based uncooled PEM detector of $x=0.2$ and a bandgap of 0.155 eV for the medium IR range (3 to $7\mu \mathrm{m}$).⁸³ The calculated results show the specific voltage response and detection versus the doping surface of the semiconductor for this detection structure. The optimal level of an acceptor doping of the semiconductor is about $p=5-6n_i$ for an uncooled PEM detector. They showed that the voltage responsivity and the specific detectivity are $R_{\lambda \max }\approx (0.8-1.0)\mathrm{V}/\mathrm{W}$ and $D_{\lambda \max }^{*}=\approx (0.8-0.9){10}^8\mathrm{cm}{\mathrm{Hz}}^{1/2}{\mathrm{W}}^{-1}$, respectively, and the sensitivity makes about 60% of the maximum responsivity of $\lambda \approx 6.7\mu \mathrm{m}$.

In 2011, Jafarzadeh and Hazrati optimized a special detector from an uncooled MCT-based PEM detector for a wavelength of $10.6\mu \mathrm{m}$ and an atmospheric window of $4\mu \mathrm{m}$. They solved the carrier transfer equations with the Poisson equation in an external magnetic field in a self-consistent process. For $10.6\mu \mathrm{m}$, the obtained optimum values of Cd molar fraction ($x$), applied magnetic field ($B$), the thickness of semiconductor ($d$), and the acceptor doping ($Na$) are $x=0.165$, $B=1.5\mathrm{T}$, $d=4.2\mu \mathrm{m}$, $Na=1.4\times {10}^{23}{\mathrm{cm}}^{-3}$, respectively. By taking these values into account, the maximum value of specific detectivity is ${3.5910}^7\mathrm{cm}.{\mathrm{Hz}}^{1/2}{\mathrm{W}}^{-1}$. Also the responsivity ($R_v$) and response time (${\tau }_r$) are $0.28\mathrm{V}/\mathrm{W}$ and 1.2 ns. For the wavelength of $4\mu \mathrm{m}$, the maximum value of detectivity is and aforementioned values are: $x=0.308$, $d=25\mu \mathrm{m}$, $Na=1.9\times {10}^{22}{\mathrm{cm}}^{-3}$, $=12.14\mathrm{V}/\mathrm{W}$, $=0.369\mu \mathrm{s}$.⁸⁴

Figure 16 shows the variation of the detector parameters versus the acceptor concentration. Initially, these parameters raise with increasing acceptor concentration and will reach a maximum value. Further increase of this doping will reduce these parameters, and the reason is the changes of Auger 1 and Auger 7 with the acceptor concentration.

Fig. 6

(a) Schematic cross-section and (b) profile of electric field of typical APD.⁵³

Fig. 7

$64\times 1$ linear mode Si APD FPA.⁵³

Fig. 8

One chip of Si SPAD array.⁵³

Fig. 9

Epitaxial layer structure of an SCM APD with 10 gain stages.⁶³

Fig. 10

Schematic of MWIR APD structure under top illumination.⁶⁵

Fig. 11

Responsivity (A/W) of Si, GaAs, and InGaAs APDs structures at different levels of temperature.⁶⁶

Fig. 12

Photocurrent of Si, GaAs, and InGaAs APDs for operating wavelengths 1300 and 1550 nm at different radiation power incident.⁶⁶

Fig. 13

Bandgap and the corresponding wavelength for alloy compounds ${\mathrm{Hg}}_{1-x}{\mathrm{Cd}}_x\mathrm{Te}$ as a function of temperature as calculated from Eq. (1).⁶⁹

Fig. 14

Schematic illustration of the HDVIP HgCdTe APD: (a) side view and (b) top view.⁷¹

Fig. 15

Schematic of PEM effect.⁸²

Fig. 16

Specific detectivity, responsivity, noise, and response time versus acceptor concentration for $4\mu \mathrm{m}$ wavelength⁸⁴ ($x=0.308$, $l=w=1\mathrm{mm}$, $d=25\mu \mathrm{m}$, $B=1\mathrm{T}$).

Fig. 17

Calculated normalized resistivity (RA), normalized responsivity ($R_{v}A$), detectivity ($D^{*}$), and bulk recombination time ($\tau$) of an uncooled $10.6\mu \mathrm{m}$ ${\mathrm{Hg}}_{1-x}{\mathrm{cd}}_x\mathrm{Te}$ Dember detector as a function of the acceptor concentration.⁸²

Fig. 18

Quadrant PD.²⁷

Fig. 19

Layer structure of IRPD.⁹⁹

7.

Laser Beam Detection Based on Photodiffusion (Dember Effect) Detectors

Another structure used in laser threat detection is Dember effect detectors. This type of detector is a type of PV device based on bulk photodiffusion voltage in a simple structure with only one kind of semiconductor doping supplied with two contacts. When radiation strikes the surface of a semiconductor, electron and hole pairs are produced. Because of the difference in the diffusion of electrons and holes, a potential difference usually occurs in the radiation direction. The electric field in the Dember effect prevents electrons with higher mobility while the holes accelerated. Hence, both fluxes will be equal. Theoretical design of and practical Dember effect have been reported. The best performance is achieved for a device with a thickness slightly larger than the ambipolar diffusion length. More thin devices show a lower voltage responsivity, while thicker devices have excessive resistance and significant related Johnson noise. Figure 16 shows the responsivity, detectivity, and bulk recombination time for an optimized thickness Dember detector. One of the exciting aspects of this device is its impressive photoelectric gain, more considerable with a zero bias condition (Fig. 17).

By using the interference effect, the Dember effect detectors can be achieved faster and with better performance. The use of optimized optical resonance cavities for uncooled Dember detectors can increase the detection by five times for non-interfering devices. HgCdTe Dember detectors have found application in high-speed laser beam diagnostics of industrial ${\mathrm{CO}}_2$ lasers and other applications that require fast operational speeds.⁸²

8.

Quadrant Photodetector in Laser Warning System

Obtaining the angle of incidence (AOI) radiation is one of the crucial expectations from LWSs. Different ideas will be used to achieve this importance. One of these ideas is using 4-QD. A QD consists of four active detection areas separated by a small gap on a single chip (Fig. 18). This detector can measure small changes in the location of the incident beam and will be used to detect the location of laser beams and position displacements.

In a 4Q-based laser beam detection and tracking system, the detector is usually placed on the focal length of imaging lenses to measure the laser spot displacement.⁸⁵ One of the disadvantages of beam detection methods by 4-QDs is the blind area at the photosensitive surface, affecting the position accuracy.²⁰ Of course, when this photosensitive surface is small enough, this blind area is negligible. Otherwise, the blind area cannot be easily ignored. In LWSs based on 4-QDs, received power ratios and associated photocurrents allow evaluating input radiation angle. This solution suffers from disadvantages, such as low and limited FOV, low angular accuracy, and strong radiation beam distribution dependence.²⁷

Usually, in many applications, conventional algorithms will be used to detect the spot location in 4-QDs from an ideal circular spot with a uniform energy distribution, which has the advantage of simplicity in calculations. However, in reality, the energy distribution in a circular mode is not ideal and is usually Gaussian. Hence, the Gaussian radius and the centroid of the measuring spot are affected accurately in actual measurements. Many studies have improved the detection algorithm and location of these types of detectors.²² Zhang et al.⁸⁶ proposed that the detection accuracy of a QD was related to the spot radius, spot position, and the QD output SNR. Wu et al.⁸⁷ analyzed the nonlinear characteristic of a QD and proposed a linear correction method based on the Boltzmann function. Gao⁸⁸ calibrated the absolute spot position and the ratio of the photocurrent output from a QD and then established a database. However, establishing a database requires a large amount of data, which involves many hardware resources.

In 2018, Jun Zhang et al. improved the spot energy distribution model to fit the elliptical Gaussian distribution. The width of the blind area has been added to the detector response model so that the output of each quadrant and the tracking algorithm error can be calculated correctly. Their simulation results show that the accuracy of quadrant measurement decreases with increasing blind area, shape, and point of slight deviation.⁸⁵

On the battlefield, the protection of moving assets (war machines) will be an important goal. These assets must have an LWS based on a laser tracker. In 2019, Wugang Zhang et al. proposed an improved method for detecting point location in a laser tracker and location system by a 4-QD. This method will use a 4-QD to receive the spatial information of the laser point, which will be returned to the moving target and adjust a 2D galvanometer to change the laser direction. Then the tracking function will be achieved.⁸⁹

9.

Laser Warning System Based on Multi-Color Infrared Photodetector

In the last few years, one of the areas that attracted researchers’ interest in improving the detector’s performance, especially for detecting objects and imaging under varying atmosphere conditions, was multicolor detectors. Since different wavelength regions are set aside for a particular empirical application, exposing an IR beam of an object at different wavelengths can be used to reduce the number of false positives.⁹⁰ The performances of two-wavelength detectors in the terahertz and mid-IR regions are inspiring.⁹¹ Several processes enable us to have a multi-wavelength response. One of these approaches is to construct several stacks of square QW with different peak responses.^92–94 Other approaches have used asymmetric or coupled QW structures to transition from zero states to several excited states.^95,96 In this approach, only a set of QW is needed, making the manufacturing process more straightforward. In this case, photogenerated carriers will be extracted by controlling the bias voltage across the device.

Ye and Campbell⁹⁷ proposed a multi-wavelength IRPD, which is controlled with bias. Its active area consists of five layers of InAs quantum dots (QDs) with InGaAs cap layers. The photoresponse peak is observed at 5.5, 5.9, 8.9, 10.3, and $10.9\mu \mathrm{m}$. The applied bias will control the relative amplitude of these peaks. For $5.9\mu \mathrm{m}$ detection, a peak detectivity, $D*$, of $5.8\times {10}^9\mathrm{cm}{\mathrm{Hz}}^{1/2}\mathrm{W}$ at 77 K and 0.3 V was achieved.

Ali Rostami et al. presented a structure based on the quantum cascade, and it can detect two different wavelengths simultaneously across two independent current paths. This structure consists of two paths (right and left) that can detect any specific wavelength in its active region.⁹⁸

Nguyen et al. presented spectral response characteristics of a dual-band IRPD based on nBn structure with GaSb and InGaAsSb absorbent layers and a ternary layer of AlGaSb, which acts as a unipolar barrier layer and is independent access to both sides (Fig. 19). The results show the spectral response in the short-range IR region, especially in the wavelength of 1.6 and $2.65\mu \mathrm{m}$, which is voltage bias dependent. $D^{*}$ values are $2.3\times {10}^{12}\mathrm{cm}.{\mathrm{Hz}}^{1/2}.{\mathrm{W}}^{-1}$ (at $1.5\mu \mathrm{m}$) and $2.1\times {10}^{11}\mathrm{cm}.{\mathrm{Hz}}^{1/2}.{\mathrm{W}}^{-1}$ (at $2.25\mu \mathrm{m}$) at 300 K.⁹⁹

Yao Zhai et al. proposed a dual-band PD in the mid and long IR regions. This PD is capable of voltage-controllable detection. This structure is based on multiple stacks of sub-monolayer (SML) QDs and self-assembled QDs. The detection band in MWIR, or LWIR, or both with high photodetectivity and low crosstalk between the bands will be achieved by varying the detector voltage bias. The schematic of this structure is shown in Fig. 20.

Fig. 20

(a) Schematic structures of the hybrid QDIP. (b) Conduction band structure of the SML and self-assembled QDIP.¹⁰⁰

Fig. 21

Distribution of image elements array and laser spot.¹⁰³

Fig. 22

Block diagram of laser beam profile analysis.¹⁰⁴

Fig. 23

Illustration a block diagram of a panoramic LWR.¹⁰⁵

Fig. 24

Principle of sinusoidal grating LWR.¹⁰⁶

Fig. 25

Designed TIA circuit.¹⁴³

Fig. 26

Illustration of the dynamic range in the designed circuit. (a) The correct output value when the DC level is 22 mA and (b) incorrect output value when the DC level is 23 mA.¹⁴³

Fig. 27

Amplification stages.¹⁴³

Fig. 28

3D illustration of front-end designed circuit.¹⁴³

Fig. 29

(a) Output waveform, (b) circuit bandwidth, and (c) total noise of the circuit.¹⁴³

As we can see, this structure consists of 10 layers of SML QDs and 10 layers of self-assembled QDs sandwiched between the top and bottom contact layers. The electrons will be collected by the top and bottom electrodes, forming a photocurrent. The QDs sizes and the energy levels are unalike due to different strains informing. Thus, allowing them to detect various bands. By optimizing the QD growth conditions, such as substrate temperatures, growth rate, and the V to III ratios, the detection of the MWIR and the LWIR bands will be achieved. The growth of strain-driven band structure engineering offers an alternative way to achieve dual-band PDs.¹⁰⁰

10.

Image Sensors Based Laser Warning System

The use of imaging elements in LWSs is also prevalent. Imaging LWSs will include fisheye lenses to provide a high FOV, narrowband spectral filters, and imaging components.¹⁰¹ Because the imaging element is so small, it can accurately orient the laser threat. In the past decade, due to the limitations of CMOS technology, CCD cameras have been used more for LWSs. With the advancement of CMOS technology, these cameras took advantage of advantages, such as high integration, low power dissipation, high speed, and good radiation resistance, which led to the use of these cameras in this application.¹⁰² One of the challenges in CMOS-based LWSs is the effect of the fill factor on the exact location of the laser spot. When the laser spots fall on the CMOS element, the non-photosensitive area of the CMOS element will cause the laser spot to be missing.

If we consider a laser beam that is an ideal point (Fig. 21) (impact function) on the focal plane through the lens of the fisheye and the laser spot distribution, we have the following relation:

In addition, we assume that the CMOS imaging device consists of two parts: photo-electricity conversion (photosensitive) and electro-circuit control (non-photosensitive). In that case, the fill factor is the ratio of the photosensitive area to the whole imaging element. It should be noted that the fill factor for CMOS is between 30% and 70%.

In 2009, Jiaju Ying et al. analyzed the effect of the fill factor on the exact location of a laser spot in an LWS. The locating standard deviation was performed by Monte Carlo simulation, and an experiment was performed to confirm this simulation. It has guiding value for analyzing the orientation precision and selecting the parameters of the CMOS sensor and fisheye lens, when imaging LWS is built up.¹⁰³

Yas A. Alsultanny reported the analysis of helium–neon (He-Ne) laser profiles and a laser diode. He showed that the He-Ne laser emits a pure Gaussian beam while the laser diode emits elliptical shape beam. This report analyzes the intensity of distribution, laser power, and the number of possible modes necessary for many applications. Image profile analysis can determine the distance between objects, depending on the laser beam line distribution. The block diagram of this setup is shown in the figure below (Fig. 22).¹⁰⁴

In 2016, Alistair developed a method and device for detecting lasers in an LWR. In this invention, a set of panoramic lenses used in cameras is combined with a laser focal plane (Fig. 23). Collision laser light is refracted in panoramic lenses to create several single sensor elements. By determining the corresponding intensity of the laser light on the sensors, the angle of arrival (AoA) resolutions superior to the part angular resolutions can be achieved. Combining a panoramic lens with a laser detection focal plane provides a low-cost laser warning for wrap-around ground-based situational awareness.¹⁰⁵

Some LWS configurations use the diffraction grating method to measure wavelength and AOI. The advantage of using diffraction grating is the simultaneous measurement of wavelength and incident angle. Figure 24 shows that this receiver will consist of a linear CCD camera, a plane-convex cylinder lens, and a sinusoidal grating. The beam strikes the grating at $\lambda$ wavelength and $\alpha$ angle. After scattering by the grating and focusing on the lens, two diffraction patterns will be formed. These two diffraction patterns will be converted into an electrical signal by the CCD camera and then into the processing unit. See Ref. 106 for more information.

11.

Nanostructures to Enhance the Electronic Characteristics of Infrared Detectors

Imagers and IR sensors using nanostructure-based materials are being advanced for various defense applications.¹⁰⁷ Engineering the electronic characteristics of the absorbent material to reach high quantum efficiency and detectivity is one of the methods to enhance the performance of PDs (the product of bandwidth detectivity). Increasing the light coupling to electronic states of absorbent material is another method. Both of these methods can be implemented using micro and nanofabrication techniques.⁸¹

12.

Nanostructures to Enhance the Light Coupling in Infrared Detectors

By increasing the radiation flux coupling to the optical sensing medium (area), IRPD’s performance can be improved. There are several ways, such as a reflective concentrator, refractive concentrator, and anti-reflective coating (AR), by which this importance can be achieved.

As we know, most semiconductors that are used for IRPDs have an extensive refractive index of about 3 to 4. Hence, anti-reflex structures are essential since about 25% to 40% of the losses are due to the reflection of the surface. Typically, the most superficial AR coating can be made by depositing one or several optically thin films on the surface utilizing interference effects to cancel the back-reflection.¹³³ However, this method also has limitations in that the reflection varies by changing the incident angle. Another limitation of this approach is the narrow spectral response band that provides weak coating layers in new multi-color and multi-band IR detectors. Usually, anti-reflection conditions create by placing slow-wave structures against the detector.

13.

Conclusion

Due to security threats and electromagnetic interference affecting RF, the use of visible and IR beams using lasers has become an integral part of today’s battlefields. Low divergence and laser beam coherence provide a secure data transmission. Laser threats can use any wavelength in the IR region. The most common will be in the NIR range and can be detected at short intervals (about 10 ns). Hence, as one of the subsystems of LWSs, IR detectors must have the best performance (response time, low dark current, and responsivity). Depending on the azimuth and horizontal coverage axis and the accuracy of the angular resolution, it is possible to choose how to configure the PDs (singular mode or 2D arrays of detectors vertically and horizontally).

PIN detectors are one of the most common detectors to be used in this system and have the advantage of high speed and low noise. It was shown that the various factors influence the electrons and holes lifetime and mobility, surface recombination velocity on the spectral sensitivity characteristics of p-i-n-structures. The breakdown voltage in the detector will increase as the active area increases. In the SWIR region, InGaAs has the best performance due to their high efficiency and low dark current at room temperature. PIN PD on a GaAs substrate using a graded InGaAs buffer showed a very low dark current. The product of bandwidth-responsivity was obtained 0.21 GHz from theoretical analysis, which is very suitable for this particular application. For narrow-band gap materials such as InGaAs, high tunneling current limits their efficiency. InGaAs APDs performed better than other structures at higher SNR and lowered BER at higher temperatures. It has been observed that reverse bias voltage, incident beam power, an optical signal wavelength affect the quality and performance of the PD significantly.

The adjustable bandgap, direct bandgap with a high absorption coefficient, average dielectric constant and index of refraction, the average CTE, and availability of wide bandgap lattice-matched substrates for epitaxial growth made MCT suitable material for IR detectors. PEM and Dember effect detectors are usually optimized for the best performance at $10.6\mu \mathrm{m}$. The blind area at the 4-QD photosensitive area will affect the position accuracy. By adding the width of the blind area to the detector response model, each quadrant’s output and the tracking algorithm’s error are calculated correctly. Two methods to have multiwavelength PD are several stacks of square QW with different peak responses and asymmetric or coupled QW structures to transition from zero states to several excited states. With a linear CCD camera, a plane-convex cylinder lens, and a sinusoidal grating simultaneous measurement of wavelength and incident angle can be obtained. QD-based detectors are similar to QWs based sensors. Still, they have many advantages over QWs, due to the confinement of the three dimensions of this nanostructure and the weaker thermionic emission (lower dark current). High sensitivity and high anti-reflection due to anisotropic geometry and high surface-to-volume rates are unique electronic and photonic characteristics of semiconductor nanowires. By locating the active region inside the resonant cavity between resonance mirrors (usually including a reflecting mirror and a Bragg distribution mirror), the absorption within the detector will increase significantly. Plasmonic and nanoantenna provide a high concentration of field and light–matter strong interaction over a broad spectrum band.