There has been a flurry of studies on the utilization of reconfigurable intelligent surfaces (RISs) in wireless remote networks to develop intelligent radio environments. In RIS, surfaces can control the propagation of electromagnetic incident waves in a programmable smart radio environment [1]. It provides a way of consciously changing the realization of the channel, which transforms the channel into a block of a controllable device that can be optimized to maximize system performance overall. Therefore, RIS is an artificial surface of electromagnetic (EM) material, electronically controlled with integrated electronics. It is a novel and cost-effective solution to obtain enhanced energy and spectral efficiency for wireless communications.

These surfaces have unique wireless communication capabilities. Recent studies on RIS are based on estimation of theoretical signal to noise ratio (SNR), signal to interference ratio (SINR) maximization, physical layer security solutions, cognitive radio applications, and artificial intelligence solutions (such as deep learning) [2]. Many scholars have published numerous studies and novel solutions related to RISs in the last few months. Reconfigurable intelligent surfaces, intelligent reflecting surfaces, artificial radio space, and other concepts have been used by different writers to describe RISs [3]. Scholars have also looked into machine learning methods, physical layer protection solutions, and the ability of intelligent surfaces for millimeter-wave (mmWave), visible light communication (VLC), and free-space optics (FSO). Furthermore, recently, the first attempt to integrate RISs with OFDM and SM/space shift keying (SSK) schemes have been reported in [4]. Researchers have evaluated the Theoretical SNR and SEP derivations, channel estimation, signal-to-interference-ratio (SINR) improvement for RIS, and joint active and passive beamforming optimization problems in the past few years [5].

Moreover, researchers have studied outage probability, asymptotic data rate, and uplink spectral efficiency when using RISs for transmission and reception in many novel research areas such as mmWave, FSO, VLC system, and unmanned aerial vehicles (UAV), as mentioned earlier. Additionally, the use of machine learning tools, physical layer security solutions, and the potential of intelligent surfaces for mmWave/terahertz applications have been examined in recent years [6,7].

By supporting MIMO transmission with better throughput and increasing spectrum efficiency in mmWave communication, RIS brings up new opportunities in mmWave communication. A RIS can alter radio propagation for mmWave MIMO Channels by passively adjusting the directions of impinging electromagnetic waves. Due to their higher performance over traditional MIMO systems, RIS has recently attracted much attention as a potential technique for FSO and hybrid RF/FSO communications. Furthermore, RIS as a wireless transmission technique in combination with hybrid RF/FSO can achieve significantly higher gain while decreasing design complexity and cost compared to multi-antenna amplify and forward relaying networks with fewer antennas.

RIS-Assisted VLC and Hybrid VLC-RF Networks show remarkable capacity, throughput, and coverage augmentation potential. This integration has the potential to provide a powerful solution for future wireless applications as it can efficiently overcome line-of-sight blockage in highly dynamic settings, such as vehicle application scenarios. Furthermore, it helps to avoid bottlenecks while allowing for intricate interactions between network elements [8,9]. According to recent studies, RIS can significantly improve UAVs’ energy efficiency and connectivity, mainly when multiple devices are supported simultaneously, and channel impairments vary. The authors in [10,11] focused on improving the system’s secure energy efficiency by maximizing the UAV’s direction, the RIS’s phase shift, user association, and transmit power all at once.

The main contribution of this paper is to provide a detailed state-of-the-art survey for RIS-assisted technologies and metasurfaces based on large surfaces with their merits and demerits. Further, various novel implementations of RIS, such as active RISs and their signal models, are presented and compared with their performance with a passive one. Their differences are discussed concerning the link budget for the connectivity.

To benefit the overall system performance, we then address the challenges in RIS implementation and highlight the possible opportunities if the challenges can be overcome. Furthermore, we review some novel field-programmable gate array (FPGA) based dynamically controlled RIS structures. Finally, we provide a detailed overview of RIS-assisted communication systems, their performance parameters comparison, applications, and ongoing and future research.

RISs can be reconfigured electrically, mechanically, or thermally, based on the tuning mechanisms. The electromagnetic properties of the RIS, such as phase discontinuities, can be controlled by tuning the surface impedance through various techniques. In addition to electrical voltage, other processes for tuning are thermal excitation, optical pump, and physical stretching. Electrical control is the most suitable way as the electrical voltage is easier to quantize and control with FPGA chips. In another way, it is possible to consider the RIS as a broad disjoint beamformer from the transmitter. One appealing feature of some of these approaches is that they do not need improvements to the wireless protocol used. The digital description of coding metasurfaces is well adapted to integrating active elements such as PIN diodes, varactors, and micro-electro-mechanical systems (MEMS). As a result, an FPGA may control all coding elements of a digital coding metasurface separately. Many different functionalities can be switched in real-time by modifying the coding sequences stored in the FPGA, leading to programmable metasurfaces. An FPGA-based dynamically controlled RIS structure is demonstrated in Figure 6, and its related work is discussed next.

A system based on a microcontroller unit (MCU), DACs board, and FPGA is commonly used to dynamically control the RIS [95,96]. The RIS in [96] is made up of 328 REs, each of which is loaded with two varactor diodes to achieve a 450-phase reflection range. The varactor is then tuned for the appropriate phase distribution using a central controller, FPGA, and DAC to provide bias voltage. When the RIS reflects the incident wave, the signal is placed onto the carrier.

The reflection phase of each RE in the proposed RIS is controlled by an upper computer (UC) through FPGA. The UC first codes the designed quantized phase before sending it to the FPGA, which uses its output pin to link the PIN diodes on the RIS. Then, to achieve the appropriate phase distribution, each PIN diode loaded on the RE is switched to the ON or OFF states. As a result, the reflected radiation pattern can be dynamically modified using different codes given to the FPGA (Figure 6). The simulated radiation pattern of beam scanning is obtained using this control mechanism [97].

In [25], the authors employed the on/off status of PIN diodes to alter the phase responses of meta-material elements. Finally, the authors developed FPGA hardware that uses PIN diodes to control programmable meta-surfaces. The authors claim that these programmable metamaterials can be used to lower the scattering properties of targets and modify antenna radiation beams. In [98], to control the scattered EM waves using the coding metasurface, in which each unit cell loads a pin diode to produce binary coding states of “1” and “0”, the authors presented a direct digital modulation scheme. Instant communications between the coding metasurface and the internal memory of FPGA are established via data lines. As a result, electromagnetic wave digital modulation is achieved, and it offers a field-programmable reflecting antenna with good measurement performance. The proposed method and functional device have a lot of potential for use in next-generation radar and communication systems. Basically, the binary units are realized by loading pin diodes to sub-wavelength artificial structures in a field-programmable reflective antenna based on the coding metasurface in the microwave frequency.

A field-programmable reflective array antenna made up of a horn antenna, and a reflective coding metasurface is presented in this paper. The coding metasurface is built using a binary-phase element and a chessboard configuration approach. The binary codes of the metasurface are controlled directly by field-programmable gate arrays (FPGA) by loading a pin diode in each element. The role of FPGA is to configure the code distributions on the coding metasurface. As a result, the scattered major lobes can be digitally reconfigured at the same frequency. Additionally, it is worthwhile to point out that a real-time switch among these functionalities is also achieved by using an FPGA. In [99], the authors proposed dynamic multi-functional properties of a digitally controlled metasurface (relatively large aperture size >20 wavelengths). The proposed programmable metasurface can be used for a variety of future applications, such as smart stealth missions and novel phased array techniques without expensive phase-shifting components. Each sub-metasurface consists of 320 active unit cells and the proposed metasurface is made up of 5 similar sub-metasurfaces.

A reconfigurable phase for a single polarization is achieved by incorporating one PIN diode into each unit cell. The reconfigurable polarisation conversion is achieved first by utilizing this anisotropic characteristic. Then, a FPGA is used to switch between these functionalities in real-time. In contrast to earlier work that typically controls a lattice and focuses on a single sort of steerable function, each unit cell in the proposed metasurface can be controlled individually, allowing for more versatile functions to be achieved simultaneously. In ref. [26], a prototype of the proposed coding metasurface is validated by building an FPGA-controlled prototype. The coding metasurface uses an FPGA hardware control board (ALTERA Cyclone IV) to generate dynamic biassing voltages, with each column sharing a control voltage. The FPGA is a low-cost system with a clock speed of 50 MHz and a preloaded code that generates eight control voltages according to time-coding sequences. By modifying the coding components on a 2D plane with predesigned coding sequences, reconfigurable meta-surface structures may be used to manipulate EM waves simply and effectively [100]. A Field Programmable Gate Array (FPGA) is used to regulate and construct multiple coding sequences independently. As a result, many distinct functionalities can be swapped in real-time by modifying the coding sequences recorded in the FPGA, resulting in programmable metasurfaces. In ref. [101], a similar approach has been used to formulate the exact features of a meta-surface for vehicular communication applications in the frequency band of 5–5.9 GHz.

An Intelligent metasurface imager and recognizer is proposed in which a network of artificial neural networks (ANNs) is used for adaptively controlling data flow. It transforms the measured microwave data into images of the whole human body, classifying specifically designated spots (hand and chest) within the entire image. Thus, it recognizes human hand signs instantly at a Wi-Fi frequency of 2.4 GHz. An FPGA with its changing coding sequences, the large-aperture programmable metasurface is designed to dynamically and adaptively regulate ambient EM wavefields. First, FPGA acts as an information relay station or an electrically controllable random mask, sending EM signals with finer information about the specimen to the receivers. Secondly, the programmable metasurface with optimized coding patterns can focus EM wavefields on the desired areas while suppressing irrelevant interference and clutter, allowing body language recognition and respiration monitoring to be realized [102]. In [96], a reconfigurable reflectarray with the feature of fast steerable monopulse patterns has been proposed and tested at X-band. To reconfigure the reflective phase between 0° and 180°, the reflectarray element integrates one PIN diode. One hundred sixty field-programmable gate arrays (FPGAs) are used to regulate the “ON/OFF” states of every PIN diode connected in parallel to provide quick beam steering. This FPGA-based reflectarray design approach is proved to be feasible for fast beam-switching.