Large-scale Arrays Research Articles

The gate injection-based field-effect synapse transistor with linear conductance update for online training

Neuromorphic computing, an alternative for von Neumann architecture, requires synapse devices where the data can be stored and computed in the same place. The three-terminal synapse device is attractive for neuromorphic computing due to its high stability and controllability. However, high nonlinearity on weight update, low dynamic range, and incompatibility with conventional CMOS systems have been reported as obstacles for large-scale crossbar arrays. Here, we propose the CMOS compatible gate injection-based field-effect transistor employing thermionic emission to enhance the linear conductance update. The dependence of the linearity on the conduction mechanism is examined by inserting an interfacial layer in the gate stack. To demonstrate the conduction mechanism, the gate current measurement is conducted under varying temperatures. The device based on thermionic emission achieves superior synaptic characteristics, leading to high performance on the artificial neural network simulation as 93.17% on the MNIST dataset.

Chaos single photon LIDAR and the ranging performance analysis based on Monte Carlo simulation.

With the advent of serial production lidars, single photon lidar faces an increasingly severe threat of crosstalk. In this paper, we first propose the concept of Chaos Single Photon (CSP) lidar and establish the theoretical model. In CSP lidar system, chaos laser replaces pulsed laser, and the physical random sequence generated by a Geiger mode avalanche photodiode (GM-APD) responding to chaos laser substitutes the traditional pseudo-random sequence. The mean density of '1' code of the CSP lidar system can exceed 10 million counts per second (Mcps) with a dead time immunity. The theoretical models of detection probability and false alarm rate are derived and demonstrated based on Poisson distribution. The bit error rate (BER) is introduced into the CSP lidar system for evaluating the range walk error intuitively. Additionally, the simulation results indicate that the CSP lidar system has a robust anti-crosstalk capability. Compared with the traditional pseudo-random single photon (PRSP) lidar system, the CSP lidar system not only overcomes range ambiguity but also has a signal to noise rate (SNR) of 60 times, reaching 10000 when the mean echo photoelectron number is 10 per nanosecond. Benefited from large-scale arrays and extremely high sensitivity properties of GM-APDs, we are looking forward to the applications of the CSP lidar in weak signal detection, remote mapping, autonomous driving, etc.

A dense light field reconstruction algorithm for four-dimensional optical flow constraint equation

Dense light field sampling is an important basis for refocusing, depth estimation and 3-D imaging. It is difficult to obtain high resolution dense light field with a large-scale camera array and expensive equipment. At the same time, the current storage devices and transmission bandwidth also limit this technology's post-processing and application. In order to effectively reconstruct the angle domain of the light field based on the sparse light field data, this paper analyzes the correlation and constraint relationship between the optical flow field and the motion field of multi-view images in the same scene, extends the traditional optical flow constraint equation of two-dimensional imaging to the optical flow constraint equation of four-dimensional light field, and establishes an effective mathematical model. The coordinate position of the original pixel in the new angle image is determined by coordinate search, and its intensity is reconstructed. The experimental results of multi-scene dense reconstruction show that the proposed method can reconstruct the texture, shadow and color information in the light field of a long-baseline scene with high quality. The quantitative evaluation results show that the algorithm can be applied to dense light field reconstruction of complex scenes. The algorithm in this paper is only suitable for the case of optical flow constraint in a linear light field, and the follow-up research will focus on the case of nonlinear optical flow constraint.

(Digital Presentation) Sub-THz Front Ends for Future Communication and Sensing Technologies in 130nm Sige Bicmos Technology

Introduction: Beyond the-5G technologies are targeting the available bandwidth (BW) in the D-band and J-band. This promises for very high speed communication and sensing accuracy. Although the advantages of operating beyond 100 GHz from the application point of view, it requires a lot of research and development to design a reliable and power efficient chipsets. In this presentation we will go through the design of wideband 240 GHz transmitter (Tx) and receiver (Rx). Firstly an overview about the link budget analysis will be given, then the design of the local carrier signal generator will be discussed. The wideband Txs and Rxs are to be presented with frequency channelizing concept. At the end the results of the wireless links will be demonstrated. Link Budget Analysis: Operating at sub-THz frequencies approaching the transistor fT put a lot of constraints on the system link budget. The elevated noise figure together with the limited maximum Pout of the transistors lead to solutions which might be limited in BW or power efficiency. Hence the design architecture need to be optimized taking into account the elevated power consumption and heat dissipation. Carrier Generation: Several approaches could be followed to generate the sub-THz carrier frequency. For a power efficient solution the fundamental oscillators promise for a power efficient solution but with limited phase noise performance and tuning range. Such a solution is more common for imaging systems operating at a single frequency. In order to achieve wider tuning ranges for multi-channel scenario or FMCW radar systems, the frequency multiplication chains are common, whereas the multiplication factor depends on the system parameters. Sub-THz Rx: Different approaches were followed to realize a wideband sub-THz Rx. For some technologies the transistors ft does not promise the realization of low noise amplifiers (LNA) with reasonable noise figure, gain and BW. Hence the mixer first approach were followed trying to optimize the noise figure of the down conversion mixer as much as possible. This also enhances the input compression point (IP1dB) of the Rx which is better for monostatic radar applications. On the other hand, if the technology in use allows the development of well performing LNA’s, the LNA first approach was followed to enhance the all over NF of the receiver and ease the implementation of IQ Rxs. In our work an LNA was implemented followed by IQ downconversion mixers and baseband chain as presented in [3]. Sub-THz Tx: The Tx main challenging performance parameter in the sub-THz frequency range is to achieve high output compression point across the required BW with as high efficiency as possible. Increasing the power handling capabilities of the power amplifiers (PA) by increasing the transistors sizes of the output stages, leads to low output impedance and hence higher impedance transformation ratios for the matching structures and eventually narrow band designs. Hence the power combining architectures were proposed as an alternative to increase the output power either by combining on-chip or combining in air. In this work a fully integrated IQ Tx is to be presented with on-chip LO chain and 4-way power combined power amplifier. Baseband Signal Generation and Processing The baseband topology depend on the application. For mobile communication or localization use cases a wideband channels are not available directly from the digital processors, a kind of channel bonding is required either in the digital domain or in the analog domain. As a proof of concept we present here 3-IQ frequency interleaving combiner. Wireless Link Demonstrator Two demonstrators are to be presented. The first demonstrator is transmitting a broadband IQ signal generated from an arbitrary wave generator wirelessly utilizing an IQ 240GHz Tx and Rx. Data rates up to 100 Gbps were demonstrated across a 1m of wireless link. The second demonstrator includes the implementation of 3-channel IQ channelizer to create the complex modulated signal at several intermediate frequencies (IFs) and then upconverted to the 240GHz carrier signal. On the Rx side the de-channelizer convert the down modulated signal to three IQ channels. Data rates up to 8Gbps were demonstrated wirelessly with the channelization with potential of further enhancements for more channels integration. Outlook The future of the sub-THz wireless links for high speed communication and localizations goes towards implementing phased array and MIMO systems. This requires the research community to develop novel architectures to develop large scale arrays with acceptable power consumptions and suitable packaging solution. On the baseband side also the channel bonding solutions are still in an early stage, requiring more work to bond modularly larger number of channels.

A signal separation method based on the subarray beam synthesis

Large-scale arrays are widely used due to their advantages of high gain, high resolution, and strong beam control capability. The large number of antennas leads to a sharp increase in computational complexity, and thus, it is necessary to reduce the complexity of signal processing. A signal separation method is proposed based on subarray beam synthesis. This method can improve signal reception and separation performance through zero-forcing calculation and realise system gain and phase error corrections in the front end of an antenna. Compared with back-end processing, it has the advantage of low implementation difficulty. The feasibility and performance of the method are verified by simulations.

Manifold learning for dynamic array geometries

Large-scale distributed arrays can obtain high spatial resolution, but they typically rely on a rigid array structure. If we want to form distributed arrays from mobile and wearable devices, our models need to account for motion. The motion of multiple microphones worn by humans can be difficult to track, but through manifold techniques we can learn the movement through its acoustic response. We show that the mapping between the array geometry and its acoustic response is locally linear and can be exploited in a semi-supervised manner for a given acoustic environment. We will also investigate generative modelling of microphone positions based on their acoustic response to both synthetic and recorded data. Prior work has shown a similar locally linear mapping between source locations and their spatial cues, and we will attempt to combine these findings with our own to develop a localization model suitable for dynamic array geometries.

Guest Editorial Special Issue on Antenna Array Enabled Space/Air/Ground Communications and Networking

With the rapid development of electronic and information technologies, the Internet of Everything (IoE) has become one of the trendiest topics in both academia and industry. Therein, many types of space/air/ground platforms need to be connected to networks for breaking down the isolation of information islands and providing various services. Space/air/ground platforms, such as satellites, unmanned aerial vehicles (UAVs), airships, balloons, terrestrial vehicles, and high-speed trains (HSTs) have emerged for accomplishing various complex tasks. Wireless communication is one of the most important technologies to support the real-time delivery of control commands and mission-related data. On the other hand, the space-air-ground integrated network has become a promising paradigm for the six-generation (6G) mobile communication network, where the aerospace and terrestrial vehicles may need to connect to existing mobile cellular networks or act as base stations (BSs) or relays to assist terrestrial wireless communications. To meet the ever-increasing demands of high capacity, wide coverage, low latency, and strong robustness for communications, it is promising to adopt large-scale antenna arrays at the transceivers to obtain considerable array gains and improve the channel quality. Antenna array-enabled beamforming technologies can facilitate spectrum reuse, interference mitigation, coverage enhancement, and physical-layer security. Antenna arrays can also be used to promote the sensing capability of space/air/ground networks, where the sensing information may be carefully processed to assist communications. However, enabling antenna array for space/air/ground communication networks poses specific, distinctive, and tricky challenges in antenna array design, physical layer, multiple access control layer, and network layer. As a result, numerous new research issues require to be addressed, which cover a wide range of disciplines including communication theory, network theory, antenna theory, signal processing, protocol design, resource allocation, optimization, hardware implementation, and experimentation.

Multilayer doped-GeSe OTS selector for improved endurance and threshold voltage stability

Selector devices are indispensable components of large-scale memristor array systems. The thereinto, ovonic threshold switching (OTS) selector is one of the most suitable candidates for selector devices, owing to its high selectivity and scalability. However, OTS selectors suffer from poor endurance and stability which are persistent tricky problems for application. Here, we report on a multilayer OTS selector based on simple GeSe and doped-GeSe. The experimental results show improving selector performed extraordinary endurance up to 1010 and the fluctuation of threshold voltage is 2.5%. The reason for the improvement may lie in more interface states which strengthen the interaction among individual layers. These developments pave the way towards tuning a new class of OTS materials engineering, ensuring improvement of electrical performance.

Massive MIMO Empowered Wireless Powered Sensor Networks: An Optimal Design With Statistical CSI

Wireless energy transfer (WET) has been anticipated as a viable solution to combat the challenges of scarce energy supply in Internet-of-Things (IoT) applications. To reap the benefit brought by large scale antenna arrays, massive multiple-input-multiple-out (MIMO) empowered wireless powered sensor networks (WPSNs) is investigated in this letter to meet the urgent demand for reliable and self-sustainable IoT applications. Considering a large-scale antenna array deployed at the access point for energy transfer to and information collection from the multiple-antenna wireless powered sensors, the hardening property in energy transfer is generalized and theoretically proved based on random matrix theory. Based on the hardening properties, the harvesting energy and the uplink sum-rate for sensor data collection are found to be asymptotically deterministic and depend on the statistical channel state information (CSI) only. The hardening properties in both energy and information transfers enable a novel optimal design on the WPSNs based on statistical CSI only. Particularly, the designs of the downlink energy beamforming and multiple uplink information beamforming can be decoupled, and global optimal solutions for the beamforming and time allocation to maximize the uplink sum-rate given downlink power constraint are found in semi-closed forms. Different from prior WPSNs designs, statistical CSI but not instantaneous CSI is exploited and frequent channel estimation and heavy communication overhead can be avoided with a much lower computation complexity, thus improving the practicability and sustainability of the WPSNs. The effectiveness of the proposed design is finally demonstrated by numerical simulations.

Degrees of Freedom in 3D Linear Large-Scale Antenna Array Communications—A Spatial Bandwidth Approach

For wireless communications using linear large-scale antenna arrays, we define a receiving coordinate system and parameterization strategy to facilitate the study of the impact of three-dimensional position and rotation of the arrays on the achievable spatial degrees of freedom (DoF) in line-of-sight (LOS) channels. An analytical framework based on spatial bandwidth analysis is developed, under which three elementary problems corresponding to three basic orthogonal receiving directions are investigated. For each of them, accurate, simple, and interpretable closed-form approximations for the achievable spatial DoF are derived, and the spatial region where a sufficient amount of spatial DoF is expected available is determined. The expressions can easily be integrated into large-scale system-level simulations. Some interesting and surprising observations are made from simulation studies based on the analytical results. For instance, the spatial bandwidth is shown to be approximately constant in almost the entire spatial multiplexing region. Moreover, in significant parts of this region, the optimal receive array orientation is not parallel with the transmitting array.

Ka-Band Single-Layer High-Efficiency Circular Aperture Microstrip Array Antenna

Based on microstrip antenna technology, a compact single-layer circular aperture array is proposed in this paper. The proposed method is suitable for designing large-scale and high-efficiency patch array in Ka band. By using the high-impedance characteristic of the coupled microstrip line, a novel multi-stage feeding network that can realize large power division ratio is proposed, and its structure is center-symmetrical and is fed through a coaxial probe at the center. The parallel branch of the multi-stage divider is connected to the row array, and the array elements can obtain uniform excitation through the design of the division ratio of each stage. A 256-element patch array operating at 35 GHz is proposed, the length and width of the antenna unit are 2.74 mm and 3 mm, respectively, the horizontal spacing between the array elements is 0.742λ0, the vertical spacing varies from 0.677λ0 to 0.737λ0, and the radius of the proposed array is 60 mm (7λ0 at 35 GHz). At the end, a prototype is fabricated and measured, the measured results show that the maximum gain is 29.5 dBi at 35 GHz, which correspond to the radiation efficiency and aperture efficiency of 61.66% and 46.07%, respectively, and the results show that the proposed antenna achieves high efficiency in Ka band, which verifies the correctness of the design. The proposed array antenna is compact in structure and only needs single-layer design, which can achieve high efficiency and low in manufacture cost, which is an excellent candidate for millimeter-wave large-scale array application.

Large-Scale Integrated Flexible Tactile Sensor Array for Sensitive Smart Robotic Touch.

In the long pursuit of smart robotics, it has been envisioned to empower robots with human-like senses, especially vision and touch. While tremendous progress has been made in image sensors and computer vision over the past decades, tactile sense abilities are lagging behind due to the lack of large-scale flexible tactile sensor array with high sensitivity, high spatial resolution, and fast response. In this work, we have demonstrated a 64 × 64 flexible tactile sensor array with a record-high spatial resolution of 0.9 mm (equivalently 28.2 pixels per inch) by integrating a high-performance piezoresistive film (PRF) with a large-area active matrix of carbon nanotube thin-film transistors. PRF with self-formed microstructures exhibited high pressure-sensitivity of ∼385 kPa-1 for multi-walled carbon nanotubes concentration of 6%, while the 14% one exhibited fast response time of ∼3 ms, good linearity, broad detection range beyond 1400 kPa, and excellent cyclability over 3000 cycles. Using this fully integrated tactile sensor array, the footprint maps of an artificial honeybee were clearly identified. Furthermore, we hardware-implemented a smart tactile system by integrating the PRF-based sensor array with a memristor-based computing-in-memory chip to record and recognize handwritten digits and Chinese calligraphy, achieving high classification accuracies of 98.8% and 97.3% in hardware, respectively. The integration of sensor networks with deep learning hardware may enable edge or near-sensor computing with significantly reduced power consumption and latency. Our work could empower the building of large-scale intelligent sensor networks for next-generation smart robotics.

Demonstration of 2D beam steering using large-scale passive optical phased array enabled by multimode waveguides with reduced phase error

Optical phased arrays (OPAs) have received considerable attention as solid-state beam scanners. However, conventional OPAs that actively control the phase difference between arrays are characterized by excessive power consumption for high-precision beam emission. In this study, we fabricated an OPA comprising Bragg grating and arrayed waveguide grating (AWG). Multi-mode waveguide is used in AWG to reduce the effect of manufacturing error. This device realizes wide and high-resolution two-dimensional beam steering only by sweeping wavelength. FWHM of the emitted beam is 0.534° × 2.27°, and the steering range is 43.9° × 13.5° with 1/64 of the power consumption of conventional OPA.

Wafer-level calibration of large-scale integrated optical phased arrays.

We present the wafer-level characterization of a 256-channel optical phased array operating at 1550 nm, allowing the sequential testing of different OPA circuits without any packaging steps. Using this, we establish that due to random fabrication variations, nominally identical circuits must be individually calibrated. With this constraint in mind, we present methods that significantly reduce the time needed to calibrate each OPA circuit. In particular, we show that for an OPA of this scale, a genetic optimization algorithm is already >3x faster than a simple hill climbing algorithm. Furthermore, we describe how the phase modulators within the OPA may be individually characterized 'in-situ' and how this information can be used to configure the OPA to emit at any arbitrary angle following a single, initial calibration step.

Compressive Channel Estimation Based on the Deep Denoising Network in an IRS-Enhanced Massive MIMO System.

Integrating large intelligent reflecting surfaces (IRS) into a millimeter-wave (mmWave) massive multi-input-multi-output (MIMO) technique has been a promising approach to enhance the performance of the wireless communication system with the channel state information (CSI). Most existing work assume that ideal channel estimation can be obtained, but the proposed high-dimensional cascaded MIMO channels and passive reflectors pose a great challenge to these methods. To address the abovementioned problems, we proposed a new method for the reduction of training overhead in IRS with a partial ON/OFF model and an optimizing strategy for pilot design approach. The energy consumption of large-scale antenna arrays and the pilot overhead in the training phase of signal transmission are greatly reduced. Besides, we proposed an improved deep residual shrinkage denoising network, which possesses better denoising performance with a soft thresholding model. The channel data can be denoised by deep learning methods, which greatly improve the accuracy of channel estimation. Simulation results demonstrate that the superiority of the proposed network over prior solutions.

Compact low‐cost W‐band large‐scale high‐gain circular‐aperture slot array antenna

Based on substrate-integrated waveguide (SIW) antenna technology, a compact high-gain slot array is proposed in this paper. The proposed antenna is suitable for designing circular aperture and large-scale array in W-band. A method of dividing the array into multiple sections is proposed, which provides a solution for widening the bandwidth of large-scale SIW slot arrays. The proposed antenna is composed of the backside waveguide feeding network and the upper SIW array. A novel method of loading biased metal vias on SIW is proposed to realize unequal power distribution. And it makes the antenna array to be compact, low-cost, and easy to manufacture. An antenna prototype is fabricated and measured. The measured bandwidth of the antenna array is 92.2–96.2 GHz within |S11|<−10 dB, and the measured maximum gain is 31.8 dBi at 94 GHz, which correspond to the radiation efficiency and aperture efficiency of 27.5% and 15.7%, respectively. The proposed antenna is an excellent candidate for millimetre-wave radar applications.

Omnidirectional precoding for massive MIMO with uniform rectangular array in presence of mutual coupling

The massive Multiple-Input and Multiple-Output (MIMO) system is one of the technologies in 6-Generation (6G) communication networks. Omnidirectional or broad coverage precoding is used in massive MIMO to broadcast the common signals from public channels to all active and non-active users so that the radiation power of these signals is constant in all spatial directions. On the other hand, in large-scale array antennas, the presence of mutual coupling leads to deteriorating the performance of MIMO systems. In this paper, we propose omnidirectional precoding of the massive MIMO equipped with a Uniform Rectangular Array (URA) in presence of the mutual coupling. In the proposed method, omnidirectional transmission and maximum achievable ergodic rate conditions are developed for the case that mutual coupling exists, then an omnidirectional precoding matrix is designed to satisfy three constraints: 1) Omnidirectional transmission, 2) Equal average power transmission per antenna, and 3) Maximum achievable ergodic rate. For this purpose, we show that an omnidirectional matrix can be reconstructed using a pair of vectors with elements' absolute value equal to one that forms a Complementary Set (CS), and the Complete Complementary Codes (CCC) that satisfied some conditions. The intended CS is selected from the Golay sequence and the intended CCC is obtained by solving a gradient algorithm with projection on the Grassmann and Stiefel manifolds. Simulation results confirm the expected results.

Technique for Large-Scale Antenna Beamforming Based on Neural Network

The commercialization of the fifth-generation technology and the rapid development of Internet of Things technology have made mobile communication networks increasingly complex. Simultaneously, the massive terminal connections have caused serious interference between networks. Large-scale array antennas have become a hot topic of recent studies to improve the wireless transmission characteristics and spectrum resource utilization efficiency. This study was aimed at exploring using large-scale array antennas combined with neural networks in ultradense cells to improve the wireless signal transmission quality. The simulation results showed that the scheme realized not only the wireless signal transmission with a neural network but also the effective recovery of the source signal. Similarly, the application of this scheme effectively reduced the power consumption during signal transmission, the delay, and the interference.

An Efficient OTA Calibration and Pattern Estimation Method for 5G mmWave Large-Scale Arrays

In this article, an efficient over-the-air (OTA) calibration and pattern estimation method under the midfield (MF) condition for the fifth-generation (5G) millimeter-wave (mmWave) large-scale arrays is proposed. A new transformation and revision algorithm is developed to calibrate the antenna array fast and accurate. Meanwhile, the time consumed in the accurate beamforming pattern estimation is reduced significantly. The proposed method requires a much smaller size anechoic chamber, a fixed probe antenna, and a vector network analyzer (VNA). The effectiveness is proven with the detailed theoretical analysis and experimented with a 64-element phased array in the 26 GHz band. The calibrated root-mean-square (rms) errors of the phase and magnitude are about 1.8°/0.4 dB. In addition, the proposed calibration and pattern estimation method is tremendously cost-effective and time-saving with similar accuracy as the far-field test method.

Ultranarrow Linewidth Coupling Resonance in Flexible Plasmonic Nanopillar Array for Enhanced Biomolecule Detection (Adv. Mater. Interfaces 27/2022)

Plasmonic Nanostructure Sensor In article number 2201011, Shuwen Chu, Yuzhang Liang, and co-workers experimentally demonstrate a flexible plasmonic nanostructure sensor based on a large-scale periodic array of nanopillars. The structure utilizes the coupling of Wood's anomaly and Fabry-Perot modes to generate an ultra-narrow plasmon resonance mode with high sensing figure of merit (FOM), providing an effective strategy for label-free biomolecular sensing with low cost, high sensitivity and high throughput.