This PDF file contains the front matter associated with SPIE Proceedings Volume 12973, including the Title Page, Copyright information, Table of Contents, and Conference Committee information.

Although conventional structure Pseudo Resistors (PRs) can provide high value resistance in a small chip area, they still need to be improved in terms of linearity. A new PR structure with good frequency characteristics and high linearity was proposed for Capacitively-coupled Chopper Instrumentation Amplifier (CCIA). On the other hand, CCIA using a classical analog DC-servo loop (DSL) can only block biological Electrode DC Offset (EDO) ±50mV or so. To calibrate a larger EDO in a dry gel application, a bulk feedback DSL with the new structure PR is proposed which uses the parasitic capacitance at the input of the amplifier and the body effect to block the EDO. The CCIA was designed using a 180nm CMOS process. It occupies only 0.27mm² chip area and consumes 5.4μA current from a 1.8V supply. Its input-referred noise varies from 0.44 to 2.2μVrms, total harmonic distortion of 60dB. It can calibrate ±300mV EDO and the maximum capacitance required for digital DSL is 1.37pF. Compared with the traditional DSL method, the capacitor area is reduced by 29%.

In today's trains where AC power supply is commonly used, circuit breaker operation is the main source of overcurrent in trains. This paper investigates the transient circuit model of interference to the axle counting system by train VCB operation and analyzes the overcurrent interference caused by the axle counters. This paper analyzes the propagation path of overcurrent interference to the axle counter by introducing the main circuit structure of high-speed trains, the axle counting principle of the axle counter and the overcurrent principle generated by the circuit breaker operation. By using the Q3D and Simplorer of ANSYS software platform to build a three-dimensional equivalent model of car body and high-voltage cable and equivalent circuit models of traction network, high-voltage system and grounding system respectively, a joint simulation was carried out in Simplorer to obtain the overcurrent in the rails and the magnitude of the overcurrent and the magnetic field generated around the axle counter, and the simulation results show that the overcurrent through the rails The simulation results show that the magnetic field generated by the overcurrent through the rail in the Z-axis direction may have an effect on the operation of the axle counter.

During researching on GaN-based quantum well diodes, an intensive and continuous current oscillation phenomenon was discovered. To describe the discovered phenomenon and release its mechanism and foreseeable applications, an In_0.21GaN/GaN/In_0.14GaN/Al_0.1GaN/In_0.07GaN stacked DBS structure was proposed and modeled by coupled 1D Schrodinger and Poisson equations. Then, method of ‘non-equilibrium Green function (NEGF)’ was adopted to carry out simulation experiments on the proposed samples with different structure parameters. Simulation results indicate that the discovered phenomenon occurs at bias voltages higher than about 3.15 V and accompanying main RT oscillations some time. Through studies and analyses, it was released the generation of the discovered phenomenon is mainly determined by band structure, distribution of Bound State Energy Levels (BSELs), electron distribution on BSELs, superstition of electron transportation through multi-sub-band BSELs and the impact of polarization field. The proposed RTDs with intensive and continuous current oscillation characteristic is very suitable for versatile applications such as Voltage Controlled High Frequency Small Signal Voltage-Current Converter (VCHFSSVCC), Voltage Controlled Impedance Matching (VCIM), Voltage Controlled RF Tuning (VCRFT), Voltage Controlled Electron Wave Filter (VCEWF), Multi-Valued Half-Voltage Domain and Full-Voltage Domain Pure Quantum Logic (MVPQL), Novel Neuron Logic (NNL) and some other circuits and systems.

In this paper, the bonding wire used for multi-chip interconnection is studied, and the model of non-parallel double bonding wires is proposed based on the single bonding wire model. Then, a microstrip impedance matching structure is designed based on the equivalent circuit model of non-parallel double bonding wires. In order to verify the effectiveness of microstrip impedance matching structure, experimental measurements are carried out. The experimental results show that the return loss is better than – 15dB in the frequency band of 25.8GHz - 40GHz, which is optimized by more than 5dB. The insertion loss is optimized by 0.7dB at most in the frequency band of 1GHz - 40GHz.

As the scale of high-performance System-on-Chip (SoC) continues to grow, the need for frequent data exchange among various Intelligent Property (IP) cores such as processors, memories, and peripherals becomes crucial. However, existing on-chip bus structures often compromise area and complexity to enhance performance for accommodating the data transfer requirements of complex SoCs. In this study, we propose a high performance, low area interconnect architecture based on the Advanced eXtensible Interface(AXI) bus specification, which can be effectively utilized in conjunction with several processors to construct efficient and compact embedded SoC. The proposed design employs a hierarchical approach for read and write operations to minimize area consumption and resource utilization. For read and write transactions, a shared address bus and a shared data bus structures are utilized, respectively, to support synchronous read and write operations. Additionally, to enhance the data transfer capability and reliability of the interconnect, our approach involves the utilization of weighted arbiters for prioritization and the implementation of random access queues as a caching mechanism. Experimental results demonstrate that this design enables efficient data transfer with exceptional performance and scalability, while maintaining the benefits of reduced area and complexity.

The adoption of IoT has raised concerns about security as life-threatening attacks become more common. To address this, ultra-lightweight secure SoCs become a demand to ensure trusted operation in the presence of untrusted IPs while meeting power and area constraints. Integrity, one of the main components of security, is the prevention of protected data modification without detection. Hash functions are used to ensure data integrity in secure systems. In this paper, we propose a low-power implementation of a hash function by using approximate adders in SHA-256 implementation. First, approximate adders’ designs were investigated, where we tested the output uniqueness of each of them, as the output uniqueness of an adder improves the properties of hash results. Then, the chosen adders are utilized for hash implementation. Next, the Avalanche test is applied to evaluate the properties of the proposed hash functions. Synthesis and simulation results show that using an approximate adder reduces the power consumed in the combinational part by up to 66% while the avalanche test results are reduced gradually from 50% to 25%.

In this paper, artificial surface equipartition excitations are applied to phase shifter to realize the miniaturization of phase shifter. A coplanar waveguide liquid crystal phase shifter is designed based on the artificial surface equipartition exciton structure, which operates in the frequency band of 42-48 GHz. In the case of achieving a large phase shift of 360°, the liquid crystal cavity length of the conventional coplanar waveguide structure-based liquid crystal phase shifter is 78.6mm, while the designed phase shifter uses a metal strip structure with periodic uniform slotting, which increases the contact area between the transmission line and the liquid crystal, making the length of the liquid crystal cavity only 28.8mm, which greatly reduces the size of the phase shifter. The final design of the liquid crystal phase shifter in the operating band, the liquid crystal dielectric constant changes from 2.4 to 3.2 in the case of the return loss are less than -15.0dB, the maximum insertion loss is 5.1dB, the quality factor is FOM = 76.3 °/dB.

Inverse modelling with deep learning algorithms involves training deep architecture to predict device’s parameters from its static behaviour. Inverse device modelling is suitable to reconstruct drifted physical parameters of devices temporally degraded or to retrieve physical configuration. There are many variables that can influence the performance of an inverse modelling method. In this work the authors propose a deep learning method trained for retrieving physical parameters of Level-3 model of Power Silicon-Carbide MOSFET (SiC Power MOS). The SiC devices are used in applications where classical silicon devices failed due to high-temperature or high switching capability. The key application of SiC power devices is in the automotive field (i.e. in the field of electrical vehicles). Due to physiological degradation or high-stressing environment, SiC Power MOS shows a significant drift of physical parameters which can be monitored by using inverse modelling. The aim of this work is to provide a possible deep learning-based solution for retrieving physical parameters of the SiC Power MOSFET. Preliminary results based on the retrieving of channel length of the device are reported. Channel length of power MOSFET is a key parameter involved in the static and dynamic behaviour of the device. The experimental results reported in this work confirmed the effectiveness of a multi-layer perceptron designed to retrieve this parameter.

Ground Penetrating Radar (GPR) is a widely used technology for non-destructive near-ground detection. It is often difficult to balance detection depth and resolution when using a single-frequency antenna. To solve this problem, a novel multi-frequency data fusion algorithm is proposed. The two-dimensional Gaussian weighted window is utilized as a sliding window to analyze regional energy characteristics. By matching multi-frequency data with these properties, the benefits of high resolution for high-frequency data and high amplitude for low-frequency data can be leveraged. The matching threshold is determined using a genetic algorithm. Fusion is then performed based on the matching results obtained. The effectiveness of the proposed algorithm was evaluated through experimental comparisons with other algorithms at frequencies of 900MHz and 1600MHz. The experiment results indicate that the proposed algorithm can enhance the resolution and information content, from which a superior radar profile is acquired.

Indoor positioning technology plays a critical role in Location Based Services (LBS). However, due to the indoor environment's complexity, it is difficult to develop a real-time indoor positioning system with low cost and high precision. This paper presents a real-time localization system based on the fusion of visible light with multiple sensors. In this system, micro-controllers regulate the positioning frequencies of LEDs to realize transmission of visible light signals. An interface circuit with an external light sensor receives visible light signals. The interface circuit then delivers the collected data to its connected smartphone for visible light positioning. Simultaneously, the smartphone collects and processes multi-sensor data to achieve Pedestrian Dead Reckoning (PDR). The improved particle filtering on the smartphone then fuses the two pieces of location information to real-time obtain the precise position of the device. The experimental results indicate that the accuracy of the real-time positioning system is 0.36m, which is 68.7% and 69.5% higher compared with visible light positioning and PDR.

In active noise control system, the noise control effect based on the traditional method is reduced under nonlinear interference from the electronic devices like speakers and microphones. In this paper, a combined CNN-LSTM network active noise control system is proposed, and an experimental scenario for imitating a real auditory system is built using an artificial head. The primary and secondary channel models are estimated using BP neural networks to simulate the nonlinear properties from the electronic devices. Sever type noise data from various scenes are chosen for simulation. As compared with the conventional FxLMS algorithm, FNN, CNN and LSTM, the trials demonstrate that the proposed network can effectively suppress the noise at both low and high frequencies with high convergence speed.

The dynamic programming track before detect (DP-TBD) algorithm, with its high target detection performance, has garnered significant attention. However, as the processing dimension increases, the computational complexity inevitably rises. The high computing load poses challenges for the unoptimized DP-TBD algorithm to meet real-time requirements in engineering applications, considering hardware limitations. Therefore, effective reduction of the calculation load is crucial to enable the application of TBD technology in real-time systems. To address the issue of high computational complexity in the DP-TBD algorithm, this work proposes a DP-TBD algorithm based on the Hungarian assignment algorithm. The core concept of this algorithm involves filtering low-amplitude data through a low-threshold process and then using the Hungarian assignment algorithm to establish data associations between adjacent frames. By filtering out a large amount of low threshold data, the proposed DP-TBD algorithm significantly reduces computational complexity. Simulation results demonstrate that the proposed algorithm exhibits good detection performance even in low signal-to-noise ratios (SNR).

The study of tropospheric scattering channel propagation characteristics is the basis for establishing a reliable scattering communication system. Before studying the scattering channel propagation characteristics, the propagation mechanisms of scattering are introduced, including turbulence incoherent scattering, irregular layer incoherent reflection, and stable layer coherent reflection, as the theoretical support. For the loss characteristics, several more applicable prediction models for convective-layer scattering link transmission loss are introduced. For the fading characteristics, the causes of fast and slow fading are analyzed, and the power delay spectrum and Doppler power spectrum of the scattering channel are studied. It is shown that the power delay spectrum can be obtained from the inverse Fourier transformation of the autocorrelation function in frequency domain of the received signal, and the Doppler power spectrum can be obtained from the Fourier transformation of the autocorrelation function in time domain of the received signal. The propagation characteristics of the scattering channel and its significance are initially explored theoretical levelly.

Free-space continuous variable quantum key distribution has attracted attention due to its ability to provide a more widely available transmission channel. However, the key rate of free-space continuous variable quantum key distribution itself is not high, and a portion of the original key needs to be allocated for parameter estimation. In this article, we analyze the spatial correlation of free-space continuous variable quantum key distribution and propose the use of a nearby system to transmit an inexpensive signal. Parameter estimation is performed using a deep learning model to predict and reduce the amount of original key used for parameter estimation, thereby improving the key rate.

There are deficiencies in the traditional single-factor authentication based on static password: Dangers of intrusion hidden after “connect first, then authenticate”, single factor of “user” is authenticated, the single-factor security is low, and it is difficult to balance security and ease of use. A multi-factor biological factor authentication scheme based on PKI architecture is designed. It overcomes the above problems and users can use Iris, vein pattern, fingerprint USBKey to complete two-factor authentication (2FA) or multi-factor authentication (MFA) for higher security. In addition, this solution can provide wide-area identity authentication across multiple trust domains based on a hierarchical authentication architecture. This scheme can resist replay attack, masquerade attack, and network transmission interception attack. The new schema is applicable to a wide range of business systems.

This essay explores the pivotal role of blood oxygen level-dependent functional magnetic resonance imaging (BOLD fMRI) in advancing neuroimaging research, with a focus on its application in analyzing two distinct tasks: the Flanker Task and the Board Task. The study outlines essential preprocessing steps for brain imaging data and employs the General Linear Model (GLM) for statistical analysis. The Flanker Task assesses cognitive control, requiring participants to respond to arrow stimuli. Brain activation maps reveal significant activity in the prefrontal cortex, parietal lobules, and occipital lobes during this task. The Board Task, designed to study motor control, involves participants tapping a flashing checkerboard. The analysis identifies activation in the occipital V1 region, parietal cortex, and prefrontal cortex.