Journal of Circuits, Systems and Computers

Floating point (FP) multiplication goes down in the scientific application when it sustains the subnormal inputs either in the implementation of software or hardware. Any high-level language executes the FP instructions in the graphics processing unit (GPU) and floating-point unit (FPU) for supporting the normalized numbers alone. In FP multiplication, execution times for normalized and subnormal numbers are not equal. Execution time variations create unintentional delay and data timing channels (DTCs). A circuit is proposed for floating-point multiplication to minimize the unintentional delay for the holistic support of subnormal numbers. In this proposed four-path FP multiplication, the circuit produces the four types of output in four paths having different delays for all cases of input combination. These four paths are establishing the DTCs. A maximum delay path is taken into account to combine and equalize the four paths into a single output path. Two levels of the control circuit combine the four paths to a single path for reducing the DTC effect. To evaluate the performance after path equalization, the proposed FP multiplier is implemented in Stratix-IV and Cyclone-IV FPGAs with a delay of 57.25 and 82.82 ns, respectively. Here, eight pipeline stages reduce the delay and improve the operating speed of the entire circuit. Stage delay and operating speed for this FP multiplier in both FPGA implementations are 12.44 and 16.86ns, and 153.19 and 116.78MHz, respectively.

The purpose of this paper is to address the problem of assessing the stability of singular time-varying delay systems. In order to highlight the relations between the delay and the state, the singular system is transformed into a neutral form. Then, a model transformation using a three-terms approximation of the delayed state is exploited. Based on the lifting method and the Lyapunov–Krasovskii functional (LKF) method, a new linear matrix inequality (LMI) is obtained, allowing conclusions on stability to be drawn using the scaled small gain theorem (SSG). The use of SSG theorem for stability of singular systems with time-varying delay has not been investigated elsewhere in the literature. This represents the main novelty of this article. The result is applicable for assessing the stability of both singular systems and neutral systems with time-varying delays. The less conservativeness of the stability test is illustrated by comparison with recent literature results.

In this paper, an adaptive multi-mode PWM control PSR flyback converter is proposed. In constant voltage (CV) mode, the converter adopts primary-side regulate (PSR) scheme to detect load information through the auxiliary winding, which reflects the load information on the voltage of $V_{COMP}$. The converter adjusts the switching frequency according to the voltage of $V_{COMP}$ under different load conditions, realizing adaptive multi-mode PWM control to significantly improve light- load efficiency and thus the overall average efficiency. Besides, it does not give compromise to other system performance, such as audible noise, standby power consumption, and regulation. To verify the feasibility and performance of the proposed circuit, the converter has been designed and fabricated in HHGRACE_0.35$\mu$m BCD process and verified in a 5V/1A circuit prototype. The experimental results show that 25% load power efficiency of the proposed converter is 78.1%, which is improved by up to 3.4% compared to the conventional converter.

With the rapid growth of online videos, it is crucial to generate overviews of videos to help audiences make viewing decisions and save time. Video summarization and video captioning are two of the most common solutions. In this paper, we proposed a new solution in the form of a series of scene-person pairs generated from our proposed video description scheme. This new formation takes substantially less time than watching video summaries and is more acceptable than video captions. In addition, our method can be generalized to different types of videos. We also proposed a face clustering method and a scene detection method. The experimental results indicate that our methods outperform other state-of-the-art methods and are highly generalizable. As an example, a demo application is developed to demonstrate the proposed description scheme.

Wireless sensor networks (WSNs) have attracted a lot of interests over last years in security to ensure integrity, authenticity and data confidentiality. WSN applications are numerous, which range from smart home in indoor deployment offices to smart agriculture in outdoor deployments. Due to the deployment of nodes in remote areas, the solution brought to the wireless sensor network to increase the information security consists in protecting the messages exchanged between nodes. In the paper, modifications in the key management scheme are presented. Due to resource constraints in the sensor nodes, the proposed scheme ensures a distribution of the secured keys between nodes. New session keys transmitted between sensor nodes are defined during the discovery and path key phases. The experimental results prove that the proposed scheme is adapted to the WSN applications in terms of connectivity and resiliency.

Cloud computing has become a growing technique due to its on-demand service and scalable features. Most of the applications of the cloud today are in data storage. Data protection and privacy have thus become a major concern. There are huge risks of storing transaction data on third-party hosts. Service providers need to make right investments in proving and ensuring appropriate security over time. Malicious users can exploit weak spots in the data protection model to increase unauthorized access to data. To overcome the problem, in this paper, security-aware data transaction on the cloud using the blowfish algorithm (BA) is proposed. To improve the security of the system, initially, the proposed system verifies the authentication of the user. After the authentication process, the uploaded data are initially split using a pattern-matching algorithm. Then, the separated data are encrypted with the help of BA. Finally, the encrypted data are stored in the cloud at an optimal location. This method is more secure as the data are column-separated and optimized in the cloud, so it is difficult to attack. This method is highly guaranteed because the user cannot recover the document without authentication checking. The performance of the proposed approach is analyzed in terms of different metrics, namely, memory, encryption and decryption time. The experimental results clearly say that the proposed method took minimum encryption time compared with the existing approach.

Distributed amplifiers (DAs) are one of the most important and common wideband amplifiers that can use various arrangements in their gain cell structure. One of the gain cells that can be effective in increasing the bandwidth of a distributed amplifier is the pseudo-differential amplifier (PDA). Although pseudo-differential distributed amplifiers (PDDAs) have a wide bandwidth and amplify DC, they have a small voltage gain. In this paper, various circuits with the same power and chip area are proposed to improve the performance of PDDAs. For evaluating and comparing the performance of the proposed circuits, they are implemented in 0.18$\mu$m RF-CMOS technology. The simulation results reveal that two cascaded PDDAs with three stages have a better performance than three cascaded PDDAs with two stages; in two cascaded PDDAs with three stages, a gain of 9dB can be achieved for a bandwidth of 50GHz in 0.18$\mu$m RF-CMOS technology. In this amplifier, parameters S${}_{11}$, S${}_{22}$ and S${}_{12}$ are $-$12, $-$10 and $-$18 dB, respectively; noise figure is 4.3–5.8dB, and $P_{1dB}$ is +4dBm. This amplifier consumes 220mW power and has a chip area of 0.58mm².

A new electronically tunable quadrature oscillator is presented using a biquad band pass filter. The new band pass filter is designed using two dual X second-generation multi-output current conveyors (DX-MOCCIIs), two grounded capacitors, two grounded resistors and one NMOS transistor working in the triode region. It is used to design a four-phase voltage and three-phase current output quadrature oscillator simultaneously. The frequency of the oscillator can be tuned externally through the MOS gate voltage without affecting the condition of oscillation. The phase noise, frequency stability and nonideality analysis are given. The functionality of the oscillator circuit has been confirmed by SPICE simulation and also hardware realization using commercially available IC AD844.

Controller area network (CAN) bus-based connected and even self-driving vehicles suffer severe cybersecurity challenges because connections from outside the vehicle and an existing security vulnerability on CAN expose passengers to privacy and security threats. Generative adversarial nets (GAN)-based intrusion detection systems (IDSs) for in-vehicle network can eliminate the limit of insufficient types of attack data suffered by the deep learning-based IDSs. The existing GAN-based IDS is a hybrid deep learning model built by DNN and GAN, which is too complex to have a short detection time. The evaluation performance of this model can be further improved. To mitigate this issue, we propose another GAN-based intrusion detection method for in-vehicle network, which is a single improved GAN. The proposed model can have better evaluation metrics, e.g., the testing accuracy rate is up to 99.8% and poor detection performance is addressed when a single GAN is used in intrusion detection for the in-vehicle network. In this paper, we design a new loss function for generator in GAN to enhance an ability to produce fake abnormal data, and utilize a sparse enhancement training method helping discriminator in GAN to correct the arbitration bias for fake attack data every 100 steps. In addition, we utilize fewer convolution and de-convolution layers for constructing GAN model, which can reduce the calculation time theoretically and ultimately shorten the detection time to $0.12\pm 0.03$ms for a data block built by 64 CAN messages. We evaluate this improved GAN-based intrusion detection by test set. The results demonstrate that our method has not only a capacity of five classifications, but also better evaluation performance than the existing method in the area of GAN-based IDSs for the in-vehicle network.

Leakage energy has become an increasingly large fraction of total energy consumption, making it important to reduce leakage energy for improving the overall energy efficiency for embedded processors. In this paper, we explore how to reduce the cache leakage energy efficiently in a hybrid Scratch-Pad Memory (SPM) and cache architecture. Different from stand-alone cache, since the frequently used data may be allocated to the SPM for rapid retrieval in the hybrid architecture, the access frequency to the cache is reduced. It is possible to place the cache lines of the hybrid SPM-cache into the low power mode more aggressively than traditional leakage management for regular caches, which can reduce more leakage energy without significant performance degradation. Also, we propose a Hybrid Drowsy-Gated VDD (HDG) technique, which can adaptively exploit both short and long idle intervals of cache accesses to minimize leakage energy with insignificant performance overhead. In addition, we discussed the impact of cache size on the idle intervals of accesses, which will affect the efficiency of leakage management methods that exploit the idle intervals to reduce leakage energy.

In order to address the exponential growth of data traffic, the concept of device-to-device (D2D) communication has been introduced in fifth generation (5G). The seamless handover to the finest available access network is very essential in furnishing high quality of experience (QoE) to the devices involved in D2D communication. An intelligent, fast and reliable vertical handover (VHO) decision scheme is required for D2D to achieve seamless mobility and uninterrupted service. To address these issues, a two-stage fuzzy logic-based VHO scheme is proposed in this work. Here, fuzzy logic-based handover requirement analysis (HRA) is carried out in stage 1 based on received signal strength (RSS), data rate (DR) and latency (LA). This stage minimizes the unnecessary handovers. In stage 2, fuzzy-analytic hierarchy process (Fuzzy-AHP), fuzzy-technique for order preference by similarity to ideal solutions (Fuzzy-TOPSIS) are used in ranking and selecting the target access network. Through simulations, it has been proved that the proposed scheme offers improved performance than the classical and fuzzy-based multi-attribute decision-making (MADM) schemes. The proposed scheme offers 3.45$\%$, 7.33$\%$, 13.51$\%$, 40$\%$ minimum improvement in data rate, handover decision delay, handover blocking probability and number of handovers executed over the classical MADM schemes considered.

CAN controllers are responsible for the processing and transmission of CAN signals, which render them indispensable electronic devices in automotive cyber-physical systems. SJA1000T is one of the most common CAN communication chips which can be implemented through the use of bulk silicon CMOS technology. A susceptibility test platform is built based on the direct power injection method following IEC62132-4 to explore the electromagnetic susceptibility characteristics and the effects of thermal stress on CAN controller. Test results show that the susceptibility threshold curve is corresponded directly to frequency and temperature. The influence of thermal stress on the input and output characteristics of MOSFET is simulated and analyzed to verify the anti-interference ability and reduced reliability of the chip in a high temperature environment.

This research paper introduces Unified Active Power Filter (UAPF) based on the $Z$-Source Inverter (ZSI) assisted by a Solar Photovoltaic (SPV) to improve the quality of power. This inverter offers significant benefits as it offers better buck/boost functionality, can control the phase angle output, has fewer harmonics, does not require filters and has high performance over the traditional inverter. For the Maximum Power Point Tracking (MPPT) feature, a low step-up Direct Current (DC)–DC converter module is used as a DC–DC switched power apparatus. The compensation feature of the UAPF depends on the method of control used to generate reference signals. In this proposed method, the Improved Second-Order General Integrator (ISOGI) is used to obtain the ZSI-UAPF reference current. The ISOGI-based control system only utilizes converter and load current measurement, since the feedback signals do not need voltage information and associated sensors. The proposed controller works efficiently with the SPV and battery system and enables distributed power to be coupled with loads under various operating conditions. The ZSI-based SPV-UAPF provides mitigation for long-range current and voltage interference and meets the grid’s active power requirements. The experimental findings based on a digital real-time platform justify the efficiency of PV with ZSI topology under balanced and unbalanced voltage and load conditions.

Soft errors have always been a concern in the design of digital circuits. As technology down-scales toward Nanometer sizes, emergence of aging effects, process variations, and Multiple Event Transients (METs) has made the soft error rate (SER) estimation of digital circuits very challenging. This paper intends to characterize the challenges by investigating the cross effects of theses issues in overall SER of a circuit. To this regard, we employ a simulation-based SER estimation approach in which the aging effect, process variations and METs are jointly considered in our fault injection process. In our simulation-based SER estimation approach, a statistical gate delay model is used. The fault injection results into ISCAS85 circuit benchmark reveal that the SER estimation without taking into account the aging effects, the process variations, and METs is significantly inaccurate.

As a typical cyber-physical-social system (CPSS), the waste collection system profoundly changes the current waste processing mode and greatly relieves the dilemma of waste disposal. However, the existing waste collection system does not provide the function that guides people to deliver the waste into the correct trash bin. In order to improve the efficiency of waste collection system, we propose an automatic question answering system based on convolutional neural network (CNN) to help people classify waste correctly. The construction process of automatic question answering system is divided into the following steps. We first construct a question answering dataset about waste classification, in which question answering pairs from the four waste categories (recyclable waste, harmful waste, dry waste, and wet waste) are included. After the dataset is constructed, we perform text preprocessing on the dataset, which includes denoising, Chinese word segmentation, and removing stop words. After text preprocessing, we use the Word2vec model as feature representation. Then, we construct a CNN and utilize the word embeddings as an input to train model. Finally, we deploy the trained model to the waste collection system, which can answer the question of waste classification that people ask. We also present a comparative analysis of the proposed method and traditional machine learning methods. The experiment shows that the proposed method has higher accuracy of waste classification than that of traditional machine learning methods.

The solar photovoltaic (PV) source is preferred to be functioned at low voltages. The practical systems such as grid-tied systems require a high output voltage causing a reduction in conversion efficiency. To handle this problem, this paper presents a new nonisolated boost DC–DC converter with a single switch suitable for solar PV applications. This converter comprises dual voltage multiplier (VM) cells, leakage energy recovery scheme, and coupled inductor (CI) techniques to get the desired output voltage from the converter. The double-clamp capacitors are connected to the primary side of the CI. The clamp capacitors can share the current through CI, and it ensures that the leakage inductance energy of CI can be recovered, leading to an expansion in the converter voltage gain and conversion efficiency. In addition, the clamp circuit clamps the voltage stress of the MOSFET switch, and the clamp capacitors will discharge at a specific time to increase the converter gain. A high duty cycle is not necessary for getting a high voltage gain, which avoids the diode reverse recovery issues. The converter can be operated in both continuous conduction mode (CCM) and discontinuous conduction mode (DCM). A 500-W experimental prototype is developed for experimental validation, and the supporting simulation results are presented. The maximum efficiency of the converter is 93.94%, and, at full load, the efficiency is 92.55%.

The impedance space of the continuous inverse class-F power amplifier (PA) is extended in this paper. Traditionally, for extending impedance space, a reactive term is introduced into the current waveform of the continuous inverse class-F PA. In this paper, two reactive terms are introduced into the current and voltage waveforms simultaneously. The proposed method results in an expansive impedance space for designing broadband continuous inverse class-F PAs. For validating the proposed theory, a broadband PA working over 1.2–4.0GHz is designed and fabricated. Experimental results indicate that the fabricated broadband PA delivers a saturation power of 40.2–42.8dBm and a drain efficiency of 50.7–69%.

Geometric analysis of three-dimensional (3D) surfaces with local deformations is a challenging task, required by mobile devices. In this paper, we propose a new local feature-based method derived from diffusion geometry, including a keypoint detector named persistence-based Heat Kernel Signature (pHKS), and a feature descriptor named Heat Propagation Strips (HeaPS). The pHKS detector first constructs a scalar field using the heat kernel signature function. The scalar field is generated at a small scale to capture fine geometric information of the local surface. Persistent homology is then computed to extract all the local maxima from the scalar field, and to provide a measure of persistence. Points with a high persistence are selected as pHKS keypoints. In order to describe a keypoint, an intrinsic support region is generated by the diffusion area. This support region is more robust than its geodesic distance counterpart, and provides a local surface with adaptive scale for subsequent feature description. The HeaPS descriptor is then developed by encoding the information contained in both the spatial and temporal domains of the heat kernel. We conducted several experiments to evaluate the effectiveness of the proposed method. On the TOSCA Dataset, the HeaPS descriptor achieved a high performance in terms of descriptiveness. The feature detector and descriptor were then tested on the SHREC 2010 Feature Detection and Description Dataset, and produced results that were better than the state-of-the-art methods. Finally, their application to shape retrieval was evaluated. The proposed pHKS detector and HeaPS descriptor achieved a notable improvement on the SHREC 2014 Human Dataset.

Approximate computing allows compromising accuracy to attain energy and performance efficient designs. However, the accuracy requirements of many applications change on runtime and it has been often observed that traditional approximate hardware tends to either provide unacceptable results or leads to an unnecessary computational effort. Quality scalable configurations can overcome these limitations. With the same motivation, we propose a low-power quality scalable approximate multiplier (LPQ-SAM) in this paper. This low power multiplier has various accuracy reconfigurable modes, including an accurate one and thus, it can be used for both error-resilient and exact applications. LPQ-SAM is exhaustively tested for different error metrics and it has been observed that in the approximate mode, it provides up to 19% and 55% power reduction compared to the exact Booth and Wallace multipliers, respectively. For illustration purposes, we demonstrated the effectiveness of LPQ-SAM on a real-time application, i.e., image masking.

In this paper, we propose the hybrid deep neural network-based cross-modal image and text retrieval method to explore complex cross-modal correlation by considering multi-layer learning. First, we propose intra-modal and inter-modal representations to achieve a complementary single-modal representation that preserves the correlation between the modalities. Second, we build an association between different modalities through hierarchical learning to further mine the fine-grained latent semantic association among multimodal data. The experimental results show that our algorithm substantially enhances retrieval performance and consistently outperforms four comparison methods.