Search

With high storage density and large capacity, three-dimensional (3D) NAND flash utilizing multi-level storage technology has become the mainstream storage medium. Furthermore, by storing multiple bits in each flash cell, 3D NAND flash memory can achieve much larger storage capacity. However, the threshold voltage distribution in 3D NAND flash memory tends to shift after repeated program/erase and long retention time, leading to more detection error when adopting conventional fixed read reference voltage (RRV). To address this issue, in this work we investigate error characteristics of 3D floating-gate (FG) and charge-trap (CT) NAND flash memory, including the reliability variations of different layers and pages, and threshold voltage shifting. We propose an efficient dynamic threshold voltage detection (EDTVD) scheme by exploiting the error characteristics and the features of the data writing process of NAND flash to optimize RRV. Based on the Nanocycler test platform, the test results show that our proposed scheme can significantly reduce raw bit error rates (RBER) during reading processes and the step count is relatively low. The RBER of the EDTVD scheme is almost equal to the optimal read scheme, and the number of step count is close to 3 fixed-step read scheme.

This letter proposed two novel error correction strategies that leverage the error correction and deletion correction capabilities of RRNS (redundant residue number system) codes: the RRNS+MSA (multiple sequence comparison) scheme and the RRNS+marker code scheme. These strategies aimed to address the issues of insertion, deletion and substitution in the DNA storage channel. In consideration of the specific characteristics of the DNA storage channel, the proposed study also incorporated the probabilities of deletion and insertion into the original DNA storage channel model. The simulation results demonstrated conclusively that the two error correction schemes proposed in this letter can ensure the reliability of DNA storage information. It is noteworthy that the RRNS+marker code scheme proposed in this letter demonstrated a superior balance between error correction performance and DNA reading cost.

The copyback command can be used to accelerate data migration in solid-state drives. However, the reliability of this command is not guaranteed, posing challenges for broad utilization within the consumer electronics device. The existing algorithms for predicting the copyback threshold do not consider two issues: 1) Due to page/wordline restrictions restrictions, the accumulation of errors across pages during copyback operations is asymmetrical; and 2) The controller lacks prior knowledge about the raw bit error rate (RBER) before decoding. As a result, it is impossible to predict the frequency of copyback command execution. In this paper, we explore the mathematical models of flash programs and read operations. We further develop the state transition probability matrix for wordline-level data migration within the same plane. Therefore, we propose a predictive model for the maximum copyback threshold within the same plane. To address these two issues, we conduct tests and analyse the characteristics of error accumulation in copyback using actual chips and substitute the relative entropy of the state distribution in our prediction model for RBER. By integrating these insights, we introduce an estimated copyback count prediction model (ECCPM). The simulation results demonstrate that the ECCPM can significantly reduce latency while minimally impacting write amplification.

Recently, a promising technique of aggregating massive data from distributed Internet-of-Things Nodes (IoTNs) has been proposed, which is referred to as over-the-air computation (AirComp). Since the computational accuracy degrades as the number of IoTNs increases, reconfigurable intelligent surfaces (RISs) may be considered for improving the computation performance by improving the degrees of freedom. However, the passive beamforming performance of RISs degrades in the face of environmental factors, e.g., snow, dirt, and precipitation. To address this impediment, we investigate the robust design of RIS-assisted AirComp systems under realistic imperfections. We first characterize two models of imperfections in terms of their statistical features, namely by the long-term and the short-term model. To deal with the long-term imperfections, a minimum-mean-squared-error (MMSE) estimator is harnessed first for estimating the imperfections. Then we propose an optimization algorithm to optimize the computational accuracy by alternatively updating the transmit coefficients of the IoTNs, the active receive beamforming of the fusion center (FC), and the passive beamforming of the RIS. In particular, the penalty dual decomposition (PDD) technique is invoked for updating the passive beamforming vector. As for the short-term imperfections, we resort to optimize the computational accuracy in the worst-case scenario owing to imperfect passive beamforming at the RIS. Another alternating optimization algorithm is developed for solving the associated Min-max problem. Our numerical results verify that our developed beamforming designs substantially outperform their non-robust counterparts. Specifically, our algorithm conceived for mitigating the long-term imperfections approaches the performance of idealized perfect beamforming designs relying on unimpaired RISs.

The accuracy of index-bit detection greatly affects the overall bit-error-rate (BER) performance of index modulation aided differential chaos shift keying (IM-DCSK). To improve the BER performance of index bits, a novel neural network (NN)-aided IM-DCSK detection scheme, referred to as NN-IM-DCSK detection scheme, is proposed in this article. The proposed scheme can be applied to various IM-DCSK systems, such as carrier index DCSK, pulse position modulation DCSK, and code index modulation DCSK systems. The NN-IM-DCSK detection scheme uses a two-layer long short-term memory unit and multiple fully connected layers to extract the features and the correlation of IM-DCSK signals. By estimating the index bits from the configured NN framework, the scheme can make correct correlation to retrieve modulated bits. Therefore, the NN-IM-DCSK detection scheme that benefits from the advantage of both the NN and traditional energy detection can improve transmission reliability. The complexity of the NN-IM-DCSK system is analyzed. Simulation results show that the proposed detection scheme can achieve better BER performance than the traditional detection scheme, such as the energy detection scheme, in IM-DCSK systems over multipath Rayleigh fading channels.

In 3-D NAND flash memory, the channel is always offset due to the complicated interference from the program/erase (PE), including the data retention and layer interference, so that the channel estimation is desired. However, due to the physical structure of 3-D flash memory, there exists significant variation in channel parameters and inconsistent channel offsets among different wordlines. As a result, the process of estimating channel parameters for each individual wordline typically requires a considerable amount of computational resources, resulting in the high latency of system, which becomes a new challenge. To tackle this problem, two unsupervised learning algorithms are proposed to estimate the channel parameters, based on analyzing the error distribution in 3-D flash memory. To address the longer read latency introduced by the unsupervised learning algorithm for the channel estimation, we further propose a low-latency detection algorithm, which first detects whether the current channel needs to be updated. In the event that an update is required, the algorithm only periodically estimates the channel parameters during system idle times, resulting in a more efficient and streamlined process. Compared to the existing methods, the proposed algorithms can efficiently estimate channel parameters with lower-computational complexity. Moreover, combining with the search algorithm, a correction scheme is proposed to minimize the error between the estimated read reference voltage (RRV) and the optimal RRV in the actual device. Theoretical analysis and simulation results demonstrate that the proposed method can improve the lifetime of flash memory and reduces the number of read retries.

Low-density parity-check (LDPC) codes have been widely adopted to guarantee data reliability in 3-D NAND flash memory. However, the iterative LDPC decoding algorithm leads to high-decoding latency due to the iterative message transfer mechanism. Using a field-programmable gate array (FPGA) testbed, we first present the binary channel in NAND flash and analyze the single-symbol log-likelihood ratio (LLR) variation with the decoding iterations. Subsequently, we investigate the raw bit error ratio (RBER) characteristics of intrapage frames. To reduce the number of iterative decoding, we propose a frame feedback information aware decoding algorithm (FFIA-DA), combined with the single-symbol LLR variation and the similar error characteristics among intrapage frames. The proposed method uses the decoding feedback information of one frame to decrease the number of decoding iterations of other frames with similar RBER. Experiments show that the proposed approach can improve the decoding performance of LDPC and speed up decoding convergence.

In the protograph-coded short reference multifold rate differential chaotic shift keying (SRMR-DCSK) systems, the signal transmission reliability is affected by the channel noise, especially when the noise is correlated. To solve this problem, this letter proposes a fully convolutional residual network (FCResNet)-based iterative detection structure to suppress the correlated noise in the protograph-coded SRMR-DCSK systems. Specifically, we design a new FCResNet tailored for the considered systems to accurately extract the characteristics of correlated noise, and thus the correlated noise can be easily removed from the received signal in an iterative fashion. In addition, we derive the log-likelihood ratio (LLR) expression of the protograph-coded SRMR-DCSK systems to facilitate the proposed detection. Simulation results demonstrate that the proposed FCResNet-based detection structure can obtain more desirable performance gain compared with the existing detection over Gaussian and multipath Rayleigh fading channels.

In this letter, we investigate the robust beamforming design for a hybrid active-passive reconfigurable intelligent surface (RIS) assisted integrated sensing and communication (ISAC) system, where the RIS consists of both active and passive reflecting elements. Different from the existing works, we consider the target location estimation error. We aim to maximize the minimum sensing beampattern gain among multiple targets, subject to the signal-to-interference-plus-noise ratio (SINR) constraint for each communication user (CU), the transmit power constraints for the BS and the active elements. To address the non-convexity of the max-min RIS-assisted ISAC design problem, we propose an alternating-optimization based design solution to alternately optimize the BS’s transmit beamforming design and the hybrid RIS’s coefficient matrix in a low-complexity manner. Simulation results demonstrate the benefit of the proposed design scheme over the existing baseline schemes.

In orthogonal frequency-division multiplexing (OFDM) systems, channel estimation is required for guaranteeing the transmission quality over rapidly changing wireless channels. At the same time, it is crucial to recover the clipping signal when resolving the high peak average power ratio (PAPR) of the transmitted OFDM signal. In this letter, we apply the deep learning (DL) technique to boost the performance of channel estimation in the protograph low-density parity-check (PLDPC)-coded bit-interleaved coded modulation (BICM)-aided OFDM (BICM-OFDM) systems. Specifically, we propose a channel estimation residual network (CERNet) to estimate the channel state information (CSI). Furthermore, a signal recovery residual network (SRRNet) is designed to recover the clipped signal. Our proposed residual networks can capture the frequency-specific features of the channel matrices and the correlation of the clipping noise over the received data signal adequately without relying on explicit priori channel information to further improve the channel estimation performance. Both analytical and simulation results show that the CERNet achieves superior performance compared with the conventional estimation scheme, and the joint CERNet and SRRNet can further improve the system performance.

In consumer electronics, 3-D NAND flash memory has become the state-of-the-art storage medium due to the multi-bit storage per cell and 3-D stacking technology. However, flash memory cells are commonly vulnerable to numerous types of circuit-level noise, resulting in threshold voltage distribution distortion and decreased reliability. Characterization of the threshold voltage distribution and modeling of optimal read reference voltages can help to mitigate the reliability deterioration. We provide a solution for improving the reliability of flash memories while reducing their implementation cost in this paper. The proposed pre-processing method is shown to yield a smaller number of scanning voltage steps reducing the time overhead and energy consumption of multiple read operations during the characterization of threshold voltage distributions. We propose a single state asynchronous estimation (SSAE) method to estimate optimal read reference voltages. Compared with the advanced two-state asynchronous estimation (TSAE) method, it is shown to reduce the computational and space overhead by half. Furthermore, compared with using the default read reference voltages, it is also shown to reduce RBER without exact knowledge of noise.

With the rapid growth of data, DNA is attracting attention as a promising storage medium due to its durability, large storage capacity, and high bulk density. However, errors occur in DNA synthesis and sequencing processes, with substitutions being the most common errors in DNA storage process. To address this issue, we select low-density parity-check (LDPC) codes as the error correction method to guarantee data reliability. However, unlike traditional data storage channels, errors occurring between nucleotides in the DNA channel are asymmetric. Therefore, traditional belief propagation (BP) decoding algorithm is not sufficient for effective error correction under DNA channel. In this letter, we propose to use two LDPC codes and take error types as additional information for calibrating log-likelihood ratio (LLR) value. In addition, compared with previous works, simulation results show that our proposed decoding algorithm can reduce bit error rate (BER) by $33\% \sim 94\% $ .

An $(n,k)$ recoverable property array code is composed of $m\times n$ arrays such that any $k$ out of $n$ columns suffice to retrieve all the information symbols, where $n > k$ . Note that maximum distance separable (MDS) array code is a special $(n,k)$ recoverable property array code of size $m\times n$ with the number of information symbols being $km$ . Expanded-Blaum-Roth (EBR) codes and Expanded-Independent-Parity (EIP) codes are two classes of $(n,k)$ recoverable property array codes that can repair any one symbol in a column by locally accessing some other symbols within the column, where the number of symbols $m$ in a column is a prime number. By generalizing the constructions of EBR and EIP codes, we propose new $(n,k)$ recoverable property array codes, such that any one symbol can be locally recovered and the number of symbols in a column can be not only a prime number but also a power of an odd prime number. Also, we present an efficient encoding/decoding method for the proposed generalized EBR (GEBR) and generalized EIP (GEIP) codes based on the LU factorization of a Vandermonde matrix. We show that the proposed decoding method has less computational complexity than existing methods. Furthermore, we show that the proposed GEBR codes have both a larger minimum symbol distance and a larger recovery ability of erased lines for some parameters when compared to EBR codes. We also present a necessary and sufficient condition of enabling EBR codes to recover any $r$ erased lines of a slope for any parameter $r$ , which was an open problem. Moreover, we show that EBR codes can recover any $r$ consecutive erased lines of any slope for any parameter $r$ .

Vehicular edge computing (VEC) is a new computing paradigm that enhances vehicular performance by introducing both computation offloading and service caching, to resource-constrained vehicles and ubiquitous edge servers. Recent developments of autonomous vehicles enable a variety of applications that demand high computing resources and low latency, such as automatic driving, auto navigation, etc. However, the highly dynamic topology of vehicular networks and limited caching space at resource-constrained edge servers calls for intelligent design of caching placement and computation offloading. Meanwhile, service caching decisions are highly correlated to the computation offloading decisions, which pose a great challenge to effectively design service caching and computation offloading strategies. In this paper, we investigate a joint optimization problem by integrating service caching and computation offloading in a general VEC scenario with time-varying task requests. To minimize the average task processing delay, we formulate the problem using long-term mixed integer non-linear programming (MINLP) and propose an algorithm based on deep reinforcement learning to obtain a suboptimal solution with low computation complexity. The simulation results demonstrate that our proposed scheme exhibits an effective performance improvement in task processing delay compared with other representative benchmark methods.

The performance of over-the-air computation (AirComp) systems degrades due to the hostile channel conditions of wireless devices (WDs), which can be significantly improved by the employment of reconfigurable intelligent surfaces (RISs). However, the conventional RISs require that the WDs have to be located in the half-plane of the reflection space, which restricts their potential benefits. To address this issue, the novel family of simultaneously transmitting and reflecting reconfigurable intelligent surfaces (STAR-RIS) is considered in AirComp systems to improve the computation accuracy across a wide coverage area. To minimize the computation mean-squared-error (MSE) in STAR-RIS assisted AirComp systems, we propose a joint beamforming design for optimizing both the transmit power at the WDs, as well as the passive reflect and transmit beamforming matrices at the STAR-RIS, and the receive beamforming vector at the fusion center (FC). Specifically, in the updates of the passive reflect and transmit beamforming matrices, closed-form solutions are derived by introducing an auxiliary variable and exploiting the coupled binary phase-shift conditions. Moreover, by assuming that the number of antennas at the FC and that of elements at the STAR-RIS/RIS are sufficiently high, we theoretically prove that the STAR-RIS assisted AirComp systems provide higher computation accuracy than the conventional RIS assisted systems. Our numerical results show that the proposed beamforming design outperforms the benchmark schemes relying on random phase-shift constraints and the deployment of conventional RIS. Moreover, its performance is close to the lower bound achieved by the beamforming design based on the STAR-RIS dispensing with coupled phase-shift constraints.

Over-the-air computation (AirComp) has received substantial attention, given its ability to aggregate massive amounts of data from distributed wireless devices (WDs). However, the computation accuracy at the fusion center (FC) may be severely affected by receiving data corrupted by the poor channel conditions. To mitigate this issue, we consider the employment of reconfigurable intelligent surfaces (RISs) in the AirComp system considered for improving the quality of received data, and hence improve the computation accuracy. However, most previous contributions on RIS-assisted AirComp systems only employ a single RIS in the resultant single-RIS-assisted (SRIS-assisted) AirComp systems. We develop this concept further for mitigating the deleterious channel effects by conceiving a double-RIS-assisted (DRIS-assisted) AirComp system, where one of the RISs is located near the WDs and the other in the vicinity of the FC. We theoretically prove that the DRIS-assisted AirComp system outperforms its SRIS-assisted counterpart in terms of the resultant computation mean-squared-error (MSE). Furthermore, we propose a pair of algorithms for jointly optimizing the transmit power at the WDs, the receive beamforming vector at the FC, and the passive beamforming matrices at the RISs for minimizing the computational MSE. Specifically, the transmit power is updated by exploiting the Lagrange duality method, while the receive beamforming vector is optimized by utilizing the first-order optimality condition. Furthermore, a pair of techniques are developed for optimizing the passive beamforming matrices at the RISs based on semidefinite relaxation (SDR) and penalty-duality-decomposition (PDD), respectively. Both the complexity and the convergence of the proposed algorithms are analyzed. Finally, simulation results are provided for quantifying the overall performance of the resultant DRIS-assisted AirComp system.

In this work, a novel electrochemically switched ion permselective membrane (ESIPM) has been applied to the recovery of lithium from brine in the electrochemically switched ion permselective (ESIP) membrane separation system. This ESIPM was prepared by multi-layer assembly, with polyacrylonitrile ultrafiltration membrane (PAN) as the base membrane, and CNTs/CS membranes pressed and filtered on the base membrane as the current collector, crosslinking a lithium manganate/carbon black/polyvinyl alcohol/polyacrylic acid (LiMn2O4/C/PVA/PAA) composite material, which coated on a polyacrylonitrile ultrafiltration membrane/carbon nanotubes (PAN/CNTs) composite membrane as the electroactive lithium ion exchange layer. In the ESIP system, by alternately applying oscillation potentials between the ESIPM membrane and counter-electrodes to regulate the placement and release of the target ions on the membrane, Meanwhile the target ions are driven by the electric field to be enriched in the ESIPM and directionally transported through the ESIPM. As a result, it was found that compared with the electrodialysis (ED) system, the flux of Li+in ESIPM in the ESIP system increased by 12 times and the selectivity factor for Li+/Mg2+increased by 1.4 times for the brine with Mg2+/Li+=1. Besides, the selectivity factor for Li+/Mg2+reaches 4.4 for the brine with Mg2+/Li+=20 in ESIPM in the ESIP system, which was 2.2 times higher than that of the commercial ion exchange membrane CIMS in the ED system.

Advances in single-cell mass cytometry have increasingly improved highly multidimensional characterization of immune cell heterogeneity. The immunoassay multiplexing capacity relies on monoclonal antibodies labeled with stable heavy-metal isotopes. To date, a variety of rare-earth elements and noble and post-transition metal isotopes have been used in mass cytometry; nevertheless, the methods used for antibody conjugation differ because of the individual metal coordination chemistries and distinct stabilities of various metal cations. Herein, we provide three optimized protocols for conjugating monoclonal IgG antibodies with 48 high-purity heavy-metal isotopes: (i) 38 isotopes of lanthanides, 2 isotopes of indium, and 1 isotope of yttrium; (ii) 6 isotopes of palladium; and (iii) 1 isotope of bismuth. Bifunctional chelating agents containing coordinative ligands of monomeric DOTA (1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid) or polymeric pentetic acid (DTPA) were used to stably sequester isotopic cations in aqueous solutions and were subsequently coupled to IgG antibodies using site-specific biorthogonal reactions. Furthermore, quantification methods based on antibody inherent absorption at 280 nm and on extrinsic absorption at 562 nm after staining with bicinchoninic acid (BCA) are reported to determine metal-isotope-tagged antibodies. In addition, a freeze-drying procedure to prepare palladium isotopic mass tags is described. To demonstrate the utility, experiments using six palladium-tagged CD45 antibodies for barcoding assays of live immune cells in cytometry by time-of-flight (CyTOF) are described. Conjugation of pure isotopes of lanthanides, indium, or yttrium takes ~3.5 h. Conjugation of bismuth takes ~4 h. Preparation of palladium mass tags takes ~8 h. Conjugation of pure isotopes of palladium takes ~2.5 h. Antibody titration takes ~4 h.

We have performed an in-depth single-cell phenotypic characterization of high-grade serous ovarian cancer (HGSOC) by multiparametric mass cytometry (CyTOF). Using a CyTOF antibody panel to interrogate features of HGSOC biology, combined with unsupervised computational analysis, we identified noteworthy cell types co-occurring across the tumors. In addition to a dominant cell subset, each tumor harbored rarer cell phenotypes. One such group co-expressed E-cadherin and vimentin (EV), suggesting their potential role in epithelial mesenchymal transition, which was substantiated by pairwise correlation analyses. Furthermore, tumors from patients with poorer outcome had an increased frequency of another rare cell type that co-expressed vimentin, HE4, and cMyc. These poorer-outcome tumors also populated more cell phenotypes, as quantified by Simpson’s diversity index. Thus, despite the recognized genomic complexity of the disease, the specific cell phenotypes uncovered here offer a focus for therapeutic intervention and disease monitoring.

With the development of manufacture technology, the multi-level cell (MLC) technique dramatically increases the storage density of NAND flash memory. As the result, cell-to-cell interference (CCI) becomes more serious and hence causes an increase in the raw bit error rate of data stored in the cells. Recently, low-density parity-check (LDPC) codes have appeared to be a promising solution to combat the interference of MLC NAND flash memory. However, the decoding complexity of the sum-product algorithm (SPA) is extremely high. In this paper, to improve the accuracy of the log likelihood ratio (LLR) information of each bit in each NAND flash memory cell, we adopt a non-uniform detection (N-UD) which uses the average maximum mutual information to determine the value of the soft-decision reference voltages. Furthermore, with an aim to reduce the decoding complexity and improve the decoding performance, we propose a modified soft reliability-based iterative majority-logic decoding (MSRBI-MLGD) algorithm, which uses a non-uniform quantizer based on power function to decode LDPC codes. Simulation results show that our design can offer a desirable trade-off between the performance and complexity for high-column-weight LDPC-coded MLC NAND flash memory.

Staying single: A DNA hybridization assay employs gold‐nanoparticle (AuNP) probes and single‐particle inductively coupled plasma spectrometry (ICP‐MS; see picture). The degree of aggregation of AuNPs is characterized by simultaneous measurement of decreased concentrations of the AuNP population and increased individual sizes.

Eins, zwei, drei: Ein DNA‐Hybridisierungsassay nutzt Goldnanopartikel(AuNP)‐Sonden und Massenspektrometrie einzelner Partikel mithilfe eines induktiv gekoppelten Plasmas (ICP‐MS; siehe Bild). Der Aggregationsgrad der AuNPs wird durch simultane Messung der Größe der einzelnen Partikel und der Verringerung der AuNP‐Konzentration bestimmt.

No Abstract

Welch ein Bild!Siegelmuster auf Gemälden und Kalligraphien wurden mit einer massenspektrometrischen Technik abgebildet. Dabei kam eine Niedertemperatur‐Plasmasonde zum Einsatz, die eine gute räumliche Auflösung liefert. Mit dieser Methode kann zwischen echten und gefälschten Siegeln unterschieden werden.