Search

As a low-power and low-cost wireless protocol, the promising ZigBee has been widely used in sensor networks and cyber-physical systems. Since ZigBee based networks usually adopt tree or cluster topology, the convergecast scenarios are common in which multiple transmitters send packets to one receiver, leading to the severe collision problem. The conventional ZigBee adopts carrier sense multiple access with collisions avoidance to avoid collisions, which introduces additional time/energy overhead. The state-of-the-art methods resolve collisions instead of avoidance, in which mZig decomposes a collision by the collision itself and reZig decodes a collision by comparing with reference waveforms. However, mZig falls into high decoding errors only exploiting the signal amplitudes while reZig incurs high computational complexity for waveform comparison. In this paper, we propose CmZig to embrace channel estimation in multiple-packet reception (MPR) of ZigBee, which effectively improves MPR via lightweight computing used for channel estimation and collision decomposition. First, CmZig enables accurate collision decomposition with low computation complexity, which uses the estimated channel parameters modeling both signal amplitude and phase. Second, CmZig adopts reference waveform comparison only for collisions without chip-level time offsets, instead of the complex machine learning based method. We implement CmZig on USRP-N210 and establish a six-node testbed.

The huge amount of data produced in the 6G networks not only brings new challenges to the reliability and efficiency of mobile devices but also drives rapid development of new storage techniques. With the benefits of fast access speed and high reliability, NAND flash memory has become a promising storage solution for the 6G networks. In this paper, we investigate a protograph-coded bit-interleaved coded modulation with iterative detection and decoding (BICM-ID) utilizing irregular mapping (IM) in the NAND flash-memory systems. First, we propose an enhanced protograph-based extrinsic information transfer (EPEXIT) algorithm to facilitate the analysis of protograph codes in the IM-BICM-ID systems. With the use of EPEXIT algorithm, a simple design method is conceived for the construction of a family of high-rate protograph codes, called irregular-mapped accumulate-repeat-accumulate (IMARA) codes, which possess excellent decoding thresholds and linear-minimum-distance-growth property. Furthermore, motivated by the voltage-region iterative gain characteristics of IM-BICM-ID systems, a novel read-voltage optimization scheme is developed to acquire accurate read-voltage levels, thus minimizing the decoding thresholds (in dB) of protograph codes. Analyses and simulations indicate that the proposed IMARA-aided IM-BICM-ID scheme and read-voltage optimization scheme remarkably improve the convergence and decoding performance of flash-memory systems. Thus, the proposed protograph-coded IM-BICM-ID can be viewed as a reliable and efficient storage solution for the new-generation mobile networks, such as Internet of Vehicles.

We consider the multiuser detection (MUD) problem, i.e., how to separate and decode colliding data streams, in the uplink of massive Machine Type Communications (mMTC) at millimeter wave (mmWave). Operating on factor-graphs by passing messages, the sum-product algorithm and its variants are widely applied in many other scenarios. However, in this paper, we find that their performance in mMTC at mmWave could be dramatically degraded due to the ill-conditioned MUD channel gain matrix and the existence of enormous short cycles in their corresponding factor-graphs, which are caused by the limited scattering of mmWave and the sharing of a same codebook for error correction among densely located user equipments. Assuming LDPC codes are used for error correction, we further propose a novel sum-product based approach to dealing with the MUD problem in mMTC at mmWave. It first leverages the propagation characteristics of mmWave to optimize the factor-graph for MUD by removing short cycles based on node-split and node-contraction, and then takes a dynamic-programming based method to approximate the messages passing on the resulted factor-graph, which can achieve a higher decoding accuracy. Extensive simulation results show that our approach outperforms the state-of-the-art sum-product based approaches significantly.

Absorbing sets are combinatorial structures in the Tanner graphs of low-density parity-check (LDPC) codes that have been shown to inhibit the high signal-to-noise ratio performance of iterative decoders over many communication channels. Absorbing sets of minimum size are the most likely to cause errors, and thus have been the focus of much research. In this paper, we determine the sizes of absorbing sets that can occur in general and left-regular LDPC code graphs, with emphasis on the range of \begin{document}$ b $\end{document} for a given \begin{document}$ a $\end{document} for which an \begin{document}$ (a,b) $\end{document} -absorbing set may exist. We identify certain cases of extremal absorbing sets that are elementary, a particularly harmful class of absorbing sets, and also introduce the notion of minimal absorbing sets which will help in designing absorbing set removal algorithms.

Malware propagation by adversaries has witnessed many issues across the globe. Often it is found that malware is released in different countries for monetary gains. With the proliferation of malware spreading activities, it is made possible that now we have malware patterns that are used for training machine learning models. Thus machine learning became indispensable for malware detection. The traditional machine learning models have limitations in performance as the training depth is limited. The emergence of deep learning models paved way for more training possibilities and improvement in detection accuracy with least false positives. This paper reviews literature on deep learning techniques that are used for malware detection. The deep learning methods used for malware detection include CNN, RNN, LSTM and auto encoders. LSTM is found to have memory in the cell to have better possibilities. Auto encoders are found to have better unsupervised approach with encoding and decoding to arrive at abnormalities (malware) detection. There are many contributions found using machine learning and deep learning towards Android malware detection. This paper provides knowledge that leads to further research in deep learning which is essential to improve the state of the art.

MotivationUsing DNA as the medium to store information has recently been recognized as a promising solution for long-term data storage. While several system prototypes have been demonstrated, the error characteristics in DNA data storage are discussed with limited content. Due to the data and process variations from experiment to experiment, the error variation and its effect on data recovery remain to be uncovered. To close the gap, we systematically investigate the storage channel, i.e., error characteristics in the storage process.ResultsWe first propose a new concept named sequence corruption to unify the error characteristics into the sequence level, easing the channel analysis. Then we derived the formulations of the data imperfection at the decoder including both sequence loss and sequence corruption, revealing the decoding demand and monitoring the data recovery. Furthermore, we extensively explored several data-dependent unevenness observed in the base error patterns and studied a few potential factors and their impacts on the data imperfection at the decoder both theoretically and experimentally. The results presented here introduce a more comprehensive channel model and offer a new angle towards the data recovery issue in DNA data storage by further elucidating the error characteristics of the storage process.Contactpoh.chuehloo@nus.edu.sg

In recent years, Light Field (LF) video has grabbed much attention as an emerging form of immersive media. LF collects, through a lens matrix, light information emanating in every direction, and obtains rich information about the scene, providing users with an immersive 6 Degrees of Freedom (DoF) experience. The visual content between different viewpoints is highly homogenized, suggesting the possibility of good compression and encoding. However, most fixed-structure LF coding schemes are difficult to adapt to the real-time requirements of different LF applications and best-effort network conditions causing packet loss. In this paper, we propose a dynamic adaptive LF video transmission scheme that can achieve high compression and yet provide near-distortion-free LF video when the network condition is stable. Additionally, for unstable network conditions a description scheduling algorithm is proposed, which can decode the LF video with the highest possible quality even if partial data cannot be received completely and/or timely. We achieve this by designing a Multiple Description Coding (MDC) based solution to transport the LF video compressed by a Graph Neural Network (GNN) model. Experimental results show that the scheduling algorithm can improve the quality of the decoding results by 3% to 15%. Compared with other similar schemes, our system greatly improves the reliability of the video streaming system against packet loss/error and supports heterogeneous receivers.

String prediction (SP) is a very efficient screen content coding (SCC) tool which has been adopted in the third generation of Audio Video Standard (AVS3). It is observed that two special types of strings occur frequently. To further improve the coding efficiency for SCC on top of the original SP, a new variation of SP named Equal-value-string and Copy-above-string based SP (ECSP) is proposed. An ECSP coding unit uses only three types of string: Equal-value-string, Copy-above-string, and Unpredictable-pixel-string. Compared with the AVS3 reference software HPM9.0 with ECSP disabled, using AVS3 SCC Common Test Condition and YUV 4:2:0 test sequences, the proposed technique achieves an average Y BD-rate reduction of 5.54% and 3.01% for All Intra and Low Delay configurations, espectively, with low additional encoding and decoding complexity. The proposed ECSP has been adopted in the AVS3 standard.

This paper presents the implementation and experimental evaluation of a short-packet physical-layer network coding (PNC) system. Implementation of short-packet PNC systems is challenging. First, short packets may have only a few pilot symbols for synchronization and channel estimation purposes. Increasing the number of pilots increases the overhead; decreasing the number of pilots, on the other hand, degrades the packet error rate performance. Second, many short-packet systems are meant for applications with very stringent delay requirements. Employing advanced but complex PNC channel decoding may result in unacceptable delay due to the processing delay. This work presents a low-complexity and low-overhead physical-layer design of OFDM-based short-packet PNC systems, implemented over the software-defined radio platform. Our design makes use of only a small number of pilots (without separate OFDM preamble symbols) to address issues such as slot synchronization, packet detection, carrier frequency offsets, and mismatched channel state information. Our design employs reduced-complexity XOR channel decoding based code-aided parameter estimation (that includes synchronization and channel estimation) to compensate for the limitations imposed by having a small number of pilots. This is the first demonstration that provides a practical framework for applying PNC to short-packet communications.

A long-standing vision of backscatter networks is to provide long-range connectivity and high-speed transmissions for batteryless Internet-of-Things (IoT). Recent years have seen major innovations in designing backscatter networks toward this goal. Yet, they either operate at a very short range, or experience extremely low throughput. This paper takes one step further towards breaking this stalemate, by presenting PolarScatter that exploits channel polarization in long-range backscatter networks. We transform backscatter channels into nearly noiseless virtual channels through channel polarization, and convey bits with extremely low error probability. Specifically, we propose a new polar code scheme that automatically adapts itself to different channel quality, and design a low-cost encoder to accommodate polar codes on resource-constrained backscatter tags. Furthermore, we devise a new metric to calculate log-likelihood ratio for accurate decoding, and present a stopping criterion of iterations to reduce decoding latency. We build a prototype PCB tag, and our experiments show that it achieves up to 11.5 throughput improvement over the state-of-the-art long-range backscatter solution. We also simulate an IC design in TSMC 65 nm LP CMOS process. Compared with traditional encoders, our encoder reduces storage overhead by three orders of magnitude, and lowers the power consumption to tens of microwatts.

Mobile edge computing and nonorthogonal multiple access (NOMA) have been considered as promising technologies that can satisfy rigorous requirements of industrial internet of things systems. However, system dynamics, including channel states and computation task requests, may continuously change NOMA decoding order and computation uploading time, making it difficult to reduce latency using conventional highly complex optimization methods. In this study, we investigate a novel scheme that effectively reduces the average task delay to improve the quality of service for all users by jointly optimizing subchannel assignment (SA), offloading decision (OD), and computation resource allocation (CRA). To deal with the high complexity, the original multi-server problem is first decomposed into multiple single-server problems. Subsequently, each single-server problem is decoupled into CRA and SA/OD subproblems. Using convex optimization, a closed-form solution is derived for the optimal CRA action. Concurrently, the optimal SA/OD action is obtained using a distributed multi-agent deep reinforcement learning algorithm. Simulation results reveal that the proposed scheme significantly outperforms the state-of-the-art schemes. In particular, it reduces the action decision duration by 30 times while achieving a near-optimal performance of up to 97% of the optimum under the exhaustive search scheme.

Indoor localization with high accuracy and efficiency has attracted much attention. Due to visible light communication (VLC), the LED lights in buildings, once modulated, hold great potential to be ubiquitous indoor localization infrastructure. However, this entails retrofitting the lighting system and is hence costly in wide adoption. To alleviate this problem, we propose to exploit modulated LEDs and existing unmodulated lights as landmarks. On this basis, we present a novel inertial-aided visible light positioning (VLP) system for lightweight indoor localization on resource-constrained platforms, such as service robots and mobile devices. With blob detection, tracking, and VLC decoding on rolling-shutter camera images, a visual front end extracts two types of blob features, i.e., mapped landmarks (MLs) and opportunistic features (OFs). These are tightly fused with inertial measurements in a stochastic cloning sliding-window extended Kalman filter (EKF) for localization. We evaluate the system by extensive experiments. The results show that it can provide lightweight, accurate, and robust global pose estimates in real time. Compared with our previous ML-only inertial-aided VLP solution, the proposed system has superior performance in terms of positional accuracy and robustness under challenging light configurations, such as sparse ML/OF distribution.

In this paper, we propose a novel successive cancellation flip (SCF) decoding to reduce the computational complexity compared to the conventional SCF decoding by using distributed CRC bits. The proposed decoding reduces the number of estimations for information bits by early termination of decodings for failed frames of the first SC decoding, while trying to minimize the additional sorting operations. Simulation results show that the proposed SCF decoding reduces the computational complexity of repeated SC decoding at least 27% compared to the conventional SCF decoding.

Wireless Multimedia Sensor Network (WMSN) is increasingly being deployed for surveillance, monitoring and Internet-of-Things (IoT) sensing applications where a set of cameras capture and compress local images and then transmit the data to a remote controller. Such captured local images may also be compressed in a multi-view fashion to reduce the redundancy among overlapping views. In this paper, we present a novel paradigm for compressed-sensing-enabled multi-view coding and streaming in WMSN. We first propose a new encoding and decoding architecture for multi-view video systems based on Compressed Sensing (CS) principles, composed of cooperative sparsity-aware block-level rate-adaptive encoders, feedback channels and independent decoders. The proposed architecture leverages the properties of CS to overcome many limitations of traditional encoding techniques, specifically massive storage requirements and high computational complexity. Then, we present a modeling framework that exploits the aforementioned coding architecture. The proposed mathematical problem minimizes the power consumption by jointly determining the encoding rate and multi-path rate allocation subject to distortion and energy constraints. Extensive performance evaluation results show that the proposed framework is able to transmit multi-view streams with guaranteed video quality at lower power consumption.

Previous neural network accelerators tailored to video analysis only accept data of RGB/YUV domain, requiring decompressing the video that are often compressed before transmitted from the edge sensors. A compressed video processing accelerator can alleviate the decoding overhead, and gain performance speedup by operating on more compact input data. This work proposes a novel deep learning accelerator architecture, Alchemist, which is able to predict results directly from the compressed video bitstream instead of reconstructing the full RGB images. By utilizing the metadata of motion vector and critical blocks extracted from bitstreams, Alchemist contributes to remarkable performance speedup of 5x with negligible accuracy loss. Nevertheless, we still find that the original compressed video coded by standard algorithms such as H.264 is not suitable to be directly manipulated, due to diverse compressed structures. Although obviating the requirement to recover all RGB frames, the accelerator must parse the entire compressed video bitstream to locate reference frames and extract useful metadata. If we combine the video codec with the proposed compressed video analysis, additional optimizations can be obtained. Therefore, to cope with the mismatch between current video coding algorithms such as H.264 and neural network-based video analysis, we propose a specialized coding strategy to generate compressed video bitstreams more suitable for transmission and analysis, which further simplifies the decoding stage of video analysis and is capable of achieving significant storage reduction.

This paper proposes a real-time multi-scale semantic segmentation network (MsNet). MsNet is a combination of our novel multi-scale fusion with matching attention model (MFMA) as the decoding network and the network searched by asymptotic neural architecture search (ANAS) or MobileNetV3 as the encoding network. The MFMA not only extracts low level spatial features from multi-scale inputs but also decodes the contextual features extracted by ANAS. Specifically, considering the advantages and disadvantages of the addition fusion and concatenation fusion, we design multi-scale fusion (MF) that balances speed and accuracy. Then we creatively design two matching attention mechanisms (MA), including matching attention with low calculation (MALC) mechanism and matching attention with strong global context modeling (MASG) mechanism, to match varying resolutions and information of features at different levels of a network. Besides, the ANAS performs the deep neural network search by employing an asymptotic method and provide an efficient encoding network for MsNet, releasing researchers from those tedious mechanical trials. Through extensive experiments, we prove that MFMA, which can be applied to numerous recognition tasks, possesses excellent decoding ability. And we demonstrate the effectiveness and necessity of implementing the ``matching'' attention mechanism. Finally, the proposed two versions, MsNet₋ANAS and MsNet₋M achieve a new state-of-the-art trade-off between accuracy and speed on the CamVid and Cityscapes datasets. More remarkably, on the Nvidia Tesla V100 GPU, our MsNet₋ANAS achieves 74.1% mIoU with the speed of 184.2 FPS on the CamVid while 72.9% mIoU with the speed of 119.9 FPS on the Cityscapes.

Privacy protection is the key to maintaining the Internet of Things communication strategy. Steganography is an important way to achieve covert communication that protects user data privacy. Steganalysis technology is the key technology to check steganography security, and its ultimate goal is extracting embedded messages. Existing methods cannot extract under known cover images. To this end, this paper proposes a method of extracting embedded messages under known cover images. First, the syndrome-trellis encoding process is analyzed. Second, a decoding path in the syndrome trellis is obtained by using the stego sequence and a certain parity-check matrix, while the embedding process is simulated using the cover sequence and parity-check matrix. Since the decoding path obtained by the stego sequence and the correct parity-check matrix is optimal and has the least distortion, comparing the path consistency can quickly filter the coding parameters, so as to determine the correct matrices, and embedded messages can be extracted correctly. The proposed method does not need to embed all possible messages for the second time, improving coding parameter recognition significantly. The experimental results show that the proposed method can identify syndrome-trellis coding parameters in stego images embedded by adaptive steganography quickly to realize embedded message extraction.

Accurate electroencephalogram (EEG) pattern decoding for specific mental tasks is one of the key steps for the development of brain-computer interface (BCI), which is quite challenging due to the considerably low signal-to-noise ratio of EEG collected at the brain scalp. Machine learning provides a promising technique to optimize EEG patterns toward better decoding accuracy. However, existing algorithms do not effectively explore the underlying data structure capturing the true EEG sample distribution and, hence, can only yield a suboptimal decoding accuracy. To uncover the intrinsic distribution structure of EEG data, we propose a clustering-based multitask feature learning algorithm for improved EEG pattern decoding. Specifically, we perform affinity propagation-based clustering to explore the subclasses (i.e., clusters) in each of the original classes and then assign each subclass a unique label based on a one-versus-all encoding strategy. With the encoded label matrix, we devise a novel multitask learning algorithm by exploiting the subclass relationship to jointly optimize the EEG pattern features from the uncovered subclasses. We then train a linear support vector machine with the optimized features for EEG pattern decoding. Extensive experimental studies are conducted on three EEG data sets to validate the effectiveness of our algorithm in comparison with other state-of-the-art approaches. The improved experimental results demonstrate the outstanding superiority of our algorithm, suggesting its prominent performance for EEG pattern decoding in BCI applications.

Parameter Identification plays an important role in electric power transmission systems. Existing approaches for parameter identification tasks typically have two limitations: (1) They generally ignored development trend of historical data, and did not mine characteristics of corresponding power grid branches. (2) They did not consider the constraints of power grid topology, and treated different branches independently. Therefore, they could not characterize correlations between the center node and its neighborhoods. To overcome these limitations, this work proposes a multi-task graph convolutional neural network (MT-GCN) which utilizes the graph convolutional network (GCN) and the fully convolutional network (FCN) as building blocks for parameter identification. Specially, GCN can extract the structure information to enhance local feature extraction. FCN is a decoding module following GCN module, and it is used to identify the parameters of each branch according to its characteristics. Compared with previous methods, the proposed method is significantly improved in accuracy. Besides, this method is robust to measurement noise and errors, and can cope with multiple conditions in real power transmission systems.

In robotics, decoding complex human movements of multiple degrees of freedom (DOFs) from surface electromyography (sEMG) remains challenging. Recently, the rapid development of artificial intelligence (AI) technology provides a new solution to this problem. In this paper, we propose an AI-based framework that consists of a series of deep learning approaches, for achieving a precise and robust decoding on three-dimensional (3D) wrist movements. A previously developed device (WMD) was utilized to tag the myoelectric signals with 3D kinematic labels. A public dataset (HIT-SimCo) was established, wherein the sEMG signals were collected from diverse wrist movements and multiple subjects. A lightweight convolutional neural network (CNN) was constructed, which can extract the motion-related features directly from raw sEMG, and make motion predictions in an end-to-end manner. In addition, several data augmentation strategies were explored to improve the robustness of the model against environmental variations; and a fine-tuning policy was proposed to further improve the subject-specific accuracy. The decoding model was finally tested in three scenarios: a robotic arm/hand system performing daily-living tasks (pouring, screwing, etc.), a supernumerary robotic hand playing blocks-building, and a virtual prosthesis conducting motor rehabilitation training.

Visible Light Positioning (VLP) has been considered as a promising indoor positioning technology due to its high precision and low cost. However, current visible light positioning techniques are greatly limited by the assumption that lights have to be turned on to emit shining light beams, which is not applicable to scenarios that do not need the illumination all the time. In this paper, we design and implement a novel visible light positioning system, DarkVLP, to achieve high-precision positioning when lights are “turned off”. The “turn off” state means lights emit the extremely-low luminance that is imperceptible to human eyes. In order to realize such a system, we have to tackle non-trivial challenges in data encoding/decoding, modulation/demodulation and positioning algorithm. Specifically, we propose the sliding rheostat based modulation scheme to eliminate the need of complex signal synchronization mechanism and landmark recognition algorithm at the receiver. We also propose the dual-photodiode based positioning algorithm, which effectively mitigates effects of substantial signal strength fluctuation, to reliably achieve visible light positioning. In the end, we design novel circuits and prototype our system DarkVLP on commercial off-the-shelf devices. Results of extensive experiments demonstrate that our DarkVLP system could achieve sub-meter precision under the extremely-low luminance, which greatly broadens application scenarios of visible light positioning.

In near coast navigation, buoys and beacons convey essential information about dangers. At night-time, selected buoys send out individual blink-sequences that are marked in sea charts. International regulations require that navigation officer on watch makes visual confirmation of objects and their type in order to navigate safely. With rapid developments of highly automated vessels, this duty needs be carried out by algorithms that detect and locate objects without human intervention. At night-time, this requires algorithms that decode blink sequences and are able to classify from this information. The paper investigates this problem and suggests an algorithm that solves the problem. Convolutional Neural Networks (CNN) with Gated Recurrent Units (GRU) are developed for classification. A dedicated architecture is suggested that includes both temporal and color decoding to obtain unique precision. We demonstrate how networks are trained on synthetically generated data, and the paper shows, on real-world data, how the suggested approach yields 100.0% accurate results on 44 real-world recordings while being robust to inaccuracy in actual blink sequences. Comparison with baseline signal processing and with a recent state-of-the-art 3D CNN model shows that the new blink-sequence classifier outperforms alternative algorithms. A showcase of the results of this work is available in this video: https://youtu.be/KEi8qNnKV2w.

IP traffic has increased significantly in recent years, and it is expected that this progress will continue. Recent studies report that the viewing of online video content accounts for a share of 1% of the global greenhouse gas emissions. To reduce the data traffic of video streaming, the new standard Versatile Video Coding (VVC) has been finalized in 2020. In this paper, the energy efficiency of two different VVC decoders is analyzed in detail. Furthermore, we propose a design space exploration that uses an algorithm based on a greedy strategy to derive coding tool profiles that optimize the energy demand of the decoder. We show that the algorithm derives optimal coding tool profiles for a subset of coding tools. Additionally, we propose profiles that reduce the energy demand of VVC decoders and provide energy savings of more than 50% for sequences with 4K resolution. Thereby, we will also show that the proposed profiles can have a lower decoding energy demand than comparable HEVC-encoded bit streams while also having a significantly lower bit rate., Accepted by IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)

The Internet of Vehicles (IoV) has always attracted attention as the emerging communication network with the most development potential in the 5G era. However, the performance of IoV under 5G ultra-dense networks is an open issue, especially in practice the outage probability and ergodic capacity of the relay cooperative IoV network under aggregate interference are still unclear. Therefore, an opportunistic Decoding and Forwarding (DF) relay cooperative transmission algorithm was proposed in this paper when the destination node of IoV has aggregated interference. In addition, based on mathematical theoretical knowledge such as numerical analysis, the closed expressions of the outage probability and ergodic capacity of the IoV system under aggregated interference was derived. Finally, simulation experiments verify the effectiveness of the proposed scheme and the correctness of the theoretical analysis, which improves the transmission rate of the system.

As an important technology in artificial intelligence, Granular Computing has emerged as a new multi-disciplinary paradigm and received much attention in recent years. Information granules forming an abstract and efficient characterization of large volumes of numeric data have been considered as the fundamental constructs of Granular Computing. By generating centroids (prototypes) and partition matrix, fuzzy clustering is a commonly encountered way of information granulation. As a reverse process of granulation, degranulation involves data reconstruction completed on a basis of the granular representatives (decoding information granules into numeric data). Previous studies have shown that there is a relationship between the reconstruction error and the performance of the granulation process. Typically, the lower the degranulation error is, the better performance of granulation process becomes. However, the existing methods of degranulation usually cannot restore the original numeric data, which is one of the important reasons behind the occurrence of the reconstruction error. To enhance the quality of reconstruction (degranulation), in this study, we develop an augmented scheme through modifying the partition matrix. By proposing the augmented scheme, we elaborate on a novel collection of granulation-degranulation mechanisms. In the constructed approach, the prototypes can be expressed as the product of the dataset matrix and the partition matrix. Then, in the degranulation process, the reconstructed numeric data can be decomposed into the product of the partition matrix and the matrix of prototypes. By modifying the partition matrix, the new partition matrix is constructed through a series of matrix operations. We offer a thorough analysis of the developed scheme. The experimental results are in agreement with the underlying conceptual framework. The results obtained on both synthetic and publicly available datasets are reported to show the enhancement of the data reconstruction performance thanks to the proposed method. It is pointed out that by using the proposed approach in some cases the reconstruction errors can be reduced close to zero by using the proposed approach.

In order to satisfy timeliness requirements arising from environmental sensing applications, age of information (AoI) was proposed to characterise the freshness of the received updates. In this paper, we consider a wireless powered relay aided communication network (WPRCN), in which a relay wirelessly powered by a hybrid-access point (H-AP) receives environmental monitoring information update from a sensor and forwards it to the H-AP. The long-term average AoI is studied since the decision correctness depends on the update timely uploaded by the relay. The wireless powered relay adopts either a decode-and-forward (DF) or an amplify and forward (AF) protocols, respectively, subject to the energy causality. We also consider the decoding cost of DF protocol owing to the relay’s limited energy storage. Since the expression of the average AoI is non-elementary, we propose a Taylor approximation based algorithm to obtain its integral. We optimise the transmit power/the equivalent average power consumption of the relay for the sake of minimising the average AoI in the whole WPRCN with different forwarding protocols. Our simulation results demonstrate the accuracy of the theoretical analysis, while the average AoI of the WPRCN is optimised by our power allocation scheme at the relay. The proposed algorithm is verified to achieve great approximation effect.

The fundamental communication paradigms in the next-generation mobile networks are shifting from connected things to connected intelligence. The potential result is that current communication-centric wireless systems are greatly stressed when supporting computation-centric intelligent services with distributed big data. This is one reason that makes federated learning come into being, it allows collaborative training over many edge devices while avoiding the transmission of raw data. To tackle the problem of model aggregation in federated learning systems, this paper resorts to multiple reconfigurable intelligent surfaces (RISs) to achieve efficient and reliable learning-oriented wireless connectivity. The seamless integration of communication and computation is actualized by over-the-air computation (AirComp), which can be deemed as one of uplink non-orthogonal multiple access (NOMA) techniques without individual information decoding. Since all local parameters are uploaded via noisy concurrent transmissions, the unfavorable propagation error inevitably deteriorates the accuracy of the aggregated global model. The goals of this work are to 1) alleviate the signal distortion of AirComp over shared wireless channels, and 2) speed up the convergence rate of federated learning. More specifically, both the mean-square-error (MSE) and the device set in the model uploading process are optimized by jointly designing transceivers, tuning reflection coefficients, and selecting clients. Compared to baselines, extensive simulation results show that 1) the proposed algorithms can aggregate model more accurately and accelerate convergence; 2) the training loss and inference accuracy of federated learning can be improved significantly with the aid of multiple RISs.

We propose a partial cooperative zero-forcing (PCZF) decoding scheme for the uplink cell-free massive MIMO system, wherein the neighboring access points (APs) around each user equipment (UE) share the channel state information (CSI) and jointly suppress the interference using the zero-forcing technique. Using asymptotic analysis, we derive a closed-form asymptotic expression for a lower-bound on the achievable rates. Considering the unique and complex form of the achievable rates, we propose power control schemes according to two criteria. The first criterion is to maximize the minimum achievable rate. For this criterion we propose a Target-SINR-Tracking (TST) based bisection algorithm. Since the power control update functions are standard interference functions, the TST based bisection method always converges to the optimal solution. The second criterion is to maximize the sum-rate, for which we propose two power control algorithms: randomization and scaling algorithm (RSA) and fractional programming algorithm (FPA). In each iteration of the RAS algorithm, we first exploit the randomization technique to transform the sum-rate maximization problem into a series of power minimization problems, and then improve the sum-rate by scaling. In the FP algorithm, we derive a lower-bound on the sum-rate, and then propose an iterative approach based on the Lagrangian dual transform and fractional programming to maximize the sum-rate lower-bound. Numerical results validate the theoretical analysis and verify the efficiency of the proposed power control algorithms.

The perception and expression of emotion, as well as the content of a text, are key factors in the success of conversational agents. However, previous models for conversation generation handled single-language pairs during training and testing, neglecting the complementary information from different languages. In this article, we propose a bilingual-aided interactive approach that can simultaneously and interactively generate bilingual emotional replies to monolingual posts. Specifically, the generation of one emotional reply relies on the output of the encoder, the generated tokens, and the interactive information from the other language decoder. The interactive approach includes 1) internal interaction to capture the change in implicit contextual information and 2) external interaction to balance grammaticality and the expression of emotion. Qualitative and quantitative experiments with NLPCC2017 show that our model performs better in terms of the content and emotion of replies than several state-of-the-art approaches.

For NAND flash memory, designing a good low-density parity-check (LDPC) decoding algorithm could ensure data reliability. When the decoding algorithm is implemented in hardware, it is necessary to achieve attractive trade off between implementation complexity and decoding performance. In this paper, a novel low bit-width decoding scheme is introduced. In this scheme, the Quasi-Cyclic LDPC (QC-LDPC) is used, and the row-layered normalized min-sum algorithm is improved by restricting the amplitude of minimum and second-minimum values in each check node (CN) updating. The simulation shows that our approach achieves a lower UBER (Uncorrectable Bit Error Rate) with a negligible increase in computational complexity, especially with low precision input log-likelihood ratio (LLR).

In this work, we propose a globally optimal joint successive interference cancellation (SIC) ordering and power allocation (JSPA) algorithm for the sum-rate maximization problem in downlink multi-cell non-orthogonal multiple access (NOMA) systems. The proposed algorithm is based on the exploration of base stations (BSs) power consumption, and closed-form of optimal powers obtained for each cell. Although the optimal JSPA algorithm scales well with larger number of users, it is still exponential in the number of cells. For any suboptimal decoding order, we propose a low-complexity near-optimal joint rate and power allocation (JRPA) strategy in which the complete rate region of users is exploited. Furthermore, we design a near-optimal semi-centralized JSPA framework for a two-tier heterogeneous network such that it scales well with larger number of small-BSs and users. Numerical results show that JRPA highly outperforms the case that the users are enforced to achieve their channel capacity by imposing the well-known SIC necessary condition on power allocation. Moreover, the proposed semi-centralized JSPA framework significantly outperforms the fully distributed framework, where all the BSs operate in their maximum power budget. Therefore, the centralized JRPA and semi-centralized JSPA algorithms with near-optimal performances are good choices for larger number of cells and users., Comment: 55 pages, 33 figures

LoRa, a representative Low-Power Wide Area Network (LPWAN) technology, has been shown as a promising platform to connect Internet of Things. Practical LoRa deployments, however, suffer from collisions, especially in dense networks and wide coverage areas expected by LoRa applications. Existing collision resolution approaches do not exploit the coding properties of LoRa and thus cannot work well for low SNR LoRa signals. We propose NScale to decompose concurrent transmissions by leveraging subtle inter-packet time offsets for low SNR LoRa collisions. NScale (1) translates subtle time offsets, which are vulnerable to noise, to robust frequency features, and (2) further amplifies the time offsets by non-stationary signal scaling, i.e., scaling the amplitude of a symbol differently at different positions. In practical implementation, we propose a noise resistant iterative symbol recovery method to combat symbol distortion in low SNR, and address frequency shifts incurred by CFO and packet time offsets in decoding. We theoretically show that NScale introduces

Ride-hailing service has witnessed a dramatic growth over the past decade but meanwhile raised various challenging issues, one of which is how to provide a timely and accurate short-term prediction of supply and demand. While the predictions for zone-based demand have been extensively studied, much less efforts have been paid to the predictions for origin-destination (OD) based demand (namely, demand originating from one zone to another). However, OD-based demand prediction is even more important and worth further explorations, since it provides more elaborate trip information in the near future as reference for fine-grained operations, such as the routing and matching of shared ride-hailing services that pick up and drop off two or more passengers in each ride. Simultaneous prediction of both zone-based and OD-based demand can be an interesting and practical problem for the ride-hailing platforms. To address the issue, we propose a multi-task matrix factorized graph neural network (MT-MF-GCN), which consists of two major components: (1) a GCN (graph convolutional network) basic module that captures the spatial correlations among zones via mixture-model graph convolutional (MGC) network, and (2) a matrix factorization module for multi-task predictions of zone-based and OD-based demand. By evaluations on the real-world on-demand data in Manhattan and Haikou, we show that the proposed model outperforms the state-of-the-art baseline methods in both zone- and OD-based predictions.

This paper provides an end-to-end latency model of real-time generated live videos composed of sub-models of capturing, encoding, network, decoding, rendering, and display refreshing. We build the model by deeply analyzing each sub part's latency with parameters like video's Spatial (frame scale s), Temporal (frame rate t), Amplitude (quantization step-size q) Resolutions (STAR), Compressed Frame Bit Numbers (nF), and other necessary parameters. The latency of 640 combinations of STAR for eight representative video sequences is measured to fit accurate models. We primarily focus on processing procedures, whose latency is becoming the principal affection, as network latency decreases due to the its development. We then introduce a baseline video sequence set to reduce the complexity when fitting the encoding and decoding sub-models by predicting the model parameters on new hardware from parameters of a known one, and improve models' robustness crossing different scenarios. These sub-models correlate well with the realistic latency data, with the average Pearson Correlation Coefficient (PCC) of 0.99 and the average Spearman's Rank Correlation Coefficient (SRCC) of 0.98, according to an independent validation test. Furthermore, we validate our end-to-end latency model in a simulated end-to-end live video transmission from generating to displaying, introducing the latency as a strict condition in adaptive bit-rate selection. The simulation shows that our model could significantly increase time during which the latency can satisfy the given tolerance at a low cost.

Recent works on learned image compression perform encoding and decoding processes in a full-resolution manner, resulting in two problems when deployed for practical applications. First, parallel acceleration of the autoregressive entropy model cannot be achieved due to serial decoding. Second, full-resolution inference often causes the out-of-memory(OOM) problem with limited GPU resources, especially for high-resolution images. Block partition is a good design choice to handle the above issues, but it brings about new challenges in reducing the redundancy between blocks and eliminating block effects. To tackle the above challenges, this paper provides a learned block-based hybrid image compression (LBHIC) framework. Specifically, we introduce explicit intra prediction into a learned image compression framework to utilize the relation among adjacent blocks. Superior to context modeling by linear weighting of neighbor pixels in traditional codecs, we propose a contextual prediction module (CPM) to better capture long-range correlations by utilizing the strip pooling to extract the most relevant information in neighboring latent space, thus achieving effective information prediction. Moreover, to alleviate blocking artifacts, we further propose a boundary-aware postprocessing module (BPM) with the edge importance taken into account. Extensive experiments demonstrate that the proposed LBHIC codec outperforms the VVC, with a bit-rate conservation of 4.1%, and reduces the decoding time by approximately 86.7% compared with that of state-of-the-art learned image compression methods., 13 pages, 13 figures, accepted by IEEE Trans. on Circuits and Systems for Video Technology

Decoding the content in neural activity through voxel-wise encoding plays an important role in investigating cognitive functions of the human brain. However, unlike multi-voxel pattern analysis (MVPA), voxel-wise encoding builds a model for each individual voxel, therefore ignores the interactions between voxels and is sensitive to noise. In this work, we propose the Feature-specific denoise (FSdenoise), a noise reduction method for encoding-based models to improve their decoding performance. FSdenoise considers the response of a voxel to a stimulus as a combination of two components: feature-relevant component, which can be predicted from stimulus features and featureirrelevant component, which shows no direct relation to the concerned features. Exploiting the correlations between voxels, FSdenoise reduces the feature-irrelevant component in voxels that exhibit more feature-relevant component, enhancing their predictive power from stimulus features. Decoding performance with the denoised voxels would be improved in consequence. We validate the FSdenoise on two fMRI datasets and the results demonstrate that FSdenoise can efficiently improve the decoding accuracy for encoding-based approaches. Moreover, the encoding-based approaches combined with FSdenoise can even outperform the MVPA-based approach in brain decoding.

Application of deep learning to error control coding is gaining special attention and neural network architectures on decoding are approached to compare with conventional ones. Turbo codes conventionally use BCJR algorithm for decoding. In this paper, performances of neural Turbo decoder and deep learning-based Turbo decoder are examined. A category of sequential codes are utilized to construct the RSC (Recursive Systematic Convolutional) codes as basic elements for Turbo encoder. Sequential codes suit the requirement of memory element present in convolution codes, which act as components for Turbo encoder. Turbo decoders are constructed by two means; as neural Turbo decoder and deep learning Turbo decoder. Both structures are based on recurrent neural network (RNN) architectures. RNN architectures are preferred due to the presence of memory as a feature. BER performance of both is compared with that of a convolutional Viterbi decoder in awgn channel. Both the structures are studied for different input data-lengths and code rates.

BACKGROUND: The analysis of brain activity in different conditions is an important research area in neuroscience. OBJECTIVE: This paper analyzed the correlation between the brain and skin activities in rest and stimulations by information-based analysis of electroencephalogram (EEG) and galvanic skin resistance (GSR) signals. METHODS: We recorded EEG and GSR signals of eleven subjects during rest and auditory stimulations using three pieces of music that were differentiated based on their complexity. Then, we calculated the Shannon entropy of these signals to quantify their information contents. RESULTS: The results showed that music with greater complexity has a more significant effect on altering the information contents of EEG and GSR signals. We also found a strong correlation (r= 0.9682) among the variations of the information contents of EEG and GSR signals. Therefore, the activities of the skin and brain are correlated in different conditions. CONCLUSION: This analysis technique can be utilized to evaluate the correlation among the activities of various organs versus brain activity in different conditions.

Human has excellent motor capability and performance in completing various manipulation tasks. During some tasks such as tightening/loosening a screw with a screwdriver, the motion is accompanied by force exertion to the environment (that is, constrained motion). To obtain natural human-robot interaction (HRI) as human interacts/collaborates with the environment, interpreting the human’s intention in a way of motion and interaction force is meaningful for carrying out constrained tasks. This paper proposes a long-short term memory (LSTM) network-based decoding method for the simultaneous estimation of human motion and interactive force from surface electromyography (sEMG) signals. The surface EMG recorded from the muscles of forearm is used to decode human’s motion intention. In order to extract smooth features from non-stationary sEMG signals, Bayesian filter is applied instead of traditional time-domain feature extraction method. From the real-time experiments on eight subjects, the LSTM-based decoding method represents high accuracy of motion estimates (91.7%) and force estimates (96.1%) despite of the existence of muscle coupling and non-stationary mapping during such constrained tasks. It indicates that the estimated motion and interactive force can be further applied for HRI in accomplishing such constrained tasks.

RGB salient object detection (SOD) has made great progress. However, the performance of this single-modal salient object detection will be significantly decreased when encountering some challenging scenes, such as low light or darkness. To deal with the above challenges, thermal infrared (T) image is introduced into the salient object detection. This fused modal is called RGB-T salient object detection. To achieve deep mining of the unique characteristics of single modal and the full integration of cross-modality information, a novel Cross-Guided Fusion Network (CGFNet) for RGB-T salient object detection is proposed. Specifically, a Cross-Scale Alternate Guiding Fusion (CSAGF) module is proposed to mine the high-level semantic information and provide global context support. Subsequently, we design a Guidance Fusion Module (GFM) to achieve sufficient cross-modality fusion by using single modal as the main guidance and the other modal as auxiliary. Finally, the Cross-Guided Fusion Module (CGFM) is presented and serves as the main decoding block. And each decoding block is consists of two parts with two modalities information of each being the main guidance, i.e., cross-shared Cross-Level Enhancement (CLE) and Global Auxiliary Enhancement (GAE). The main difference between the two parts is that the GFM using different modalities as the main guide. The comprehensive experimental results prove that our method achieves better performance than the state-of-the-art salient detection methods. The source code has released at: https://github.com/wangjie 0825/CGFNet.git.

Neural Network Coding and Representation (NNR) is the first international standard for efficient compression of neural networks (NNs). The standard is designed as a toolbox of compression methods, which can be used to create coding pipelines. It can be either used as an independent coding framework (with its own bitstream format) or together with external neural network formats and frameworks. For providing the highest degree of flexibility, the network compression methods operate per parameter tensor in order to always ensure proper decoding, even if no structure information is provided. The NNR standard contains compression-efficient quantization and deep context-adaptive binary arithmetic coding (DeepCABAC) as core encoding and decoding technologies, as well as neural network parameter pre-processing methods like sparsification, pruning, low-rank decomposition, unification, local scaling and batch norm folding. NNR achieves a compression efficiency of more than 97% for transparent coding cases, i.e. without degrading classification quality, such as top-1 or top-5 accuracies. This paper provides an overview of the technical features and characteristics of NNR.

For diagnosing cardiovascular disease, an accurate segmentation method is needed. There are several unresolved issues in the complex field of cardiac magnetic resonance imaging, some of which have been partially addressed by using deep neural networks. To solve two problems of over-segmentation and under-segmentation of anatomical shapes in the short-axis view from different cardiac magnetic resonance sequences, we propose a novel two-stage framework with a weighted decision map based on convolutional neural networks to segment the myocardium (Myo), left ventricle (LV), and right ventricle (RV) simultaneously. The framework comprises a decision map extractor and a cardiac segmenter. A cascaded U-Net++ is used as a decision map extractor to acquire the decision map that decides the category of each pixel. Cardiac segmenter is a multiscale dual-path feature aggregation network (MDFA-Net) which consists of a densely connected network and an asymmetric encoding and decoding network. The input to the cardiac segmenter is derived from processed original images weighted by the output of the decision map extractor. We conducted experiments on two datasets of multi-sequence cardiac magnetic resonance segmentation challenge 2019 (MS-CMRSeg 2019) and myocardial pathology segmentation challenge 2020 (MyoPS 2020). Test results obtained on MyoPS 2020 show that the average Dice coefficients of the proposed method on the segmentation tasks of Myo, LV and RV are 84.70%, 86.00%, and 86.31%, respectively.

The Internet of Things (IoT) is scaling rapidly to billions of low power devices, with diverse radio technologies sharing a common unlicensed spectrum. Inevitably, this results in rampant cross-technology collisions between the devices that lead to wasteful re-transmissions, draining the battery life of low-power devices significantly. We present CharIoT, the first cross-technology distributed MIMO receiver system that exploits the potential of distributed MIMO to facilitate better co-existence and decoding of a large number of simultaneous low power uplink transmissions from unmodified low-power clients. CharIoT is a recovery-based system that intelligently collects radio samples from teams of light-weight IoT gateways and streams them to the cloud to effectively resolve collisions. At the cloud, CharIoT develops a suite of technology-specific software filters that decouple collisions across diverse technologies, facilitating seamless co-existence across low power radios. An implementation of CharIoT on inexpensive RTL-SDR gateways connected to a Raspberry Pi decodes collisions of four popular IoT technologies in the 868MHz ISM bands - LoRa, XBee, Z-Wave and SIGFOX showing gains in throughput of up to 4x and battery life of up to 3.5 years.

We observe that data access and processing takes a significant amount of time in large-scale deep learning training tasks (DLTs) on image datasets. Three factors contribute to this problem: (1) the massive and recurrent accesses to large numbers of small files; (2) the repeated, expensive decoding computation on each image, and (3) the frequent communication between computation nodes and storage nodes. Existing work has addressed some aspects of these problems; however, no end-to-end solutions have been proposed. In this article, we propose DIESEL+, an all-in-one system which accelerates the entire I/O pipeline of deep learning training tasks. DIESEL+ contains several components: (1) local metadata snapshot; (2) per-task distributed caching; (3) chunk-wise shuffling; (4) GPU-assisted image decoding and (5) online region-of-interest (ROI) decoding. The metadata snapshot removes the bottleneck on metadata access in frequent reading of large numbers of files. The per-task distributed cache across the worker nodes of a DLT task to reduce the I/O pressure on the underlying storage. The chunk-based shuffle method converts small file reads into large chunk reads, so that the performance is improved without sacrificing the training accuracy. The GPU-assisted image decoding and the online ROI method minimize the image decoding workloads and reduce the cost of data movement between nodes. These techniques are seamlessly integrated into the system. In our experiments, DIESEL+ outperforms existing systems by a factor of two to three times on the overall training time.

Recent neural architectures in sequence labeling have yielded state-of-the-art performance on single domain data such as newswires. However, they still suffer from (i) requiring massive amounts of training data to avoid overfitting; (ii) huge performance degradation when there is a domain shift in the data distribution between training and testing. To make a sequence labeling system more broadly useful, it is crucial to reduce its training data requirements and transfer knowledge to other domains. In this paper, we investigate the problem of domain adaptation for sequence labeling under homogeneous and heterogeneous settings. We propose MetaSeq, a novel meta-learning approach for domain adaptation in sequence labeling. Specifically, MetaSeq incorporates meta-learning and adversarial training strategies to encourage robust, general and transferable representations for sequence labeling. The key advantage of MetaSeq is that it is capable of adapting to new unseen domains with a small amount of annotated data from those domains. We extensively evaluate MetaSeq on named entity recognition, part-of-speech tagging and slot filling under homogeneous and heterogeneous settings. The experimental results show that MetaSeq achieves state-of-the-art performance against eight baselines. Impressively, MetaSeq surpasses the in-domain performance using only 16.17% and 7% of target domain data on average for homogeneous settings, and 34.76%, 24%, 22.5% of target domain data on average for heterogeneous settings.

We present a new low-complexity interference management algorithm that exploits reconfigurable antennas, removes most of the interference signal power in a K-user interference channel, and achieves the promised but unrealized gain of interference alignment in low-to-medium signal-to-noise ratio (SNR) range. The reconfigurable antenna includes several elements, each controlled individually using a radio frequency (RF) switch. The antenna has only one RF chain, resulting in simple hardware design and enabling low-complexity algorithms. We exploit space-time diversity to find two separate combinations of activating or deactivating antenna elements to allow for interference neutralization/elimination. Finding such states is a cumbersome task for which we develop novel, efficient algorithms. We further devise simple encoding and decoding with short precoder length to exploit the benefits of our designs. Our implementation does not rely on channel state information at the transmitters; works at finite signal-to-noise ratios, unlike the typical degrees-of-freedom results; and works in slow-fading environments. We calculate the average achievable rates, the minimum number of elements to reach a specific level of interference suppression, and we provide outage analysis of our technique.