Deep Knowledge Tracing (DKT), powered by recurrent neural networks (RNNs) and their advanced variants, effectively captures complex learner progression patterns. By uncovering latent structures in learning behaviors, DKT enables precise predictions and facilitates personalized interventions, enhancing the effectiveness of intelligent educational platforms. However, real-world applications often face challenges due to limited annotated learning interactions for model training, constrained by factors such as budget restrictions and privacy concerns. To address the challenges of low-resource DKT, this study integrates semi-supervised learning and ensemble deep learning. The approach begins by training multiple DKT models, including Knowledge Proficiency Tracing (KPT), Exercise-Correlated Knowledge Proficiency Tracing (EKPT), and Dynamic Key-Value Memory Networks (DKVMN), using a combination of limited labeled learning interactions and a large volume of unlabeled learning interactions. These models are then ensembled through majority voting, effectively leveraging the strengths of both semi-supervised learning and ensemble strategies. Experimental results on benchmark datasets, such as ASSISTments, demonstrate that the proposed model enhance the performance for the low-resource DKT, significantly improving key metrics including AUC, accuracy, and precision.

Prediction reliability of deep learning based systems allows users to confirm if the prediction is reliable to real applications, which is the key to the success of recommendation systems. Current research on deep learning based recommendation systems focused on estimating model reliability, but seemed to be missing data contributions to the prediction reliability. This paper proposed a novel framework to estimate the prediction reliability for deep learning-based methods through combining the model reliability and the data reliability. The proposed framework has been validated in a case study based on named entity recognition (NER) that is from an Intuit recommendation task. It employed two NER datasets: WNUT and GMB to examine detailed performance of the proposed framework, where, specifically, we proposed a novel evaluation metric to comprehensively evaluate the performance. Experimental results demonstrated that, compared to model reliability only, combining data reliability with model reliability will significantly improve performance, as well as enhance the prediction interpretability.

In this paper, the real-time optimal transmission power control problem is investigated for large-scale mobile wireless sensor networks (MWSN). Controlling large-scale MWSN has two novel challenges, i.e., 1) increasing navigation complexity due to a large number of mobile wireless sensors, and 2) limited energy prohibits peer-to-peer communication between large-scale mobile sensors. To overcome these challenges, the novel mean field game theory is adopted and integrated along with the emerging decentralized reinforcement learning technique. Specifically, the optimal transmission control problem and the optimal navigation problem are formulated as mean field games with two objectives. Then, a novel Actor-Critic-Mass multi-agent reinforcement learning algorithm is developed to learn the decentralized optimal transmission power control and motion control. To learn the decentralized optimal navigation and transmission power control policies, the coupled Hamiltonian-Jacobian-Bellman (HJB) and Fokker-Planck-Kolmogorov (FPK) equations are derived in mean field game formulation. The learned decentralized policies can be guaranteed to converge close to the optimal value, i.e., the Nash Equilibrium, even with large-scale MWSNs in uncertain environments. Finally, the numerical simulations have been provided to demonstrate the effectiveness of the proposed design.

Resistive crossbars using non-volatile memory devices have become promising components for implementing Deep Neural Network (DNN) in hardware. However, crossbar-based computations encounter a notable challenge due to device and circuit-level non-idealities. In this study, we explore three recent crossbar simulation tools, namely, AIHWKIT, Cross-Sim, and MemTorch, and evaluate four pre-trained DNNs for image classification on CIFAR-100 dataset using these tools. All three tools have strong analog simulation functionalities that effectively mimic actual hardware environment. To the best of our knowledge, this is the first study of using all three simulation tools of analog accelerators, representing three distinctive hardware settings, to evaluate DNN robustness under analog noise and nonlinearities. We first test the robustness of four DNNs (VGGI9, InceptionV3, ResNet50, & MobilenetV2) using different levels of white Gaussian noise as baselines. Then we evaluate their inference performance on the three tools to determine their resilience to analog noise and nonlinearities in hardware environment. Results show that while all DNNs suffer performance degradation as expected, ResNet50 outperforms others in two out of three simulators despite real-world hardware imperfections due to its deep structure. Specifically, the ResNet50 model shows a mere 3 % performance drop in CrossSim and, interestingly, a 2% improvement in AIHWKIT. At the same time, InceptionV3 exhibits only a 3 % drop from its baseline, while the performance of three other models declines by 12-19% in MemTorch underscoring InceptionV3's resilience in MemTorch environment. This indicates that model design of DNNs plays an important role in their resilience to analog noise and nonlinearities, and their inference performance also depends on specific hardware implementation.

In this paper, the optimal placement and dynamic resource allocation problem has been investigated for multi-UAV enhanced reconfigurable intelligent surface (RIS) assisted wireless network with uncertain time-varying wireless channels. This paper aims to stimulate the potential of RIS by adding mobility to RIS through unmanned aerial vehicles (UAV). A novel UAV optimal placement and dynamic resource allocation technique needs to be developed jointly. A novel online rein-forcement learning based optimal resource allocation algorithm has been designed. Firstly, a deep Q-learning based K-means clustering algorithm is utilized to optimize the deployment of the multi-UAV. Then, an online actor-critic reinforcement learning algorithm is developed to learn the optimal transmit power control as well as mobile RIS phase shift control policy. Compared with conventional learning algorithms, the developed algorithm can learn the optimal resource allocation and multi-UAV placement for mobile RIS-assisted wireless networks in real-time even with uncertain and time-varying wireless channels. Eventually, numerical simulations are provided to demonstrate the effectiveness of developed schemes.

Cell type identification from single-cell transcriptomic data is a common goal of single-cell RNA sequencing (scRNAseq) data analysis. Deep neural networks have been employed to identify cell types from scRNAseq data with high performance. However, it requires a large mount of individual cells with accurate and unbiased annotated types to train the identification models. Unfortunately, labeling the scRNAseq data is cumbersome and time-consuming as it involves manual inspection of marker genes. To overcome this challenge, we propose a semi-supervised learning model "SemiRNet" to use unlabeled scRNAseq cells and a limited amount of labeled scRNAseq cells to implement cell identification. The proposed model is based on recurrent convolutional neural networks (RCNN), which includes a shared network, a supervised network and an unsupervised network. The proposed model is evaluated on two large scale single-cell transcriptomic datasets. It is observed that the proposed model is able to achieve encouraging performance by learning on the very limited amount of labeled scRNAseq cells together with a large number of unlabeled scRNAseq cells.

Analog hardware has become a popular choice for machine learning on resource-constrained devices recently due to its fast execution and energy efficiency. However, the inherent presence of noise in analog hardware and the negative impact of the noise on deployed deep neural network (DNN) models limit their usage. The degradation in performance due to the noise calls for the novel design of DNN models that have excellent noise-resistant property, leveraging the properties of the fundamental building block of DNN models. In this work, the use of L1or TopK BatchNorm type, a fundamental DNN model building block, in designing DNN models with excellent noise-resistant property is proposed. Specifically, a systematic study has been carried out by training DNN models with L1/TopK BatchNorm type, and the performance is compared with DNN models with L2BatchNorm types. The resulting model noise-resistant property is tested by injecting additive noise to the model weights and evaluating the new model inference accuracy due to the noise. The results show that L1and TopK BatchNorm type has excellent noise-resistant property, and there is no sacrifice in performance due to the change in the BatchNorm type from L2to L1/TopK BatchNorm type.

In this article, we propose a transfer learning (TL) enabled edge convolutional neural network (CNN) framework for 5G industrial edge networks with privacy-preserving characteristic. In particular, the edge server can use the existing image dataset to train the CNN in advance, which is further fine-tuned based on the limited datasets uploaded from the devices. With the aid of TL, the devices that are not participating in the training only need to fine-tune the trained edge-CNN model without training from scratch. Due to the energy budget of the devices and the limited communication bandwidth, a joint energy and latency problem is formulated, which is solved by decomposing the original problem into an uploading decision subproblem and a wireless bandwidth allocation subproblem. Experiments using ImageNet demonstrate that the proposed TL-enabled edge-CNN framework can achieve almost 85% prediction accuracy of the baseline by uploading only about 1% model parameters, for a compression ratio of 32 of the autoencoder.

Increasing concerns on intelligent spectrum sensing call for efficient training and inference technologies. In this paper, we propose a novel federated learning (FL) framework, dubbed federated spectrum learning (FSL), which exploits the benefits of reconfigurable intelligent surfaces (RISs) and overcomes the unfavorable impact of deep fading channels. Distinguishingly, we endow conventional RISs with spectrum learning capabilities by leveraging a fully-trained convolutional neural network (CNN) model at each RIS controller, thereby helping the base station to cooperatively infer the users who request to participate in FL at the beginning of each training iteration. To fully exploit the potential of FL and RISs, we address three technical challenges: RISs phase shifts configuration, user-RIS association, and wireless bandwidth allocation. The resulting joint learning, wireless resource allocation, and user-RIS association design is formulated as an optimization problem whose objective is to maximize the system utility while considering the impact of FL prediction accuracy. In this context, the accuracy of FL prediction interplays with the performance of resource optimization. In particular, if the accuracy of the trained CNN model deteriorates, the performance of resource allocation worsens. The proposed FSL framework is tested by using real radio frequency (RF) traces and numerical results demonstrate its advantages in terms of spectrum prediction accuracy and system utility: a better CNN prediction accuracy and FL system utility can be achieved with a larger number of RISs and reflecting elements.

Analog hardware implemented deep learning models are promising for computation and energy constrained systems such as edge computing devices. However, the analog nature of the device and the many associated noise sources will cause changes to the value of the weights in the trained deep learning models deployed on such devices. In this study, systematic evaluation of the inference performance of trained popular deep learning models for image classification deployed on analog devices has been carried out, where additive white Gaussian noise has been added to the weights of the trained models during inference. It is observed that deeper models and models with more redundancy in design, such as VGG, are more robust to the noise in general. Also, it is observed that the performance is affected by the design philosophy of the model, the detailed structure of the model, the exact machine learning task, as well as the datasets.

The empowering unmanned aerial vehicles (UAVs) have been extensively used in providing intelligence such as target tracking. In our field experiments, a pretrained convolutional neural network (CNN) is deployed at UAV to identify a target (a vehicle) from the captured video frames and enable the UAV to keep tracking. However, this kind of visual target tracking demands a lot of computational resources due to the desired high inference accuracy and stringent delay requirement. This motivates us to consider offloading this type of deep learning (DL) tasks to a mobile-edge computing (MEC) server due to the limited computational resource and energy budget of the UAV and further improve the inference accuracy. Specifically, we propose a novel hierarchical DL tasks distribution framework, where the UAV is embedded with lower layers of the pretrained CNN model while the MEC server (MES) with rich computing resources will handle the higher layers of the CNN model. An optimization problem is formulated to minimize the weighted-sum cost, including the tracking delay and energy consumption introduced by communication and computing of UAVs while taking into account the quality of data (e.g., video frames) input to the DL model and the inference errors. Analytical results are obtained and insights are provided to understand the tradeoff between the weighted-sum cost and inference error rate in the proposed framework. Numerical results demonstrate the effectiveness of the proposed offloading framework.

Multi-access edge computing (MEC) has already shown great potential in enabling mobile devices to bear the computation-intensive applications by offloading some computing jobs to a nearby access point (AP) integrated with a MEC server (MES). However, due to the varying network conditions and limited computational resources of the MES, the offloading decisions taken by a mobile device and the computational resources allocated by the MES can be formulated as a mixed-integer nonlinear programming (MINLP) problem, which may not be optimized with the lowest cost. In this paper, we propose a novel offloading framework for the multi-server MEC network where each AP is equipped with an MES assisting mobile users (MUs) in executing computation-intensive jobs via offloading. Specifically, we formulate the offloading decision problem as a multiclass classification problem and formulate the MES computational resource allocation problem as a regression problem. Then a multi-task learning based feedforward neural network (MTFNN) model is designed and trained to jointly optimize the offloading decision and computational resource allocation. Numerical results show that the proposed MTFNN outperforms the conventional optimization method in terms of inference accuracy and computational complexity.

Single-cell RNA sequencing (scRNAseq) enables the profiling of the transcriptomes of individual cells, thus characterizing the heterogeneity of biological samples since scRNAseq experiments are able to yield high volumes of data. Analyzing scRNAseq data will be beneficial for obtaining knowledge on cancer drug resistance, gene regulation in embryonic development, and mechanisms of stem cell differentiation and reprogramming. One common goal of scRNAseq data analytics is to identify the cell type of each individual cell that has been profiled. However, data sparsity is the main challenge due to limitations of current single-cell RNA sequencing techniques. In this paper, a novel method of representing the genes as gene embeddings is proposed to reduce data sparsity of scRNAseq data for cell type identification, which is inspired by similarities between gene system and natural language system. It contains two steps: 1) transform gene sequences into gene sentences by ranking genes in terms of their expression values; 2) employ the word2vec technique to learn gene embeddings on these gene sentences. Then we build three deep learning models, namely RNNs, Attention RNNs, and Bi-directional LSTM RNNs, for cell type classification. The proposed method is evaluated on macosko2015, a large scale scRNAseq dataset with ground truth of individual cell types. Experimental results show that the proposed method performs effectively and efficiently on identifying cell types on scRNAseq data, and it can achieve promising performance even learning on limited number of genes.

Ambient backscatter communication enabling device-to-device (D2D) communications via the ambient radio frequency (RF) signal has revealed its numerous application potential in the Internet-of-Things (IoT) networks. However, the work on the link layer for the backscatter communication is in its infancy due to the constraints of ultralow power and cost in such a system, especially a channel access protocol in the backscatter communication system for IoT networks is rarely mentioned. In this article, a distributed multiple access control (MAC) protocol is presented, which allows multiple backscatter devices (BDs) to connect with each other by relying on the ambient RF signal. By combining an analog channel sensing strategy with the dual-backoff mechanism, each BD can switch among the transmission, receiving, and energy harvesting (EH) states. Specifically, each BD starts a randomly designated time for EH once the channel is sensed to be busy. As the timer expires, the BD ceases the EH and continues its sensing and backoff procedures. With the consideration of the false alarm and the miss detection problems occurring while sensing, an enhanced 3-D Markov model is built to analyze the saturation throughput performance of the proposed MAC protocol. Extensive simulations verify the analysis and demonstrate the advantage of the proposed backscatter MAC protocol.

In this article, we propose a novel framework of mobile edge computing (MEC)-based hierarchical machine learning (ML) tasks distribution for the Industrial Internet of Things. It is assumed that a batch of ML tasks, such as anomaly detection, need to be executed timely in an MEC setting, where the devices have limited computing capability while the MEC server (MES) has rich computing resources. Thus, a small ML model for the device and a deep ML model for the MES are pretrained offline using historical data, and then they are deployed accordingly. However, offloading tasks to the MES introduces communications delay. Thus, each device must decide the portion of the tasks to offload to minimize the processing delay. Since the delay and the error of data processing are incurred by communications and ML computing, a joint optimization problem is formulated to minimize the total delay subject to the ML model complexity and inference error rate, data quality, computing capability at the device and MES, and communications bandwidth. A closed-form solution is derived analytically and an optimal offloading strategy selection algorithm is proposed. Insights are provided to understand the tradeoff between communications and ML computing in offloading decisions, and the effects of key parameters in the proposed algorithm are investigated. The numerical results demonstrate the effectiveness of the proposed algorithm.

News in social media, such as Twitter, has been generated in high volume and speed. However, very few of them are labeled (as fake or true news) by professionals in near real time. In order to achieve timely detection of fake news in social media, a novel framework of two-path deep semisupervised learning (SSL) is proposed where one path is for supervised learning and the other is for unsupervised learning. The supervised learning path learns on the limited amount of labeled data, while the unsupervised learning path is able to learn on a huge amount of unlabeled data. Furthermore, these two paths implemented with convolutional neural networks (CNNs) are jointly optimized to complete SSL. In addition, we build a shared CNN to extract the low-level features on both labeled data and unlabeled data to feed them into these two paths. To verify this framework, we implement a Word CNN-based SSL model and test it on two data sets: LIAR and PHEME. Experimental results demonstrate that the model built on the proposed framework can recognize fake news effectively with very few labeled data.

Long-term evolution (LTE) deployment in the unlicensed band has been a promising solution to handle the ever-increasing data traffic growth. However, spectrum sharing on unlicensed band poses a significant challenge regarding the interaction between LTE licensed assisted access (LAA) and Wi-Fi. In this paper, a radio access technology (RAT) heterogeneous network that consists of an LAA tier and a Wi-Fi tier is constructed, to achieve the coexistence and alleviate the intra-RAT and inter-RAT interference, a listen-and-talk (LAT) scheme is utilized in Wi-Fi while LAA adopts the listen-before-talk (LBT). By leveraging the full-duplex (FD) techniques in Wi-Fi, the collision can be avoided and the utilization of unlicensed spectrum can be improved. Furthermore, the sensing errors are derived based on the FD strategy and an enhanced Markov model is presented to analyze the performance of the heterogeneous network with consideration of the residual self-interference (RSI). The fairness between LAA and Wi-Fi is also investigated. At last, to ensure the performance of Wi-Fi when serious RSI exists, a switched MAC (S-MAC) that can adaptively switch between the FD mode and half-duplex (HD) mode is presented.

In this paper, we address the issue of how to enhance the generalization performance of convolutional neural networks (CNN) in the early learning stage for image classification. This is motivated by real-time applications that require the generalization performance of CNN to be satisfactory within limited training time. In order to achieve this, a novel hierarchical transfer CNN framework is proposed. It consists of a group of shallow CNNs and a cloud CNN, where the shallow CNNs are trained firstly and then the first layers of the trained shallow CNNs are used to initialize the first layer of the cloud CNN. This method will boost the generalization performance of the cloud CNN significantly, especially during the early stage of training. Experiments using CIFAR-10 and ImageNet datasets are performed to examine the proposed method. Results demonstrate the improvement of testing accuracy is 12% on average and as much as 20% for the CIFAR-10 case while 5% testing accuracy improvement for the ImageNet case during the early stage of learning. It is also shown that universal improvements of testing accuracy are obtained across different settings of dropout and number of shallow CNNs.

In this paper, we propose a distributed deep reinforcement learning (DRL) based medium access control (MAC) protocol, termed DRL-MAC, which is used to assist the backscatter communications for Internet-of-Things (IoT) networks. By leveraging the current Wi-Fi infrastructure, the backscatter communications can be reserved in advance to avoid the interference from Wi-Fi communications. In the proposed MAC protocol, the deep reinforcement learning is further introduced to learn the reserved information and make decisions such as 1) which backscatter device (TAG) will be serviced, and 2) the reservation step for the serviced TAG. In addition, the utility function is defined and the optimization problem is formulated to balance the backscatter communications and Wi-Fi communications. Moreover, a DRL algorithm is proposed to obtain the optimal strategy. The numerical results show the effectiveness of the proposed MAC for backscatter communications.

Multi-access edge computing (MEC) has already shown the potential in enabling mobile devices to bear the computation-intensive applications by offloading some tasks to a nearby access point (AP) integrated with a MEC server (MES). However, due to the varying network conditions and limited computation resources of the MES, the offloading decisions taken by a mobile device and the computational resources allocated by the MES may not be efficiently achieved with the lowest cost. In this paper, we propose a dynamic offloading framework for the MEC network, in which the uplink non-orthogonal multiple access (NOMA) is used to enable multiple devices to upload their tasks via the same frequency band. We formulate the offloading decision problem as a multiclass classification problem and formulate the MES computational resource allocation problem as a regression problem. Then a multi-task learning based feedforward neural network (MTFNN) model is designed to jointly optimize the offloading decision and computational resource allocation. Numerical results illustrate that the proposed MTFNN outperforms the conventional optimization method in terms of inference accuracy and computation complexity.

In this paper, a joint latency and energy minimization problem is considered for hierarchical machine learning tasks distribution (HMLTD) with mobile edge computing (MEC). Firstly, we propose a MEC based HMLTD framework enabling mobile devices embedded with shallow neural network (SNN) model to offload latency-sensitive computing-intensive tasks to a nearby MEC server (MES), which has a more powerful deep neural network (DNN) model. Then, we formulate the offloading strategy as a piecewise convex optimization problem to minimize the weighted-sum of latency and energy. A closed-form solution of the optimal tasks partition strategy is derived analytically for different scenarios, and then an optimal partial offloading strategy (OPOS) is proposed. As proof of concept, some insights are gained to demonstrate the key parameters affecting the task partition strategy. Numerical results are given to illustrate that the proposed offloading scheme outperforms the baseline scheme.

Nowadays, an Internet-of-Things (IoT) connected system brings a tremendous paradigm shift into the medium access control (MAC) design. In this paper, we present a distributed MAC framework assisted by machine learning for the Heterogeneous IoT system, where the IoT devices coexist with the WiFi users in the unlicensed industrial, scientific, and medical (ISM) spectrum. Specifically, the superframe is divided into two phases: a rendezvous phase and a transmission phase. During the rendezvous phase, the gateway that is capable of machine learning predicts the number of WiFi users and the IoT devices by performing a triangular handshake on the primary channel. The prediction takes advantage of the deep neural network (DNN) model which is pretrained on our universal software radio peripheral (USRP2) testbed offline. The gateway allocates the frequency channels to the WiFi and IoT systems based on the inference results. Then, the IoT devices and WiFi users initiate data transmissions during the transmission phase. Furthermore, system throughput is analyzed and optimized in two typical scenarios, respectively. An optimized MAC framework is proposed to maximize the total system throughput by finding the key design parameters. The analytical and simulation results that are conducted using the ns-2 demonstrate the effectiveness of the proposed MAC framework.

Rumor detection becomes increasingly important in social media. The effects of rumor propagation are dreadful in case of time-critical events, for example, during natural disasters. In this paper, we proposed a multiple time-series data analysis model to detect rumors on Twitter. Instead of checking the contents of the tweets, the proposed method only uses temporal properties of the tweets. As a result, the computational complexity measured by the training time and prediction time has been reduced significantly, which allows quick detection of rumors. Experimental results show that the proposed model combined with Gaussian Naive Bayes classifier achieved a high precision score of 94%.

Ahstract- While hard sensor fusion is a highly developed discipline with vast methods, the inclusion of soft evidence continues to gain significant interest because soft sensors in the fusion process has certain merits, such as the ability to model attributes of interest (e.g., emotional level) that hard sensor may not. However, how to combine the hard/soft sensor data efficiently is a challenging problem, especially when the data set becomes large. In this study, a novel algorithm is proposed to apply the Conditional Core Theorem (CCT) in computing the Fagin-Halpern conditionals in the fusion of bodies of evidence with disparate frames of discernment. The computational complexity of the proposed algorithm is derived analytically and simulations are carried out to demonstrate the efficiency of the proposed algorithm.

In future wireless systems, efficient spectrum usage enabled by technologies such as cognitive radio (CR), dynamic spectrum access, non-orthogonal multiuser within licensed and unlicensed band requires knowledge of prevalence situation in a frequency band through learning. It is critical for wireless devices to identify the type and number of users and their waveforms in a frequency band at a given time. In this paper, this problem is formulated as a multi-class classification problem for accurate spectrum situation prediction under complicated coexistence scenarios. Deep learning is chosen for obtaining the statistics of the different coexisting wireless systems through learning from superimposed radio frequency (RF) data. Specifically, Deep Neural Network (DNN), Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM) are adopted and designed. They are trained using real RF traces for different coexisting scenarios collected from a USRP based testbed to identify the presence of signals with varying levels of Signal to Noise Ratio, even when the signal is superimposed with other signals as sources of interference. Experimental results demonstrate the potential of deep learning for spectrum situation prediction under complicated coexistence scenarios.

The vision of the Internet-of-Things (IoT) networks calls for a large number of power constrained devices communicating with the gateway. To achieve the channel coordination in IoT, IEEE 802.15.4 standard has been considered as one of the most competitive technologies. However, the length of the Contention Access Period (CAP) of the superframe can hardly adapt to the variation of network traffic, so the performance of IoT is restricted. To resolve the problem, we propose a scalable MAC framework assisted by Machine Learning, called MML. With the implementation of machine learning algorithms such as Neural Network Predictor (NNP), the gateway can detect the number and type of devices from the overlapped signals, as demonstrated in our Universal Software Radio Peripheral (USRP2) testbed. Therefore, MML can dynamically adjust the CAP length based on the knowledge of the number of active devices and a stable throughput can be achieved. Moreover, the throughput of MML is analyzed, which is verified by conducting simulations using network simulator (ns-2.35). The analytical and simulation results demonstrate the superiority of the proposed MML.

BACKGROUND: Electronic Medical Record (EMR) comprises patients' medical information gathered by medical stuff for providing better health care. Named Entity Recognition (NER) is a sub-field of information extraction aimed at identifying specific entity terms such as disease, test, symptom, genes etc. NER can be a relief for healthcare providers and medical specialists to extract useful information automatically and avoid unnecessary and unrelated information in EMR. However, limited resources of available EMR pose a great challenge for mining entity terms. Therefore, a multitask bi-directional RNN model is proposed here as a potential solution of data augmentation to enhance NER performance with limited data. METHODS: A multitask bi-directional RNN model is proposed for extracting entity terms from Chinese EMR. The proposed model can be divided into a shared layer and a task specific layer. Firstly, vector representation of each word is obtained as a concatenation of word embedding and character embedding. Then Bi-directional RNN is used to extract context information from sentence. After that, all these layers are shared by two different task layers, namely the parts-of-speech tagging task layer and the named entity recognition task layer. These two tasks layers are trained alternatively so that the knowledge learned from named entity recognition task can be enhanced by the knowledge gained from parts-of-speech tagging task. RESULTS: The performance of our proposed model has been evaluated in terms of micro average F-score, macro average F-score and accuracy. It is observed that the proposed model outperforms the baseline model in all cases. For instance, experimental results conducted on the discharge summaries show that the micro average F-score and the macro average F-score are improved by 2.41% point and 4.16% point, respectively, and the overall accuracy is improved by 5.66% point. CONCLUSIONS: In this paper, a novel multitask bi-directional RNN model is proposed for improving the performance of named entity recognition in EMR. Evaluation results using real datasets demonstrate the effectiveness of the proposed model.

Recently opened spectrum within 3550-3700 MHz provides more accessing opportunities to secondary users (SUs), while it also raises concerns on the operational privacy of primary users (PUs), especially for military and government. In this paper, we propose to study the tradeoff between PUs' temporal privacy and SUs' network performance using the data-driven approach. To preserve PUs' temporal operational privacy, we develop an obfuscation strategy for PUs, which allows PUs to intentionally add dummy signals to change the distribution of temporal spectrum availability, and confuse the adversary. While generating the dummy signals for privacy, the PUs have to consider the utility of SUs and try their best to satisfy SUs' uncertain traffic demands. Based on the historical data, we employ a data-driven risk-averse model to characterize the uncertainty of SUs' demands. With joint consideration of PUs' privacy and uncertain SUs' demands, we formulate the data-driven risk- averse stochastic optimization, and provide corresponding solutions. Through numerical simulations, we show that the proposed scheme is effective in preserving PUs' temporal operational privacy while offering good enough spectrum resources to satisfy SUs' traffic demands.

Despite the recent advancements in glycemic control for diabetic patients, the realization of an automated closed-loop artificial pancreas is still a challenge. The purpose of this research is to develop an integrated control system for in silico closed loop administration of insulin for Type 1 diabetic patients based on patients' medical record and real-time control-relevant data. The proposed system consists of a virtual patient model from the online AIDA diabetes simulator, a neural network predictor trained on patients' data for feedback purposes, and a Proportional-Integral Controller and data logging nodes. The virtual patient takes into account the delayed and time-varying insulin and carbohydrate absorption rate associated with the existing subcutaneous insulin delivery and complex glucose metabolism, respectively. The neural network predictor was trained using 23 features including semi-static and dynamic data, with built-in knowledge of all available past blood glucose levels. Then the controller calculates the infusion bolus to be delivered by the insulin pump. Extensive simulations are performed and it is shown that the neural network predictor has less Root-Mean-Square error than the currently used continuous glucose monitors, which takes measurement from the interstitial fluid. Simulation results also demonstrate that our proposed data-driven closed loop system for glycemic control can effectively regulate the blood glucose level of Type 1 diabetic patients without hypoglycemic excursions, and with no preset instruction on meal ingestion.

In this paper, a transfer bi-directional recurrent neural networks (RNN) is proposed for named entity recognition (NER) in Chinese electronic medical records (EMRs) that aims to extract medical knowledge such as phrases recording diseases and treatments automatically. We propose a two-step procedure where the first step is to train a shallow bi-directional RNN in the general domain, and the second step is to transfer knowledge from the general domain to train a deeper bi-directional RNN for recognizing medical concepts from Chinese EMRs. Specifically, this is achieved by initializing the shallow parts of the deeper network in the second step with parameter weights from the bi-directional RNN trained in the first step. Then the deeper networks are re-trained on the Chinese EMRs. Experimental results show that NER performances are improved by the transferred knowledge significantly.

In this paper, big Radio Frequency (RF) data assisted optimization is considered for future wireless networks employing cognitive radio technology with machine learning capability. A cognitive radio network (CRN) with multiple Secondary Users (SUs) may coexist with other wireless systems such as Small Cells (SC) and Radar systems, both Primary Users (PUs) with different level of priorities. Traditional spectrum sensing typically only gives information about the presence or absence of a PU. However, when multiple heterogeneous systems coexist, it becomes imperative to acquire the knowledge of the systems operating in a specific band at a particular time so as to choose an appropriate transmission strategy. In this work, we take advantage of the learning capability of a Neural Network Predictor (NNP) to obtain the statistics of the coexisted wireless systems from the RF traces collected in our Universal Software Radio Peripheral (USRP) based test bed. The NNP is able to learn the features of the RF traces and make accurate prediction of the signals prevalent in the wireless environment. Because of the augmented information learned from the RF traces, a novel optimization problem incorporating the outputs from the NNP is formulated to maximize the throughput of the CRN. The solution is derived using Karush- Kuhn-Tucker (KKT) and extensive simulations using the real RF traces are carried out. It is demonstrated that the NNP can detect the type and number of coexisted users reliably and the proposed scheme will improve the performance of the coexisted CRN.

Summary With the proliferation of the Internet of Things, numerous sensors are deployed to monitor a phenomenon that in many cases can be modeled by an underlying stochastic process. The goal is to detect change in the process with tolerable false alarm rate. In practice, sensors may have different accuracy and sensitivity range, or they decay along time. As a result, the sensed data will contain uncertainties and sometimes they are conflicting. In this study, we propose a novel framework to take advantage of Dempster‐Shafer theory's capability of representation of uncertainty to detect change and effectively deal with complementary hypotheses. Specifically, Kullback‐Leibler divergence is used as the metric to find the distances between the estimated distribution with the before and after change distributions. Mass functions are calculated on the basis of those distance values for each sensor independently, and Dempster‐Shafer combination rule is applied to combine the mass values among all sensors. In the case of high conflict in various sensor readings, Dezert‐Smarandache combination rule is applied, and the belief, plausibility, and pignistic probability are obtained for decision making. Simulation results using both synthetic data and real data demonstrate the effectiveness of the proposed schemes.

Cognitive radio is one of the enabling technologies considered for the next generation communication systems for many mission-critical applications. In cognitive radio systems, cooperative spectrum sensing is one of the key techniques that can improve reliability and agility. In this paper, a framework that integrates quickest detection and belief propagation is applied to the cooperative spectrum sensing, where the primary user activities are heterogeneous in the space and dynamic in the time. The performance of the proposed scheme is analyzed mathematically. Using numerical simulations, detection performance measured by false alarm rate and average detection delay is obtained for different setups. The results show that the proposed scheme achieves better receiver operational curves than traditional detection method.

Although mesh and tree communication network topologies both bear some advantages, none of them alone can satisfy the robustness and real-time requirements of data collection in the Advanced Metering Infrastructure (AMI). To address this problem, a biconnected tree based topology is proposed in this work. Specifically, in order to mitigate the effects of unreliable spanning tree-based communication links, an augmented tree with 2-connectivity is designed to enhance the robustness of the AMI tree yet keeping the overhead and delay low. As a result, the proposed biconnected AMI tree will continue to sustain the communication between functioning Smart Meters (SMs) when any single communication link fails. At the same time, it maintains low signaling overhead and meets the real-time requirement of data collection. The detailed algorithms to create such biconnected AMI tree and related theoretical analysis are provided. The simulation results demonstrate the effectiveness of the proposed scheme.

Numerous applications of wireless sensor networks (WSNs) are hindered by the limited battery power of the sensors. Instead of using the fixed amount of battery power, in this paper, we propose to integrate supercapacitors into the sensors, and let the sensors wirelessly harvest the relatively unlimited Radio Frequency (RF) energy for the perpetual operation of WSNs. To further facilitate the energy harvesting of WSNs in harsh terrains, we employ an unmanned vehicle to provide dedicated RF signals, and develop a dynamic optimization scheme to control the moving of the vehicle. Specifically, we propose to amount a dedicated RF energy source on a mobile unmanned vehicle, and let the vehicle periodically localize sensors, dynamically select the sensors to recharge, and find its optimal sojourn time, so that the overall operation of the WSN is optimized. Based on the RF energy harvesting equipment of Powercast Corp., we establish the testbed and conduct a series of experiments to verify the effectiveness of the proposed scheme.

This study takes advantage of the ability of Physical-Layer Network Coding (PNC) to embrace electromagnetic interference, improve spectral efficiency and achieve throughput optimality in a femtocell operated as a relay. Specifically, we make a case for a femtocell operating as a relay and utilizing PNC to eliminate the interference that exists between macrocells user equipments (MUEs) and femtocells home users equipments (HUEs) in an Opened Subscriber Group (OSG) co-channel deployment. With PNC implemented at the HNB relay, it becomes possible for the relay to take advantage of the information signals transmitted from both the HUE and MUE rather than treating the MUE's signals as interference. To guarantee the Service Quality (QoS) of femtocell users, we develop the closed-form expressions of the cummulative distributions function (CDF) of the received signal-to-noise ratios (SNR) and the outage probability of the HUE analytically. Simulation results match the analytical results and it is demonstrated that PNC is suitable for a relay operated femtocell network.

In this work, we consider Quality of Service (QoS) guarantee for Device to Device (D2D) users co-existing with a cellular system where the D2D communication links are sharing radio spectrum resources with macrocell users in the downlink. Despite the lofty advantages associated with D2D communications, one major concern is the resulting interference from the D2D users, which should not infringe on the service quality requirements of the User Equipments (UEs). In this work, we investigate the scenario in which D2D communications operates simultaneously with downlink transmissions from the Evolved Node B (eNB). Power control problem for the D2D users is formulated in order to optimize the energy efficiency of the eNB users as well as to ensure that QoS of D2D devices and UEs does not fall below the acceptable target. The feasible conditions of the power control problem are derived and then the centralized and the distributed solutions are obtained. We further suggest jointly designed dynamic power control and channel re-allocation algorithm that will guarantee the priority of the UEs. Effectiveness of the proposed scheme is demonstrated through extensive simulations.

In wideband cognitive radio (CR) networks, spectrum sensing is one of the key issues that enable the whole network functionality. Collaborative spectrum sensing among the cognitive radio nodes can greatly improve the sensing performance, and is also able to obtain the location information of primary radios (PRs). Most existing work merely studies the cognitive radio networks with static PRs, yet how to deal with the situations for mobile PRs remains less addressed. In this paper, we propose a collaborative compressive sensing based approach to estimate both the power spectrum and locations of the PRs by exploiting the sparsity facts: the relative narrow band nature of the transmitted signals compared with the broad bandwidth of available spectrum and the mobile PRs located sparsely in the operational space. To effectively track mobile PRs, we implement a Kalman filter using the current estimations to update the location information. To handle dynamics in spectrum usage, a dynamic compressive spectrum sensing algorithm is proposed. Joint consideration of the above two techniques is also investigated. Simulation results validate the effectiveness and robustness of the proposed approach.

In this paper, the problem of spectrum situation reconstruction is considered for large cognitive radio ad hoc networks. The major challenge of such problem lies in the fact that only very limited measurements of spectrum occupancy may be obtained by the cognitive users for a certain location at any given time slot. This is due to both the hardware limitations as well as the tradeoff between spectrum sensing and data throughput of the cognitive users. By representing the spectrum sensing results across the network as an image, we formulate the problem of spectrum situation reconstruction as an image recovery problem. The method of total variation inpainting is applied to solve the problem with low recovery error. The proposed method takes advantage of the correlations in multiple dimensions and the simulation results demonstrate the effectiveness of the proposed scheme.

In this paper, we propose to study the treatment and drug effects at the molecular level using a hybrid system model. Specifically, we propose a generic piecewise linear model to analyze drug effects on the state of the genes in a genetic regulatory network. We intend to answer the following question: given an initial state, would a treatment or drug (control input) drive the target gene to a new desired state that are not reachable without the treatment or drug? assuming that the concentration level of the drug remains constant. In other words, we try to identify whether there is a chance that the treatment or drug will be effective for changing gene expressions at all. We provide detailed analysis for two cases. In the first case, there is only one target gene; while in the second case, there is also another gene interacting with the target gene. The relationships between various parameters (of the genetic regulatory network and the design of the drug) and the convergence and the steady state of the controlled genes are derived analytically and discussed in detail. Simulations are performed using MATLAB/SIMULINK and the results confirmed our analytical findings.

In this paper, the quantitative analysis of time-series gene expression data on inference of gene regulatory networks is performed. Time-series gene data are modeled by the B-Spline algorithm to improve the overall smooth expression curves which can further reduce over-fitting. The effect of the different sizes of observed time-series data on gene regulatory networks inference is analyzed. The stochastic errors introduced by the B-Spline algorithm to the system are evaluated. The precision of different sizes of time-series data on parameter estimations is compared. With application of the B-Spline to generate continuous curves, simulation results can be much more accurate and inference results are significantly improved. Both synthetic data and experimental data from microarray measurements are used to demonstrate the effectiveness of the proposed method.

Data collection is of paramount importance in many applications of wireless sensor networks (WSNs). Especially, to accommodate ever increasing demands of signal source coding applications, the capacity of processing multi-user data query is crucial in WSNs where the efficiency is one key consideration. To that end, this paper presents EDCA: an Efficient Data Collection Approach for data query in WSNs, which exploits recent matrix completion techniques. Specifically, for the efficiency of energy consumption, we randomly select a part of nodes from the sensor network to sample at each time instance and directly forward the data to the sink. Then, to recover the data precisely, we shift the rank minimization problem, which is NP-hard, to a convex optimization one. Compared with the centralized scheme, energy consumption using EDCA is significantly reduced due to lower sampling rate and fewer packets to transmit. The experimental results demonstrate that EDCA significantly outperforms the existing naive method in terms of energy consumption and the introduced errors are quite trivial.

Spectrum under-utilization calls for the open and dynamic spectrum access mechanism, which allows the unlicensed user equipped with cognitive radios to opportunistically sense and access the spectrum that not occupied by primary users. In practice due to the hardware limitations, each cognitive radio user may be only able to sense a portion of the interested wide span spectrum. Hence, a hardware-constrained cognitive MAC to conduct efficient and intelligent spectrum sense decision is desired. In this paper, we formulate the cognitive radio spectrum sensing problem under time-varying channels as an adversarial bandit problem without any assumption of the channel statistics. A fully distributed strategy is proposed to address the fundamental tradeoff between spectrum exploration and spectrum exploitation during the sensing periods. Simulation results demonstrate that significant performance gain can be achieved by the proposed algorithm when the channels are time-varying on small time-scales. A coordination scheme for the multi-user case is also presented and the effectiveness is also demonstrated by the simulation results.

Although Ordinary Differential Equations (ODEs) have been used to model Genetic Regulatory Networks (GRNs) in many previous works, their steady-state behaviors are not well studied. However, a phenotype corresponds to a steady-state gene expression pattern and steady-state analysis of GRNs can provide valuable information on the stability of the GRNs, insights into cellular regulatory mechanisms underlying disease development as well as possible interventions for disease control. In this study, the steady-state behaviors of the nonlinear GRN models are analyzed based on time series data. The steady-state solutions and stability of nonlinear GRNs including polynomial model, sigmoidal model and S-system model are discussed in details.

In this paper, a novel synchronization protocol is proposed especially for Cognitive Radio (CR) networks called CR-Sync. In a CR network, time synchronization is indispensable because of the requirements for coordinated and simultaneous quiet periods for spectrum sensing, as well as the common understanding of time frame/slot in many CR MAC designs. The proposed CR-Sync achieves network-wide time synchronization in a fully distributed manner, i.e., each node performs synchronization individually using CR-Sync. Contrary to many existing synchronization protocols that do not exploit CR attributes, the proposed protocol takes advantage of the potential multiple spectrum holes that are discovered by CR and distributes the synchronization of different pairs of nodes to distinct channels and thus reduces the synchronization time significantly. Detailed analysis of synchronization error and convergence time are provided. Results show that the proposed CR-Sync out-performs other protocols such as TPSN in CR networks.

Determining liver segmentation accurately from MRIs is the primary and crucial step for any automated liver perfusion analysis, which provides important information about the blood supply to the liver. Although implicit contour extraction methods, such as level set methods (LSMs) and active contours, are often used to segment livers, the results are not always satisfactory due to the presence of artifacts and low-gradient response on the liver boundary. In this paper, we propose a multiple-initialization, multiple-step LSM to overcome the leakage and over-segmentation problems. The multiple-initialization curves are first evolved separately using the fast marching methods and LSMs, which are then combined with a convex hull algorithm to obtain a rough liver contour. Finally, the contour is evolved again using global level set smoothing to determine a precise liver boundary. Experimental results on 12 abdominal MRI series showed that the proposed approach obtained better liver segmentation results, so that a refined liver perfusion curve without respiration affection can be obtained by using a modified chamfer matching algorithm and the perfusion curve is evaluated by radiologists.

In this paper, energy efficient spectrum access is considered for a wireless cognitive radio ad hoc network, where each node is equipped with cognitive radio, has limited energy, and the network is an OFDMA system operating on time slots. In each slot, the users with new traffic demand will sense the spectrum and locate the available subcarrier set. Given the data rate requirement and maximal power limit, a constrained optimization problem is formulated for each individual user to minimize the energy consumption per bit over all selected subcarriers, while avoid introducing harmful interference to the existing users. Because of the multi-dimensional and non-convex nature of the problem, a fully distributed subcarrier selection and power allocation algorithm is proposed by combining an unconstrained optimization method with a constrained partitioning procedure. Due to the non-cooperative behavior among new users, they will execute distributed power control to manage the co-channel interference when needed. Simulation results demonstrate that the proposed scheme performs tightly to the global optimal solution. In addition, the comparison between the proposed energy efficient allocation scheme and the well established rate or power efficient allocation algorithms is carried out to demonstrate the advantage of the proposed scheme in terms of network lifetime.

In this paper, we investigate the issues of survivability and efficiency in a heterogeneous mobile ad hoc network (MANET) and Sensor network as a combined network where the communications is achieved using a mixed MANET and Sensor Network, referred to as SMANET for supporting mission critical applications. Using the cluster managed SMANET where, heterogeneous nodes with different capabilities and nodes from different organizations deployed within the same geographical area will support variety of multi-service mission critical applications while increasing the network lifetime. In this network, nodes from different organizations collaborate with each other for mission critical data to be delivered using multi-hop radio-sensor-radio type path routing while keeping the tactical operations supported by the battlefield command control structure transparent. While keeping the collaborative encounters, it is important to keep cross-organization data transfer at the lowest possible level for minimal disruption of the traffic within each organization and for security reasons. Assuming that all nodes are equipped with cognitive radio capability maintaining interoperability, we propose a novel routing metric that maximize the benefits from the collaboration of heterogeneous nodes and take the organizational constraints into account. Furthermore, we propose joint design of cognitive radio and multi-carrier modulation to maximize the bandwidth utilization and energy efficiency in SMANET. Simulation results demonstrate that the proposed method provide much needed survivability and efficiency in battlefield environment while keeping the cross-organization data transfer at the lowest possible level as a value added enhancement compared to homogeneous MANET or sensor networks. Same network design can also find use in civilian environment for supporting applications such as emergency preparedness management and traffic efficiency in congested transportation systems.

In this paper, a wireless cognitive radio sensor network is considered, where each sensor node is equipped with cognitive radio and the network is a multi-carrier system operating on time slots. In each slot, the users with new traffic demand will sense the entire spectrum and locate the available subcarrier set. Given the required data rate and power bound, a fully distributed subcarrier selection and power allocation algorithm is proposed for each individual user to minimize the energy consumption per bit over all subcarriers, while avoid introducing harmful interference to the existing users. The multi-dimensional and non-quasi-convex/concave nature of the energy efficiency optimization problem in multi-carrier systems makes it more challenging than throughput/power optimization problems or the energy efficiency problem in the single carrier system. The optimal solution is derived by using a two-stage algorithm where the original problem is decoupled into an unconstrained problem and branch and bound method is applied thereafter to reduce the search space. In addition, a distributed power control is performed to manage the co-channel interference among new users when needed. Simulation results demonstrate that the proposed approach performs close to the centralized optimal solution, and it provides prolonged network lifetime.

In this paper, a unified approach to infer gene regulatory networks using the S-system model is proposed. In order to discover the structure of large-scale gene regulatory networks, a simplified S-system model is proposed that enables fast parameter estimation to determine the major gene interactions. If a detailed S-system model is desirable for a subset of genes, a two-step method is proposed where the range of the parameters will be determined first using genetic programming and recursive least square estimation. Then the exact values of the parameters will be calculated using a multi-dimensional optimization algorithm. Both downhill simplex algorithm and modified Powell algorithm are tested for multi-dimensional optimization. Simulation results using both synthetic data and real microarray measurements demonstrate the effectiveness of the proposed methods

In this paper, we consider the lifetime maximization problem in a wireless cognitive radio sensor network, where the joint design of cognitive radio and multi-carrier modulation is proposed to achieve high power efficiency. Sensor nodes first sense the entire spectrum and locate the available subcarriers based on a pilot tone detection scheme. After each node locates the available subcarrier set, information is transmitted over the favorite channel that has the largest channel gain. Under this setting, an adaptive modulation strategy is proposed to maximize the network lifetime by selecting the optimal constellation size. Simulation results demonstrate the effectiveness of our approach. In addition, the impact of conflicting transmissions upon the network lifetime is also investigated.

While FCC proposes spectrum sharing between a legacy TV system and a cognitive radio network to increase spectrum utillization, one of the major concerns is that the interference from the cognitive radio network should not violate the QoS requirements of the primary users. In this paper, we consider the scenario where the cognitive radio network is formed by secondary users with low power personal/portable devices and when both systems are operating simultaneously. A power control problem is formulated for the cognitive radio network to maximize the energy efficiency of the secondary users and guarantee the QoS of both the primary users and the secondary users. The feasibility condition of the problem is derived and both centralized and distributed solutions are provided. Because the co-channel interference are from heterogeneous systems, a joint power control and admission control procedure is suggested such that the priority of the primary users is always ensured. The simulation results demonstrate the effectiveness of the proposed schemes.

In this paper, joint power control and maximally disjoint routing is proposed for multihop CDMA wireless ad hoc networks. A framework of power control with QoS constraints in CDMA wireless ad hoc networks is introduced and the feasibility condition of the power control problem is identified. Both centralized solution and distributed implementations are derived to calculate the transmission power given required throughput and the set of transmitting nodes. Then a joint power control and maximally disjoint routing scheme is proposed for routing data traffic with minimum rate constraint while maintaining high energy efficiency and prolonged network lifetime. Furthermore, in order to provide reliable end-to-end data delivery, the proposed joint power control and maximally disjoint routing scheme is augmented by a dynamic traffic switching mechanism to mitigate the effect of node mobility or node failure. Simulation results demonstrate the effectiveness of the proposed scheme

In order to achieve high end-to-end throughput in a multihop wireless ad hoc network, TD/CDMA has been chosen as the medium access control (MAC) scheme due to its support for high network throughput in a multihop environment. The associated power control and scheduling problem needs to be addressed to optimize the operations of TD/CDMA. In this paper, cluster based architecture is introduced to provide centralized control within clusters, and the corresponding power control and scheduling schemes are derived to maximize a network utility function and guarantee the minimum rate required by each traffic session. Because the resulted optimal power control reveals bang-bang characteristics, i.e., scheduled nodes transmit with full power while other nodes remain silent, the joint power control and scheduling problem is reduced to a scheduling problem. In order to achieve a balance between throughput and fairness, proportional fair scheduling is considered. The multi-link version of the proportional fair scheduling algorithms for multihop wireless ad hoc networks are proposed. In addition, a generic token counter mechanism is employed to satisfy the minimum rate requirements. Approximation algorithms are suggested to reduce the computational complexity. In networks that are lack of centralized control, distributed scheduling algorithms are also derived and fully distributed implementation is provided. Simulation results demonstrate the effectiveness of the proposed schemes

In this paper, the uplink power control problem is considered for CDMA cellular systems, where stochastic SIR measurements are performed at base stations. A distributed stochastic power control algorithm is proposed assuming SIR measurements contain white noise. The proposed scheme minimize the sum of variances of mobile's transmission power and signal-to-interference error. The algorithm derived is fully distributed in the sense that each user only needs to know its own signal-to-interference measurement and channel variation. Uncertainties of wireless channels are accommodated by using a robust estimator. Simulation results indicate that the proposed power control scheme has very fast convergence. In addition, it also works fine with 4-bit quantization.

Abstract In order to achieve high end‐to‐end throughput in a multihop wireless ad hoc network, TD/CDMA has been chosen as the Medium Access Control (MAC) scheme due to its support for high network throughput in a multihop environment. The associated power control and scheduling problem needs to be addressed to optimize the operations of TD/CDMA. In this paper, cluster‐based architecture is introduced to provide centralized control within clusters, and the corresponding power control and scheduling schemes are derived to maximize a network utility function and guarantee the minimum rate required by each traffic session, given routes for multiple end‐to‐end multihop traffic sessions. Because the resulted optimal power control reveals bang‐bang characteristics, that is, scheduled nodes transmit with full power while other nodes remain silent, the joint power control and scheduling problem is reduced to a scheduling problem. The multi‐link version of the throughput‐optimal and the proportional fair scheduling algorithms for multihop wireless ad hoc networks are proposed. In addition, a generic token counter mechanism is employed to satisfy the minimum rate requirements. By ensuring different minimum rate for different traffic sessions, service differentiation is also achieved. Approximation algorithms are suggested to reduce the computational complexity. In networks that are lack of centralized control, distributed scheduling algorithms are also derived and fully distributed implementation is provided. Simulation results demonstrate the effectiveness of the proposed schemes. Copyright © 2006 John Wiley & Sons, Ltd.

We consider the problem of scheduling multiple users sharing a time-varying wireless channel. (As an example, this is a model of scheduling in 3G wireless technologies, such as CDMA2000 3G1xEV-DO downlink scheduling.) We introduce an algorithm which seeks to optimize a concave utility function /spl Sigma//sub i/H/sub i/(R/sub i/) of the user throughputs R/sub i/, subject to certain lower and upper throughput bounds: R/sub i//sup min//spl les/R/sub i//spl les/R/sub i//sup max/. The algorithm, which we call the gradient algorithm with minimum/maximum rate constraints (GMR) uses a token counter mechanism, which modifies an algorithm solving the corresponding unconstrained problem, to produce the algorithm solving the problem with throughput constraints. Two important special cases of the utility functions are /spl Sigma//sub i/log R/sub i/ and /spl Sigma//sub i/R/sub i/, corresponding to the common proportional fairness and throughput maximization objectives. We study the dynamics of user throughputs under GMR algorithm, and show that GMR is asymptotically optimal in the following sense. If, under an appropriate scaling, the throughput vector R(t) converges to a fixed vector R/sup +/ as time t/spl rarr//spl infin/ then R/sup +/ is an optimal solution to the optimization problem described above. We also present simulation results showing the algorithm performance.

The application of multi-path techniques in wireless ad hoc networks is advantageous because multi-path routing provides means to combat the effect of unreliable wireless links and constantly changing network topology. The performance of multi-path routing under wormhole attack is studied in both cluster and uniform network topologies. Because multi-path routing is vulnerable to wormhole attacks, a scheme called statistical analysis of multi-path (SAM) is proposed to detect such attacks and to identify malicious nodes. As the name suggests, SAM detects wormhole attacks and identifies attackers by statistically analyzing the information collected by multi-path routing. Neither additional security services or systems nor security enhancement of routing protocols is needed in the proposed scheme. Simulation results demonstrate that SAM successfully detects wormhole attacks and locates the malicious nodes in networks with cluster and uniform topologies and with different node transmission range.

We consider meshed optical transport networks having multiple levels of hierarchy whereby cross-connects in different levels perform multiplexing and demultiplexing functions at different granularities. We present a traffic grooming approach that can be implemented in a centralized or distributed fashion, based on the novel design that takes advantage of algorithm efficiency and a simple threshold mechanism to decide when grooming is economical. Our approach allows nodes to perform grooming and degrooming automatically, which is desired in the next-generation optical network where circuits are setup and torn-down dynamically through signaling. We present several experiments using different network topologies. The results indicate that the flow pattern influences the port requirement behavior, and that the flow thickness influences the optimal value of the threshold.

Uplink scheduling in wireless systems is gaining importance due to arising uplink intensive data services (ftp, image uploads etc.), which could be hampered by the currently in-built asymmetry in favor of the downlink. In this work, we propose and study algorithms for efficient uplink packet-data scheduling in a CDMA cell. The algorithms attempt to maximize system throughput under transmit power limitations on the mobiles assuming instantaneous knowledge of user queues and channels. However no channel statistics or traffic characterization is necessary. Apart from increasing throughput, the algorithms also improve fairness of service among users, hence reducing chances of buffer overflows for poorly located users. The major observation arising from our analysis is that it is advantageous on the uplink to schedule "strong" users one-at-a-time, and "weak" users in larger groups. This contrasts with the downlink where one-at-a-time transmission for all users has shown to be the preferred mode in much previous work. Based on the optimal schedules, we propose less complex and more practical approximate methods, both of which offer significant performance improvement compared to one-at-a-time transmission, and the widely acclaimed Proportional Fair (PF) algorithm, in simulations. When queue content cannot be fed back, we propose a simple modification of PF, Uplink PF (UPF), that offers similar improvement.

Uplink scheduling in wireless systems is gaining importance due to arising uplink intensive data services, which could be hampered by the currently in-built asymmetry in favor of the downlink. In prior work, (K. Kumaran and L. Qian, 2002, Apr. 2003), we proposed optimal algorithms for uplink scheduling in a CDMA cell that does not employ any form of interference cancellation. In this work, we modify the approach to incorporate successive interference cancellation (SIC), which has been shown to be optimal in an information theoretic sense (G. Caire and S. Shamai, 2000). As in K. Kumaran and L. Qian ( 2002, Apr. 2003), no statistical assumptions are made about channel or traffic behavior, but feedback to communicate current channel state and queue state is assumed. Our results demonstrate that the throughput optimal scheduling strategy takes a particularly simple form with SIC as compared to without, (K. Kumaran and L. Qian, 2002, Apr. 2003), apart from providing some level of performance improvement. A reasonable alternative algorithm based purely on received power can be constructed based on early work on SIC (P. Patel and J. Holtzman, June 1994). Considering decoding errors, only strongly received users can benefit from SIC. Our simulation experiments suggest that our throughput optimal scheduling improves performance over the alternative when users have similar received power. Combining the above observation, we also propose a hybrid scheduling algorithm that performs SIC for strong users and simultaneous transmission for weak users.

In this paper, a stochastic uplink power control problem is considered for CDMA systems. A distributed algorithm is proposed based on stochastic linear quadratic optimal control theory assuming SIR measurements contain white noise. The proposed scheme minimizes the sum of the variance of the mobile's transmission power and the variance of SIR error, with guaranteed stability. Simulation results indicate the effectiveness of the proposed algorithm.

The problem of achieving high accuracy positioning of a medical robot is studied. The Northeast Proton Therapy Center, at the Massachusetts General Hospital, is a new cancer research and treatment facility. A major component of the center is a robotic patient positioning system that will carry and position patients in a proton beam. The desired positioning accuracy of the robot is less than 0.5 mm. However, various sources of errors in the robot such as assembly errors or flexible deformation of the robot links result in big end-effector positioning errors. It is important to know these end-effector errors as a function of the robot joint variables and patient weight, to be able to compensate them using the manipulator controller. A multi-layer neural network is proposed to identify the robot positioning errors. The neural network is trained using the Levenberg-Marquardt method. Simulations and experimental results demonstrate the validity of the neural network.