Predicting epileptic seizures effectively allows patients to take preventive measures in advance, reducing accident risk and enhancing safety. Several modeling challenges remain open: (1) The complex spatiotemporal dependency of EEG signals makes it challenging to design a model that efficiently extracts spatial and temporal features from multi-channel EEG signals to classify epileptic EEG signals. (2)While many studies have utilized machine learning and deep learning for seizure prediction, they often lack research on model interpretability. To tackle the challenges above, this paper proposes a novel approach—an epilepsy prediction framework that combines EEG-Inception Convolutional Neural Networks (EICNN) with Feature Pattern Interpretability Post-processing (FPIP). It yielded substantial performance improvements and achieved an average sensitivity of 94.6%, a false prediction rate of 0.29/h in the study involving 22 pediatric patients from the CHB-MIT database using the leave-one-out method. The proposed FPIP provides visual interpretations and significantly enhances performance in predicting epileptic seizures.

The dynamic propagation of epileptic discharges complicates Drug-Resistant Epilepsy (DRE) seizure detection using traditional machine learning methods and Stereotactic Electroencephalography (SEEG). Several challenges remain unresolved in prior studies: (1) incomprehensive representations of epileptic brain network features; (2) lacking of flexible and dynamic mechanisms to learn brain network evolving features; and (3) the absence of model mechanisms interpretation corresponds with seizure mechanisms. In response, we propose a novel multi-band dynamic graph attention network, DynSeizureGAT, to detect and analyze DRE seizures with precision and interpretability. Specifically, a seizure network sequence is first constructed by integrating a multi-band directed transfer function matrix and enhanced epileptic index node features. Second, a dynamic graph attention module is integrated to dynamically weigh the contribution of various spatial scales. Third, spatial-spectral-temporal attention mechanisms enhance the model's capacity to better characterize and interpret the ictal and interictal states. Extensive experiments are conducted on the large-scale public clinical SEEG dataset (OpenNeuro). The proposed model demonstrates high seizure detection performance, achieving an average of 94.6% accuracy, 93.4% sensitivity, and 96.4% specificity. In addition, the importance of frequency bands and dynamic abnormal connectivity patterns is successfully quantified and visualized, which contributes most to the explainability. Experimental results indicate that DynSeizureGAT demonstrates strong dynamic propagation feature learning capability, corresponding with seizure propagation mechanisms, and is promising to assist DRE epileptogenic zone localization.

Accurate and reliable prediction of epileptic seizures has potential in epilepsy monitoring, diagnosis, and rehabilitation. Current mainstream deep learning models treat the epileptic seizure prediction problem as a binary classification task, where hard labels are assigned to all signal segments for modeling. However, since the signal classes are manually selected, this approach can lead to many false positives during classification. Moreover, using hard labels that do not vary over time results in the inability to maintain the accuracy’s monotonicity over time during prediction. This means that the system can only provide results after a certain period once the alarm has started, which may cause anxiety for patients. This paper proposes a new modeling approach that uses early event prediction to model the labels of signals preceding seizure onset. Additionally, an innovative introduction of a large EEG model into the seizure prediction task is presented. The proposed model is validated on the CHB-MIT dataset, achieving an overall sensitivity of 94.8% and reducing false positive alarms to 0.03. Comparison with state-of-the-art methods indicates that the proposed model exceeds previous prediction performance. Furthermore, our method demonstrates the potential for maintaining accuracy monotonicity over time for certain patients, allowing them to predict seizures based on the model’s time-based judgments, thereby alleviating anxiety.

Focal cortical dysplasia (FCD) is a significant cause of intractable epilepsy, with its identification and segmentation being crucial for diagnosis. While fully-supervised learning models can detect and segment FCD, they are limited by the need for extensive labeled data, which is time-consuming and requires expert annotation. To address this, we introduce a semi-supervised model for FCD segmentation, leveraging a small set of labeled data alongside unlabeled data. Our method estimates uncertainty by measuring voxel-wise deviations between predictions and ground truth, using both ground truth and pseudo-labels to refine segmentation masks. For unlabeled data, multi-scale predictions capture different frequency components, allowing for selective consistency regularization based on region reliability. This approach is the first application of semi-supervised learning in epilepsy segmentation and demonstrates superior performance over state-of-the-art methods in experiments on clinical and public datasets.

We propose a Multi-Band Adaptive Transformer (MBAT) model designed to enhance the accuracy of seizure detection and seizure onset zone (SOZ) localization using electroencephalography (EEG) data. Our model integrates multiband feature extraction with an adaptive transformer architecture to integrate cross-channel information. In order to improve the sensitivity of the model to different EEG channels, we design squeeze-excitation attention (SEAtt) layer to dynamically recalibrates the importance of channels. Then we utilize a gated recurrent unit (GRU) layer for seizure detection. The detection results are used as attention weights in an attention-weighted multi-instance pooling mechanism for SOZ localization. We validate our model through cross-validation on a dataset of 90 patients curated from the Temple University Seizure (TUSZ) corpus. Extensive experiments demonstrate that our model performs better in both seizure detection and SOZ localization tasks. Ablation studies validate the effectiveness of each component, highlighting the significance of our design choices.

Medical imaging is now a widely used test for the preoperative evaluation of focal cortical dysplasia (FCD). Deep learning-based methods can learn lesion features from image data to automatically recognize and segment FCD in epilepsy treatment. However, the existing FCD segmentation networks lack the ability to fully extract the FCD lesion information and automatically aggregate the salient features of the lesions, the segmentation accuracy needs to be improved. To this end, we propose an end-to-end 3D Convolutional Neural Network segmentation model, Multi Attention Two-Path Residual UNet (MATPR-UNet). Specifically, we propose two modules: (1) Two-Path Residual Attention module, which can extract both local and global information, and suppress invalid information by fully fusing the features in channel and space; (2) Spatially Gated Attention module, which enables the model to automatically focus on the FCD lesion region, highlighting its salient features. We combine the proposed two modules with the 3D UNet to construct the MATPR-UNet. Extensive experiments on the private FCD dataset and the public EPISURG dataset demonstrate that our method outperforms other state-of-the-art methods, and is robust.

This paper explores the joint optimization problem for trajectory planning and radio resource allocation in unmanned aerial vehicle (UAV) communications with the aim of maximizing data collection. Rather than decomposing the problem into subproblems, as most current approaches do, we express the quantity of data gathered by a UAV-assisted network as a function of both the size of the resource block allocated to all ground devices and their average upload rate. Based on this formula, it can be concluded that the problem of maximizing the average data collection can be reduced to minimizing the flight trajectory if each device communicates with the UAV within the maximum allowable coverage of the UAV. To address this issue, we propose an advanced hierarchical clustering algorithm that divides larger network-scale scenarios into many disjoint subregions to determine the initial hovering positions of the UAV. The non-convex minimization trajectory problem is decomposed into a series of convex optimizations to minimize path segments along the trajectory, based on the traveling salesman problem (TSP). Subsequently, the communication optimization process is modified to assign specific upload times for each device. The effectiveness of the optimization algorithm is demonstrated through extensive simulations, which show its superior performance in terms of average rates of data collection and upload failures.

Alzheimer’s disease (AD) is a neurodegenerative brain disorder of unknown etiology that has a significant impact on the lives of patients and their families. Currently, there are no drugs or treatments that can cure AD, but early diagnosis and intervention can mitigate the effects of the disease on patients. Of various brain imaging tools, structural magnetic resonance imaging (sMRI) has been most intensively studied for AD diagnosis as it provides imaging biomarkers of neuronal loss. However, the accuracy of visual inspection by doctors is limited. Therefore, computer-aided diagnosis (CAD) based on sMRI has important clinical significance. Previous research has employed traditional machine learning and deep learning methods for AD image classification. However, these methods face challenges due to the scarcity, high noise, and high redundancy of medical data. Recent studies demonstrate the effectiveness of self-distillation in enhancing the model robustness, but the lack of additional knowledge limits such improvement. To address the issues of high redundancy in medical data and the lack of additional knowledge in self-distillation architectures, we propose the Self-Distillation Adversarial Training Network (SDATNet), which integrates self-distillation and adversarial training into a single framework. We utilize adversarial training methods to simulate noise on medical images, thereby supplementing additional information. Through self-distillation, we achieve cross-scale information interaction, enabling the extraction of discriminative features and effectively improving model performance. Experimental validation on the ADNI dataset demonstrates that our model outperforms other methods on publicly available datasets, achieving an accuracy of 92.77% and specificity of 92.19%. Our ablation experiments and visualization results further validate the reliability and superiority of the model.

Structural magnetic resonance imaging (sMRI) plays an important role in Alzheimer’s disease (AD) diagnosis. AD-related brain atrophy exhibits subtle local structural changes. Therefore, the key to AD diagnosis lies in pathological region identification and discriminative feature learning. However, existing methods often overlook the spatial and feature relationships between brain regions. To address this, we propose a graph learning framework based on contrastive learning and graph transformer for AD diagnosis using sMRI. First, we use a supervised contrastive learning strategy to learn discriminative brain representations from sMRI data. The features of the regions of interest (ROIs) are computed using a brain template to represent the graph nodes. Then, we build a multi-level connected brain network that incorporates spatial and feature information. Finally, a graph transformer network is introduced to gradually aggregate graph information for AD diagnosis. Our method achieves good performance in AD diagnosis using sMRI data from the ADNI dataset. Additionally, important brain regions such as the hippocampus, amygdala, and insula are identified as highly associated with Alzheimer’s disease.

The success of supervised deep learning heavily depends on large labeled datasets whose construction is often challenging in medical image analysis. Contrastive learning, a variant of self-supervised learning, is a potential solution to alleviate the strong demand for data annotation. In this work, we extend the contrastive learning framework to 3D volumetric medical imaging. Specifically, we propose (1) multiview contrasting strategy to maximize the mutual information between three views of 3D image to learn global representations and (2) long-short spatial contrasting strategy to learn local representations by matching a short spatial clip to a long spatial clip in the latent space. To combine these two strategies, we propose multiview long-short spatial contrastive learning (MLSSCL) framework, which can effectively learn generic 3D representations. Our extensive experiments on two brain Magnetic Resonance Imaging (MRI) datasets demonstrate that MLSSCL significantly outperforms learning from scratch and other self-supervised learning methods on both classification and segmentation tasks.

Intracranial electroencephalogram (IEEG) is an invasive procedure widely used for preoperative assessment of drug-resistant epilepsy (DRE). In recent years, methods based on machine learning achieved high accuracy in automatic EEG recognition. However, these models have also grown in complexity, requiring a large amount of time for various feature extraction or signal transformation, which makes it difficult to efficiently process the IEEG recordings that span days to weeks. In this study, we propose IEEG-TCN, a concise temporal convolutional network that performs nicely on different IEEG classification tasks while costing less time. According to experimental results on the Bern-Barcelona EEG dataset, our method reaches an impressive accuracy of 93.76% in detecting focal signals. The latency for processing each IEEG segment is reduced by 19 seconds. We also verify the robustness of IEEGTCN based on the Multicenter IEEG dataset that contains four categories of IEEG segments (physiological activity, pathological activity, artifacts, and noise). The results show that the model can be successfully generalized to multi-class problems.

Multiple Sclerosis (MS) lesion segmentation from MR images is important for neuroimaging analysis. MS is diffuse, multifocal, and tend to involve peripheral brain structures such as the white matter, corpus callosum, and brainstem. Recently, U-Net has made great achievements in medical image segmentation area. However, the insufficiently use of context information and feature representation, makes it fail to achieve segmentation of MS lesions accurately. To solve the problem, 3D attention context U-Net (ACU-Net) is proposed for MS lesion segmentation in this paper. The proposed ACU-Net includes 3D spatial attention block, which is used to enrich spatial details and feature representation of lesion in the decoding stage. Furthermore, in the encoding and decoding stage of the network, 3D context guided module is designed for guiding local information and surrounding information. The proposed ACU-Net was evaluated on the ISBI 2015 longitudinal MS lesion segmentation challenge dataset, and it achieved superior performance compared to latest approaches.

Owing to the growing proliferation of machine type communication devices and other high‐end devices in conventional human‐to‐human (H2H) communication, it is inevitable that these types of communications will co‐exist with each other in the next generation of cellular communications. This study investigates an uplink cellular network with machine‐to‐machine (M2M) and H2H co‐existence, where a machine type communication gateway is deployed as a relay in the cellular network to forward the M2M messages to the base station (BS). Non‐orthogonal multiple access (NOMA) has been adopted to transmit the data of H2H and M2M communications to the BS simultaneously. Considering the different delay‐sensitive transmissions of M2M/H2H communications, the expressions for outage probability and effective capacity (EC) are theoretically derived with the constraints of quality‐of‐service requirements. Simulation results show that the NOMA scheme outperforms the orthogonal multiple access in terms of outage probability and EC.

Automatic lesion segmentation on conventional magnetic resonance imaging is an essential component in disease diagnosis, assessment, and follow-up. Recently, extensive deep neural networks have been designed for automatic lesion segmentation. However, these approaches are not easy to further optimize owing to the poor interpretability. In this paper, we present a novel cross attention densely-connected network(CA-DCN) for multiple sclerosis lesion segmentation, which integrates attention mechanism into the encoder-decoder architecture. Aiming for further improving the performance of the model, we propose a comprehensive cross attention mechanism module by combining the characteristics of spatial and channel domains. Our method is evaluated on the public International Symposium on Biomedical Imaging (ISBI) 2015 multiple sclerosis segmentation challenge. At the time of submission, our method was amongst the top performing solution.

Hypertension is one of the most important contributors to heart disease and stroke. Multimodality medical image fusion plays an important role in the precise diagnosis, treatment planning and follow-up studies of various diseases. In this paper, we propose an image fusion framework in patients with hypertension, which is based on Frei-Chen operators in generalized reisz-wavelet transform (GRWT) domain. The proposed method is tested on two cases of MRI/CT and MRI/SPECT images. The input medical images are first transformed by GRWT into basic and detail components. Further, they are fused respectively by Frei-Chen operators which extracts ripples, edges, lines and points well. Finally, the fused image is constructed by the inverse GRWT and evaluated by indicators such as average gradient and spatial frequency etc. The visual and quantitative evaluation of the results demonstrated the superior performance of the proposed image fusion method which assists doctors for precise and sufficient diagnosis.

Hypertension has become a health problem that seriously endangers human life and is the leading cause of cardiovascular disease. Many patients do not know exactly whether their blood pressure is well controlled or not, which makes their conditions worse. A context-aware intelligent system can help patients to analyse their control situation of blood pressure (BP) and provide feedback. It is especially important to determine whether the risk-factors input in the context-aware system of hypertension is appropriate. The choice of risk factors will affect the classification performance and accuracy of the system. The risk factors screening method for hypertension proposed in this paper combined the random forest algorithm and stability selection (RFSS). It can remove the redundant context information, and leave the key factors of BP control situation. Experimental results showed that the prediction accuracy achieved more than 77% prediction accuracy, and dimension of risk factors reduced by 59%. The results indicated that RFSS is an effective method in the screening of risk factors and the prediction of hypertension.

In this paper, we consider a cluster-to-cluster (C2C) multicast scenario where multiple transmitters in a sending cluster cooperatively transmit a common known data packet to multiple receivers of the receiving cluster. The maximization of transmission rate is firstly formulated as a max-min fairness problem given the set of receivers, and an iterative transmit beamforming optimization algorithm is proposed to obtain the optimal transmit beamforming vector. Furthermore, in order to maximize the C2C system throughput, a joint optimization algorithm of transmit beamforming and receiver selection is proposed to search the optimal set of receivers based on the above transmit beamforming algorithm. The simulation results prove that the proposed algorithm has achieved a higher throughput than the conventional schemes in a lower complexity.

For sparse code multiple access advanced (SCMAA), the quality of initial information on each resource node and the convergence reliability of the detected user in each decision process were unsatisfactory at the message passing algorithm (MPA) receiver. Driven by these problems, this paper proposes a nonuniform code multiple access (NCMA) scheme. In the codebook design of NCMA, different transmitted layers are generated from different complex multidimension constellations, respectively, and a novel basic complex multidimension constellation design is proposed to increase the minimum intrapartition distance. Then a novel criterion of permutation set is proposed to maximize the sum of distances between interfering dimensions of transmitted codewords multiplexed on any resource node, where the number of nonzero elements of transmitted codewords is more than 1. On the other side, an advanced MPA receiver is proposed to improve the reliability of detection on each transmitted layer of NCMA. Simulation results show that the block error rate performance of NCMA outperforms SCMAA and sparse code multiple access (SCMA) under the same spectral efficiency.

This paper investigates a power allocation problem for maximizing energy efficiency (EE) in downlink Coordinated Multi-Point (CoMP) systems with non- orthogonal multiple access (NOMA). First, users' achievable data rate and network throughput are analysed under three transmission schemes: 1) all users' signals are jointly transmitted by coordinated base stations (BSs); 2) only cell-edge users' signals are jointly transmitted by coordinated BSs; 3) each user's signals are transmitted by only one BS. Next, we formulate EE maximization problems for the three schemes under the constraints of minimum users' data rate and maximum BS transmit power. The considered problem is non-convex and hard to tackle. To address it, an iterative sub-optimal algorithm is proposed by adopting fractional programming and difference of convex programming. Numerical results show that the near optimality performance of EE can be achieved by using the proposed algorithm with advantages of fast convergence and low complexity. Three transmission schemes of NOMA have superior EE performance compared with conventional orthogonal multiple access scheme in the same CoMP networks.

In this paper, we propose a dynamic resource allocation with quality-of-service (Qos) guarantee for clustered machine-to-machine (M2M) communications, which provides higher throughput as well as higher resource efficiency and lower access delay in a co-existing environment of both delay-sensitive and delay-tolerant services. We develop the dynamic resource allocation into four cases. In each case, the total throughput is improved under the condition of satisfying the access delay constrain for each cluster by two folds. First, a reasonable resource tradeoff between physical random access channel (PRACH) and physical uplink shared channel (PUSCH) is achieved. Second, the PRACH resource are separated for each cluster, and the access class barring (ACB) schemes separately control the number of devices belonging to different clusters. The case whose the total throughput is maximum is selected from the four cases as the optimal resource allocation scheme. Simulation results demonstrate our scheme can significantly improve the total throughput as well as the resource efficiency and reduce the access delay for each cluster.

most cancers at early stages show no obvious symptoms and curative treatment is not an option any more when cancer is diagnosed. Therefore, making accurate predictions for the risk of early cancer has become urgently necessary in the field of medicine. In this paper, our purpose is to fully utilize real-world routine physical examination data to analyze the most discriminative features of cancer based on ReliefF algorithm and generate early risk predictive model of cancer taking advantage of three machine learning (ML) algorithms. We use physical examination data with a return visit followed 1 month later derived from CiMing Health Checkup Center. The ReliefF algorithm selects the top 30 features written as Sub(30) based on weight value from our data collections consisting of 34 features and 2300 candidates. The 4-layer (2 hidden layers) deep neutral network (DNN) based on B-P algorithm, the support machine vector with the linear kernel and decision tree CART are proposed for predicting the risk of cancer by 5-fold cross validation. We implement these criteria such as predictive accuracy, AUC-ROC, sensitivity and specificity to identify the discriminative ability of three proposed method for cancer. The results show that compared with the other two methods, SVM obtains higher AUC and specificity of 0.926 and 95.27%, respectively. The superior predictive accuracy (86%) is achieved by DNN. Moreover, the fuzzy interval of threshold in DNN is proposed and the sensitivity, specificity and accuracy of DNN is 90.20%, 94.22% and 93.22%, respectively, using the revised threshold interval. The research indicates that the application of ML methods together with risk feature selection based on real-world routine physical examination data is meaningful and promising in the area of cancer prediction.

Spatially coupling data transmission (SCDT) is a multiple access technique. Since both SCDT and low density parity-check (LDPC) codes can be represented by a factor graph, a joint sparse graph (JSG) including the single graphs of SCDT and LDPC codes is constructed. Based on the JSG, the joint demodulation and decoding (JDD) by applying belief propagation (BP) algorithm in the parallel schedule is performed, but its convergence rate is slow. In order to accelerate convergence rate, a new serial schedule is proposed. The extrinsic information transfer (EXIT) charts of iterative JDD are investigated for the two schedules and employed to evaluate their converge behaviour. EXIT analysis and simulation results demonstrate that about half iteration number can be saved at the cost of marginal system performance loss. Furthermore, compared with separate demodulation and decoding, turbo-structured JDD and the uncoupling structured JDD, the JDD based on JSG for SCDT by utilizing serial schedule achieves the best performance.

Motivated by improving the number of successful access for machine-to-machine (M2M) communications, and realizing resource allocation for any number of active M2M UEs with less computation, a computationally efficient adaptive resource allocation scheme is given in this paper. eNB first learns the resource allocation for few active M2M UEs, which is adaptively changed with the number of uplink resource blocks. Then the resource allocation for other M2M UEs is derived from those active M2M UEs. In addition, the access barring scheme is incorporated to improve the number of successful access. Experimental simulations demonstrate our scheme can significantly improve the number of successful access and reduce resource allocation computation.

On dispersing different base matrices combined with masking, different kinds of quasi-cyclic low-density parity check (QC-LDPC) codes are constructed, which may achieve good error performances. Two new algebraic constructions of regular QC-LDPC codes are presented in this paper. The two constructions are based on two base matrices by dispersing and masking. The two base matrices are based on the generators of the cyclic group of a prime field. Results show that the two classes of codes presented in this paper perform well with iterative decoding over AWGN channel and have advantages over MacKay codes in some aspects of code performances. We find that one of the constructed codes converges very fast with iterative decoding, which is a critical property in ultra-high throughput communication systems.

Chronic diseases are gradually becoming the principal factors of harm to people's health. Fortunately, the development of e-health provides a novel thought for chronic disease prevention and treatment. This paper focuses on the research of cardiovascular disease (CVDs) prevention and early warning techniques using e-health and data mining. In this paper, we will use weighted associative classification algorithm to model the data in healthcare database to determine the level of cardiovascular risk. Besides, on the basis of data mining and knowledge discovery, intelligent warning mechanisms are proposed to provide different services to patients with different levels of risk. The experimental results show that the used classification algorithm is a more effective mining algorithm in the field of healthcare with higher accuracy and better comprehension. Our study is of definite significance to help control risk level of CVDs patients.

This paper presents a healthcare monitoring architecture coupled with a wireless sensor network and wearable sensor systems which monitor chronic patients in nursing house or the elderly in their home. With this architecture, we investigate how sensor nodes are deployed in the three-dimension (3D) monitoring region to achieve uniform distribution, which directly determines the Quality of Service (QoS). Based on the existing two-dimension (2D) coverage-enhancing algorithms for wireless sensor networks, a 3D sensing model and a coverage optimization algorithm are proposed in this paper. Firstly, an intelligent optimization algorithm is utilized to adjust the position of sensor nodes. Then, we pick out the redundant nodes with the set coverage algorithm, and move them into the uncovered area to increase the coverage ratio. The simulation results show that the coverage ratio increased by the coverage optimization algorithm compared with the other coverage algorithms.

This paper focuses on the optimization of energy efficiency (EE) for BSs with distributed multiple-input multiple-output (MIMO) using an efficient antenna sleep strategy. A novel metric called sleep EE gain is proposed in this paper to evaluate the energy efficiency performance of sleep strategy in distributed MIMO systems. A related closed-form expression is developed to determine the optimum antenna sleep period and to optimize the proposed metric. Based on our analysis, we propose an efficient sleep strategy for distributed MIMO systems. Simulation results show that the proposed strategy can significantly improve the overall sleep energy efficiency gain of distributed MIMO systems.

Energy is one of the crucial factors in wireless sensor network. In this paper, we consider energy fairness problem in heterogeneous wireless sensor network. Firstly, we propose a nodes' deployment model in which we deploy heterogeneous sensor nodes judiciously to conserve energy. This prevents the problem of occurrence of 'energy holes' and ensures prolonged network lifetime. Then, we use Game Theory to model the packet transmission among sensor nodes, and get the Nash equilibrium by properly designing the measurement function. Finally, the simulation results fully prove the effectiveness of the proposed scheme in this paper, which not only make the energy consumption of the network balanced but the network lifetime also improved.

This paper focuses on linear precoding in a cognitive radio (CR) multi-input multi-output (MIMO) network, where a secondary broadcast channel shares the same spectrum with a licensed primary user (PU). Considering cooperative feedback from PU to the secondary transmitter (ST), a linear precoding scheme is proposed to minimize the expected mean square error (MSE) and null out the interference to the PU. The proposed scheme is an improved approach, which is robust to the channel uncertainties caused by quantization errors and the lack of channel quality information (CQI). Furthermore, interference power control (IPC) is applied at the ST to adjust the transmit power under the tolerable interference threshold of the PU. Simulation results show the effectiveness of the proposed scheme compared with the zero-forcing (ZF) and the conventional minimum mean square error (C-MMSE) criteria.

This paper focuses on the optimization of coordinated spectrum sensing for cognitive radio (CR) networks using an efficient cognitive terminal (CT) assignment strategy. A metric called ineffective transmission-effective transmission ratio (ITETR) is proposed to evaluate the performance of the CT assignment strategy. Considering the diversity of channel utilization and CT sensing performance, the optimization of the proposed metric is formulated as an assignment problem. We provide a mathematical analysis as well as the related closed-form solution in the simplified case when all CTs have the same sensing performance for the same channel. Based on this analysis, an improved greedy algorithm-based strategy is proposed to determine the sub-optimal solution for the more general case, i.e., each CT exhibits different sensing performance for the same channel. Simulation results show that the proposed strategy can significantly improve the ITETR of CR networks when different channels possess different utilizations. When all channels have the same utilization, the proposed strategy yields a perfect approximation to the iterative Hungarian algorithm-based strategy while offering lower computation complexity.

Cognitive spectrum sensing is a key technology to improve the spectrum utilization and meanwhile decrease the possibility of interfering the primary users(PUs). Based on the cooperation of several cognitive radio sensors(CRSs), the channel fading and hidden terminal problems are avoided. The performance of detection has great relevance with the algorithm of combining the data from different CRSs. Soft fusion scheme is proved to perform better than hard fusion scheme in many scenarios[1]. The log-likelihood ratio(LLR) based soft fusion scheme is widely studied these years, but the complex or unavailable form of the distribution of the data at fusion center(FC) largely limits the optimization of the cognitive system. This paper proposes a new soft fusion scheme based on LLR which simplifies the analysis of the data at FC, and presents the exact closed-form expressions for the distribution of the data at FC, which turn out to be the Gaussian Distribution. The closed-form expressions for the probabilities of miss detection and false alarm derived are analytically tractable, therefore they efficiently guarantee the global optimality with low computational complexity. Simulation results verify the effectiveness of the proposed data fusion scheme and the validity of the expressions.

This paper addresses relay selection(RS) and power allocation(PA) issues for a physical-layer network coding (PNC) relay based secondary user (SU) communication in cognitive radio networks, in which two secondary transceiver nodes exchange their information with the assistance of a relay under interference power threshold (IPT) constraints. We propose an optimal energy-efficient RS and PA (OE-RS-PA) scheme to minimize total energy consumption per bit with the sum rate constraint and IPT constraints. A closed-form solution for optimal allocation of transmit power among the SU nodes and relay node are then derived analytically and confirmed by numerical results. Furthermore, numerical simulations and comparisons are presented to illustrate the performance of the proposed scheme.

In the intelligent monitoring of the hypertensive patients, it's necessary to assess their treatment effect and give corresponding diagnostic feedback automatically. This paper proposed a hybrid decision support system (DSS) combining several data mining techniques using an improved weighted majority voting scheme (iWMV). The mass health data of hypertensive patients were used as data source of the data mining techniques, and iWMV was used to produce a proper final judgement on patients' control condition on the basis of the individual classifier results. The proposed system was trained and evaluated using data from 167 hypertensive patients. Performance analysis showed that the hybrid system could reach classification rate (CR) of 95.34% and kappa coefficient (KC) of 92.54%, much better than systems with a single classification algorithm or combining using the simple weighted majority voting scheme (WMV). Moreover, the proposed DSS showed high stability.

Possessing dependable communications throughout large-scale disaster circumstances is of principal significance in today's world. Especially, propagation of vital information to the survivors and the rescue groups is an essential part of crisis communications. Given that the normal communication channels generally turn into non-functional in large-scale disasters, a provisional communication is required. Thus, relay networks are proper for such situations where forward stations can be suitably situated within the disaster district to increase the coverage of the stations and enhance communication dependability. Recently, the performance analysis of the wireless communication systems that deteriorated by fading, and noise has received great attention. We propose square quadrature amplitude modulation (S-QAM) for consumed power enhancement and Cooperative communication for Ad hoc and sensor networks (WASN) to combat Rician fading. The proposed modulation technique in co-operative communication situation is evaluated by an analytical approach and simulations. The performance analysis with different Rician k-factor and diversity order with Matlab simulation is done. We observe bit error rate gain performance, in which there is a considerable gain in performance accomplished by increasing diversity order L, while the gain in performance that achieved by the increase in k is extremely small without diversity order. We conclude that using of cooperating communication can enhance the Rician fading in disaster situation and helps to save human lives.

Recently, the R&D and applications of M2M systems are booming in China, especially after it is written in the 2010 Government Work Report. In this paper, the R&D works of wireless eHealth (WeHealth) are overviewed, the concept of which was proposed by our group in 2005. Some key techniques of WeHealth system are discussed, and some practices based on the concept of WeHealth are introduced. Besides, a recent WeHealth pilot trial on chronic disease monitoring is also introduced, which is reported as “The First Wireless Healthcare Chronic Disease Monitoring Project in China Based on Internet of Things Technology”. There are 30 community hospitals up to now applying our WeHealth blood pressure monitoring system for chronic disease management. The practical data clearly demonstrate the effectiveness of our WeHealth system in hypertension disease control.

With the unprecedented aging of population, chronic disease become a serious problem in modern medical area. Recent advances of wireless technology make the health monitoring in home to be more convenient for the chronic disease. Reducing the influence of the interference from the other wireless equipment is one of the most important problems in wireless physical parameter monitoring device in healthcare field, as reliable data transmitting is vital in medical care. In this paper, we propose a novel interference avoidance strategy that can greatly reduce the influence of the interference from other wireless equipment to wireless physical parameter monitoring devices in the hospital or home environment. Experimental results show that the strategy has good effect on interference avoidance and reduce the package loss rate.

Recent advances of hardware and integrated chip have made the body-implanted medical device reality. In this paper, a chronic disease management system with implanted devices based on Wireless Sensor Networks (WSN) was proposed for people with chronic illnesses. We first described the architecture of the chronic disease management system. Then through analyzing the system architecture, we explained the advantages of our new system and pointed out the main issue of the short lifetime of the implanted medical devices. Then a cooperative strategy was proposed in the particular system to decrease the energy consumption, which extended the entire lifetime of all implanted devices. Finally, we evaluated the performance of our strategy with simulations.

To improve the performance of hospital emergency department (ED), we propose a scheduling strategy to reduce the total time of medical tests in ED for a given number of patients. We model the schedulig strategy as a constraint logic programming problem, and then solve this problem with an open source software: Minizinc. Also we compare the medical test time with the scheduling strategy and that without scheduling, which is viewed as a benchmark for our proposed scheduling strategy. The results show that the scheduling strategy can save 12%–25% of medical test time.

In this paper, we detail the design of a hypertension monitoring system via telecommunication and computer technologies and propose a method for theoretically analyzing its system accuracy. In the design of our system, we add a novel decision support unit, building on a diagnosis standard in medicine, into our monitoring system. The architecture of our system, including this decision support system, is detailed in this paper. In the analysis of our system accuracy, we theoretically evaluate the accuracy of a decision support system by linking the system accuracy with the distribution of sensors' errors. Additionally, we verify our proposed analysis of system accuracy in this paper by comparing its result to that of Monte Carlo simulation.

This paper presents a novel type 2 diabetes mellitus (T2DM) monitoring system. Decisions on the statuses of diabetes control and predictions of future blood glucose of an individual made by the system depend both on the original medical data collected by medical sensors and some contexts either are entered manually or generated automatically. When dealing with the data, first we clean and transform data and contexts, then build data mining models using several mining algorithms. After the mining process, we analyze and assess the accuracy and sensitivity of the models, and find out the appropriate models for decision making and predicting in diabetes control. Our study is of certain significance to help prevent and treat T2DM.

In this paper we proposed a novel cochannel interference avoidance strategy for outdoor remote medical monitor network. We employ simulated annealing to design the channel allocation algorithm, which is the core of the strategy. We implemented the strategy in the experimental we built at Beijing outer suburb and took a series of experiments to examine the effect of the strategy. The experimental results showed that the interference level in the network was well controlled after the strategy is used.

In this paper, we propose a novel method to estimate the accuracy of a decision support system for hypertension monitoring. The decision support system is designed building on a diagnosis standard in medicine, and one decision made by this system depends on both the blood pressure data gathered by medical sensors and some contexts that are manually entered by clinicians. When analyzing the system accuracy, we take into account both the potential sensors' errors and the errors of entering context. In addition, we propose a novel method for the estimation of system accuracy; this method is motivated by the fact that the traditional method to estimate system accuracy would overestimate the system accuracy (the details would be presented in Section III.C.). Finally, we compare the system accuracy estimated by the proposed method and that estimated by the traditional method. Our study shows that our proposed method can well estimate the system accuracy.

During the last decades, there are numerous researches in the field of wireless local area networks (WLANs). A large number of these researches focus on performance analysis of IEEE 802.11 Medium Access Control (MAC) protocols. And the most representative model for MAC mechanism in WLANs is the two-dimensional Markov model, which has been proven to be an accurate model to analyze the performance of WLANs, in the assumption of finite number of terminals and ideal channel conditions. But this two-dimensional Markov model is rather complex, as far as computation and analysis are concerned. Recently, a novel semi-Markov process model for IEEE 802.11 MAC protocols has been put forward to mitigate the complexity of analyzing the performance of WLANs. However, this semi-Markov model has some defects, such as not taking maximum retransmission number into account. In this paper, we propose an advanced semi-Markov process (ASMP) model, which not only withholds the merits of semi-Markov chain model, but also eliminates its shortcomings, thus more accurately calculating the network parameters of WLANs. Simulation results show that our proposed model achieves accurate results with less complexity, and is suitable for evaluating the performance of the MAC protocols.

This paper presents an innovative interface design and its field programmable gate array (FPGA) implementation for transparent transmission between medium access control (MAC) layer and physical (PHY) layer. In the interface design, we introduce a mechanism of central process unit (CPU) and FPGA co-processing data so as to obtain good performance compared with conventional interface architecture. In addition, enhanced memory architecture, including memory separating and co-operating among different types of memory, is also taken into consideration to achieve this goal. The design uses the intellectual property (IP) core technology on processor local bus (PLB) with high throughput, low latency and low complexity. Based on a wireless transmission system with Gigabit (Gbps) throughput, the interface prototype has been implemented on Xilinx Virtex5 FX130T FPGA. The implementation results show that the interface logic can achieve a peak throughput up to 3.93 Gbps and on average to 1.02 Gbps.

In the spectrum sharing systems, an unlicensed (secondary) user may share a frequency band with its licensed (primary) user as long as its transmission does not interfere with the primary user's communications. For the cognitive wireless relay networks, this interference regulation from the primary user may affect its cooperative scheme. In this paper, we investigate the outage performance of cognitive wireless relay networks in a spectrum sharing environment, in which the secondary users including the source and relays may take advantage of a frequency band of the primary user to transmit data to the receiver while not interfering with the primary user's communication. In particular, we quantify the relation between the outage performance of the secondary relay link and the interference inflicted on the primary user.

Distributed multiple-input multiple-output (DMIMO) system is a combination of distributed antenna systems (DAS) and co-located antenna systems (CAS). Power allocation between different antennas is a very important aspect in both DAS and CAS. In this paper, we investigate the downlink power allocation scheme in D-MIMO systems based on minimum symbol error rate (SER) criteria. We propose a power allocation scheme which can effectively approximate the optimal allocation scheme. In the proposed scheme, both Large-scale fading and transceiver antenna correlation are taken into consideration. Moreover, mobile terminal (MT) only needs to feedback largescale fading information and the rank of transmit antenna correlation matrix periodically, which significantly reduces the feedback overhead. Simulation results verify the capacity improvement based on the proposed power allocation scheme.

In many practical systems, perfect channel state information (CSI) is difficult to obtain at the base station (BS), we usually use limited feedback CSI to reduce the symbol error rate (SER). To analyze the SER performance of different transmit distributed antenna (DA) subset in limited feedback distributed antenna systems (DAS), the closed-form SER expressions for both binary phase shift keying (BPSK) and M-phase shift keying (MPSK) are derived in this paper. From these expressions, we find there exists an appropriate transmit DA subset which can obtain the best SER performance. Then, an adaptive transmit DA selection scheme with the minimum SER is proposed. Simulation results show that the proposed scheme can achieve SER performance gain than blanket transmission scheme and selection diversity scheme.

Distributed localization algorithm continues to be an important and challenging topic in today's wireless sensor networks (WSNs). The accurate estimations of distances among nodes are premises for the accurate estimations of node positions. In this paper, we propose some schemes towards the DV-Hop algorithm to improve the distance estimations. These improvements are based on accurate analysis of hop information of neighbor nodes in a WSN with randomly deployed sensors. It is demonstrated that our proposals ensure better distances estimations and hence improve the localization accuracy with low communication overheads. In addition, we discuss the deployment, communication cost and computing complexity of the novel algorithms. Simulations are performed and the results are observed to be in good agreement with the analysis.

In this paper, the pilot design aiming at mitigating the inter-cell interference in MIMO OFDM systems is addressed. We first compute the mean square error of the frequency domain LS joint channel estimation with the consideration of interfering cell. We then overview the optimal pilot sequences within a cell, which can cancel the intra-cell multiple access interference. Based on the mean square error and variation coefficient, we propose two conditions for the pilot design to mitigate the intercell interference. It is shown that if an ideal channel between interferer and receiver is assumed, we can obtain the pilots resulting in a smaller value of mean square error based on the variation coefficient of zero, which make the power of inter-cell interference distribute at all paths of the channel more uniformly and less, therefore, the performance on system level would be improved. A further discussion about proposed guidelines in practical system is made, and numerical calculations as well as simulation results are presented to verify the theoretical analysis.

In this paper, we propose a relay selection scheme based on cooperative MIMO (multi-input-multi-output) communication for network lifetime maximization in energy-constrained sensor networks. Compared with existing work, our distributions are: cooperative nodes choice is not based on minimizing energy consumption per communication but balancing remaining energy between participants including source and cooperators based on one special application. We validate the efficiency of the proposed algorithm by simulation by comparing it with the direct communication and minimizing energy consumption communication schemes.

In wireless sensor networks (WSN), the nodes have limited energy resource. So power conservation in individual node is of significant importance. In this paper we derive a cross-layer design method that combines adaptive modulation and coding (AMC) at the physical layer and two sleep modes according to monitoring type at the medium access layer (MAC) in healthcare monitoring WSN, judiciously to maximize spectral efficiency, and to minimize energy consumption, under prescribed error performance constraints. To achieve maximum spectral efficiency in the entire SNR (signal-to-noise ratio) range, we advise a combined transmission mode switching between convolutionally code and unconvolutionally code mode, based which we analyze the energy consumption of the system with multi mobile nodes and a sink node and verify that it is energy efficient to adopt AMC instead of single modulation at physical layer in healthcare monitoring WSN.

Orthogonal frequency division multiplexing (OFDM) is robust against frequency selective fading due to the increase of the symbol duration. As the symbol duration increases, the OFDM systems become more susceptible to time-variations. Users moving at a quite high speed suffer the inter-carrier interference caused by the time varying (TV) channels. Meanwhile, users at the cell boundary are known to experience a large inter-cell interference in a fully-loaded OFDM cellular environment with a frequency reuse factor equal to 1. In this paper, the design of pilots in the downlink at the cell boundary is addressed, aiming at mitigating the inter-cell interference under TV channel. The properties of these pilots are described, and the simulation results are presented.

This paper presents a high speed decoder architecture for irregular structured low density parity check (LDPC) codes and its field programmable gate array (FPGA) implementation. Algorithm transformation and architectural level optimizations are employed to reduce the critical path. The enhanced semi-parallel architecture is easily scalable and reconfigurable for larger block sizes and can be well suited for achieving high decoding throughput. Based on the proposed architecture, a (10240, 5120) irregular structured LDPC decoder is implemented on Xilinx FPGA Virtex-4 VLX80, the FPGA implementation results show that the irregular LDPC decoder can achieve a maximum (information data) decoding throughput of 223 Mbps at 18 iterations.

Orthogonal frequency division multiplexing (OFDM) systems encounter performance degradations due to the time varying (TV) channels in wireless environments. As delay spread increases, symbol duration should also increase, and then OFDM systems become more susceptible to time-variations. Time-variations introduce inter-carrier interference (ICI), which must be mitigated to improve the performance in high delay and Doppler spread environments. In this paper, a pilot design scheme for multiple input multiple output (MIMO) OFDM system is proposed under TV channels. The properties of these pilots are described, and the simulation results are presented.

Multiple input multiple output (MIMO) systems provide dramatic capacity gain through an increased spatial dimension. However, the capacity gain is reduced if the channel state information (CSI) is not perfect. The more exact the CSI is, the higher capacity gain we can obtain. In order to obtain higher capacity gain, this paper analyzes the two factors which mainly affect the reliability of CSI: the quality of channel estimation and the feedback delay, and proposes a channel modification strategy, which can optimally reduce the effect of the two factors and obtain more exact CSI. Simulation shows that the channel modification strategy is effective.

The traditional channel estimation for orthogonal frequency division multiplexing (OFDM) systems over fast-varying fading channels is usually carried out in two steps. Firstly obtain the least-square (LS) estimate over the pilot sub-carriers, and then interpolate it over the entire frequency-domain. In this paper, we propose a high precision channel estimation method by adding an intermediate step, which is based on the strong correlation in each path of the channel during the coherent time and can distinguish valid path taps and noise taps effectively, to improve the accuracy of the preliminary estimate over the pilot sub-carriers. The simulation results in the frequency band of 2.4 GHz show that the proposed method can obtain refined channel functions more efficiently and achieve a good bit error rate (BER) performance close to the theoretical bound of ideal channel estimation.

Orthogonal frequency division multiplexing (OFDM) systems encounter performance degradations due to the time varying (TV) channels in wireless environments. As delay spread increases, symbol duration should also increase, and then OFDM systems become more susceptible to time-variations. Time-variations introduce inter-carrier interference (ICI), which must be mitigated to improve the performance in high delay and Doppler spread environments. In this paper, we propose a method to estimate the TV channel. The method is designed for the comb pilot pattern. It performs a multi- symbol processing of the LS estimate over the pilot sub- carriers to detect the active paths of the channel and uses a piece-wise linear model to approximate the time-variations characteristic of each active path during each OFDM symbol to mitigate ICI. Theoretical analysis and simulation result show performance improvement in high delay and Doppler spread environments.

In this paper, we address the problem of the design of pilots aiming at mitigating the inter-cell interference in OFDM systems. Based on the conventional pilot design used in the intra-cell, we design the pilots which can make the power of inter-cell interference distribute in all paths of the channel uniformly, and result in a variation coefficient of zero, then the better performance on system level will be derived. The properties of these pilots are described, and the simulation results are presented.

We consider superimposing pilot signals onto data signals for channel estimation in multiple-input multiple-output (MIMO)-orthogonal frequency division multiplexing (OFDM) systems. This paper presents a channel estimation method, which is based on orthogonal pilot-constant amplitude zero autocorrelation code (CAZAC) for space-time block code (STBC) OFDM system. In this system, the proposed method has low-complexity and CAZAC codes have the minimum peak-to-average power ratio (PAPR). At the same time, the superimposed pilot approach can increase bandwidth efficiency and data rate compared with other classical approaches, where pilot signals and data signals are disjoined, at nearly the same system performance by the computer simulation.

A novel preamble sequence design for random access using complete complementary code sets in multi-input multi-output orthogonal frequency division multiplexing (MIMO-OFDM) scheme is proposed. By elaborate design, both the optimum cross-correlation function (CCF) property and sufficient available number of preamble sequences can be achieved. The generation method and the bound of extension code number are addressed to satisfy the requirement of practical system. In addition, the comparison between proposed code and constant amplitude zero auto-correlation (CAZAC) code which is promising in the long term evolution of 3rd generation systems as preamble sequences was presented. The numerical results show that the application of preamble sequences with complete complementary code sets can obtain high detection probability.

In this paper, we first overview the least square (LS) space-time joint channel estimation in filtered multi-tone (FMT) system in multi-input multi-output (MIMO) system. Then the design of pilots aiming at mitigating the inter-cell interference in MIMO FMT systems is addressed. It is shown that if the pilots which result in a zero variation coefficient of the estimation error are used, the better performance on system level was obtained. The properties of these pilots are described, and the simulation results are presented.

In this paper, we investigate some space-time joint channel estimation (JCE) schemes in filtered multitone (FMT) based multicarrier multibranch (multiuser and/or multiantenna) systems. Each subcarrier is proposed to be processed separately due to the spectral characteristics of the FMT signals On each subcarrier, the wireless channels experienced by all transmit branches are estimated jointly at the receiver. The least square (LS) space-time JCE and weighted least square (WLS) schemes are derived. The constant amplitude zero auto-correlation (CAZAC) code based pilots are proposed for the aforementioned multicarrier applications to reduce the mean square estimation error. Simulation results are provided to validate the theoretical analysis.

This paper discusses some issues in designing space-time trellis codes of high complexity, i.e., when the total number of candidate codes is large or the total number of pairwise error events is large. By adding windows in the search domain and combining the equal states in error state trellis, the complexity of search is greatly reduced. And at the same time, we can still find some codes which have very good performance. We present some new codes of high complexity at the end of this paper and show that our codes have better performance than many codes before.