Search

Matrix calculations are often required for the analysis of any big-data cloud computing system. It is quite common to process big-data associated matrices possessing the sparsity and low-rank properties. In order to efficiently deal with big-data matrices, we propose a new divide-and-iterate framework, which can be invoked to solve an enormously large linear system of equations by taking advantage of factored matrices. The Kaczmarz algorithm (KA) is utilized here to design the parallel iterative algorithms which are capable of solving a large system of equations by iteratively updating the solution through the reduction into the factorized subsystems in parallel. The convergences of our proposed new iterative algorithms are justified by the rigorous proofs. Besides, the time- and memory-complexities are studied to demonstrate the resource efficiency of the proposed algorithms. Numerical experiments are also presented to illustrate the effectiveness of this proposed new framework.

Manufacturing firms operating in today’s competitive global markets must continuously find the appropriate manufacturing scheme and strategies to effectively meet customer needs for various types of quality of merchandise under the constraints of short order lead-time and limited in-house capacity. Inspired by the offering of a decision-making model to aid smooth manufacturers’ operations, this study builds an analytical model to expose the influence of the outsourcing of common parts, postponement policies, overtime options, and random scrapped items on the optimal replenishment decision and various crucial system performance indices of the multiproduct problem. A two-stage fabrication scheme is presented to handle the products’ commonality and the uptime-reduced strategies to satisfy the short amount of time before the due dates of customers’ orders. A screening process helps identify and remove faulty items to ensure the finished lot’s anticipated quality. Mathematical derivation assists us in finding the manufacturing relevant total cost function. The differential calculus helps optimize the cost function and determine the optimal stock-replenishing rotation cycle policy. Lastly, a simulated numerical illustration helps validate our research result’s applicability and demonstrate the model’s capability to disclose the crucial managerial insights and facilitate manufacturing-relevant decision making.

Indoor localization of human objects has many important applications nowadays. Here, we propose a new multiresolution device-free indoor localization scheme using wireless radio frequency (RF) fingerprints. In our proposed new device-free approach, all transceiver devices are fixed in an indoor environment, so human targets do not need to carry any transceiver device with them. First, the indoor geometry is divided into several zones, and then, the received signal strength indicators measured by the receiving antennas are input features to our designed innovative machine learning model to identify within which zone the target is. Our proposed machine learning model, i.e., a multiresolution random forest classifier, is composed of a cascade architecture, which integrates and distills learned results over various zoning resolutions. The proposed new multiresolution approach greatly outperforms the existing random forest classifier. The average Euclidean distance error resulting from our proposed new technique is 1.0376 m.

Autonomous vehicles become more and more popular in our daily life. Mobile computing schemes to be installed on these vehicles have drawn a lot of recent research interest. In this paper, we address the important path-planning problem for autonomous vehicles. We introduce and formulate the novel paintbrush coverage problem . We present a theoretical study on the minimum trajectory length of a paintbrush to cover an arbitrary convex region, which is derived as a function of the area of the region and the size of the cover. Three commonly-used patrolling/scouting methods, namely boustrophedon, spiral, and sector, are manifested in details as the potential solutions to the paintbrush coverage problem. The theoretical minimum trajectory lengths any algorithm can achieve are also demonstrated as the benchmarks for different shapes of regions.

In this paper, we propose a novel spectrum-efficiency (SE) optimization scheme for amplify-and-forward (AF) cooperative relaying systems based on non-orthogonal multiple-access (NOMA). In this network, a weighted-sum-minimum mean-squared-error (WSMMSE) method is utilized to transform the original non-convex optimization problem into a tractable optimization problem. Then, the relay beamforming-matrix (BM) and the power-allocation (PA) coefficients are alternately optimized by use of the iterative beamforming method and the Lagrange-multiplier (LM) method subject to the quality-of-service requirements. Monte Carlo simulation results verify that our proposed new scheme can attain better performance than the optimal PA-NOMA technique and the conventional orthogonal multiple-access (OMA) technique.

Visible-light communications (VLCs) are energy-efficient communication means and a popular alternative to the radio-frequency technology. VLCs are referred to as free-space optical communication systems which can be built upon the existing lighting infrastructure. Due to the cost-effectiveness concern, light emitting diodes (LEDs) are adopted as the visible-light sources. In this paper, we demonstrate a VLC prototype using white LEDs and pulse position modulation (PPM), which is aimed for short-range communications. Based on the acquired VLC signal waveforms, characteristics related to electronic components, LEDs, and photo diodes (PDs) are studied thoroughly. The non-square shape of received pulses is modeled by new template functions involving rising and setting time intervals. A new PPM demodulator is devised thereupon. Meanwhile, the probability density function of noise is derived using the Edgeworth expansion. As all of the aforementioned aspects are taken into account, the VLC quality-of-service (QoS) in terms of package error rate and throughput is studied in this work. Experiments are taken to justify the theoretical QoS metric and evaluate our proposed new PPM demodulator.

A new modulation scheme using multiwavelets for downlink non-orthogonal multiple-access (NOMA) transceivers is presented in this work. Multiwavelets leading to remarkable spectral diversity are exploited to modulate/demodulate the superimposed multi-user signals in downlink NOMA systems, where the discrete multiwavelet transform and the inverse discrete multiwavelet transform are invoked in the new demodulator of the user equipment and the new modulator of the base station, respectively. The performance evaluation and the computational-complexity analysis of our proposed new NOMA scheme are also conducted. Simulation results demonstrate that the proposed new scheme can significantly increase the system capacity while dramatically suppressing the peak-to-average-power ratio (PAPR) compared to the conventional OFDM-based NOMA approach. Meanwhile, our new scheme can achieve the same system capacity as the scalar-wavelet-based NOMA approach but with a lower PAPR at little extra cost of computational-complexity. All of these three aforementioned NOMA schemes can result in the identical bit-error-rate under the same signal-to-noise ratio condition.

The physical layer security (PLS) has become crucial in the fifth generation (5G) wireless technology where non-orthogonal multiple-access (NOMA) plays an essential role for sharing the spectrum among multiple users. Multiple-input multiple-output (MIMO)-NOMA based cognitive radio network (CRN) enhances the spectral-efficiency at the cost of PLS. In this paper, a transmit-zero-forcing beamforming (ZFBF) technique with signal alignment is proposed to improve the PLS for MIMO-NOMA based CRN when perfect channel state information (CSI) is available at the base station (BS). In this study, a generalized cellular model with multi-cell is considered where each cluster is formed with a primary user (PU) and a secondary user (SU) to adopt the CRN. The locations of PU and SU are kept arbitrary within the cells. We also propose an eigen beamforming technique to improve the PLS for MIMO-NOMA based CRN when the imperfect CSI is available at the BS. For both beamforming techniques, optimum power allocation algorithms are developed for further enhancement of the PLS. Finally, a closed-form expression of the secrecy outage probability for the Nakagami- $m$ fading channels is derived and compared with simulation results and other existing techniques.

Biometric information acquired by mobile devices has been explored for various emerging applications in recent years. One of these new applications is advanced authentication. On-line user-identification based on biometrics collected by smartphones has been drawing tremendous research interest lately. The accelerometer data are available pervasively on cellular phones and therefore there incurs no extra cost to facilitate accelerometer-based biometric systems. Individual time-series constituted by accelerometer data are quite complex and therefore impose challenges to the user-identification research. In this work, we propose a novel on-line user-identification approach using accelerometer data by exploiting new homological analysis. The inherent walking patterns of different users in the accelerometer data are monitored in the embedded phase space. Then the expected persistence diagrams (EPDs) are formed and transformed into the corresponding probability distribution functions. Thus, the discrepancy in walking patterns can be measured by the Kullback-Leibler divergence (KLD) of those ultimate distribution functions. Users can be identified according to the KLDs resulting from the sampled data and pre-stored training data. Experimental results from the real-world accelerometer data have demonstrated that the accuracy of our proposed new scheme reaches up to 94.5% and it greatly outperforms other existing methods.

It is well known that recursive least squares (RLS) filter is a prevalent adaptive filter that recursively updates the filter-coefficients to minimize a weighted cost function related to an input signal. Conventionally, such an input signal is a sequence of scaler numbers. Multichannel signals generated from one or more sources are often encountered nowadays. Therefore, there emerges a urgent demand to generalize the conventional RLS filter to accommodate multichannel signals with implicit interrelations. The key issue of this new generalization is to convert the correlation matrix involved in the conventional RLS filter for one-dimensional input to the correlation tensor required by the new RLS filter for multi-dimensional input. Due to lack of the sufficient mathematical framework of the required tensor operations, here we make the first attempt to established the new tensor recursive least squares filter (TRLS) theory based on our newly derived Woodbury tensor identity. We apply the Woodbury tensor identity to design a recursive algorithm for updating the weight-tensor. Furthermore, the performance analysis of the proposed new TRLS algorithm, including mean-squared deviation and misadjustment, are facilitated using the established tensor calculus theory. Numerical experiments are also presented to investigate the convergence speed, mean-squared deviation, and misadjustment for input signals with various statistical characteristics and different dimensions.

This letter introduces a novel joint power-allocation (PA) and relay beamforming optimization (JPBO) algorithm for non-orthogonal multiple-access (NOMA) amplify-and-forward (AF) relay systems to achieve the maximum weighted sum-rate (WSR). In our proposed new framework, the WSR maximization problem is transformed into a tractable weighted-sum minimum mean-square-error (WSMMSE) problem. Then, the PA and the relay beamforming-matrix (BM) are alternately optimized by use of successive convex approximation (SCA) and Lagrange-multiplier (LM) algorithm. Simulation results demonstrate that our proposed new algorithm can significantly improve the WSR with a faster convergence than the existing method.

Mosaic is a popular approach to provide privacy of data and image. However, the existing demosaicing techniques cannot accomplish efficient perfect-reconstruction. If the receiver wants to recover the original image, the extra transmission of the original subimage to be mosaicked is necessary, which consumes much channel resource and is therefore inefficient. In this paper, we propose a novel efficient recoverable cryptographic mosaic technique by permutations. A mosaic, or a privacy-protected subimage, can be constructed through either of the three permutations (Busch’s, Wu’s, and Sun’s/Minmax). These three permutations are designed to maximize the objective function as the sum of the absolute row/column index-differences. This objective is related to the sum of the pixel-to-pixel cross-correlation by our pertinent theoretical study. To measure the effectiveness of the image-mosaicing methods, we propose two image-discrepancy measures, namely summed cross-correlation (SCC) and Kullback-Leibler divergence of discrete cosine transform (DCT-KLD). Compared to the big majority of random permutations for image-mosaicing, our proposed three permutation methods can achieve much better performances in terms of SCC. Nevertheless, the advantage of the three proposed permutation methods over random permutations is not obvious according to DCT-KLD.

For addressing the One-Dimensional Road side unit Deployment (D1RD) problem, a greedy approximate algorithm named Greedy2P3E was proposed two years ago, and its approximation ratio was proved to be at least 2/3 for the D1RD problem with EQual-radius RSUs (D1RD-EQ problem). Can better or even tight approximations for Greedy2P3E be found? In this paper, approximation ratio of Greedy2P3E is re-inspected and tight approximation ratio is found. To this end, a greedy algorithm named Greedy3P4 is first proposed and proved to have a tight approximation ratio of 3/4 for the D1RD-EQ problem. Then, by using Greedy3P4 as a bridge, 3/4 is also proved to be the tight approximation ratio of Greedy2P3E and it is tight for all $n{\geq }2$ . Comparative evaluations are performed on real cases using a real vehicle trajectory dataset. The results show that these greedy algorithms usually return near optimal solutions with a profit more than 98% of the optimal solutions, and the greedy algorithms well outperform the other typical algorithms tested.

Femtocells can extend the last-mile accessibility for the current cellular networks so as to serve more users indoors and improve the spectrum utility. However, since the indoor femtocell establishment can be ad hoc, HeNBs may also be inevitably vulnerable to the rogue-HeNB (uncoordinated or illegitimate HeNB) attack in practice. For detecting rogue-HeNBs, a two-dimensional received signal-to-interference-plus-noise ratio (SINR) map is exploited to construct a spatial radio-activity feature map (RAFM). Mathematical proof is derived to justify that the existence of any femtocell interference corresponds to a flat-bottom convex region in the RAFM. Advanced image processing techniques are designed here to localize the radio interference (rogue-HeNB) over the RAFM. Thus, a novel positioning scheme is proposed in this paper to undertake the accurate rogue-HeNB localization for the femtocell networks. Different indoor scenarios are evaluated through the real experiments for our proposed method. The average accuracy can be within one meter for a 6.75 meter by 8.1 meter laboratory.

Although the clock synchronization is of utmost importance in communications, it is quite difficult to achieve in practice. Consequently, the asynchronous effect emerges as a critical problem for all pulse-position modulation (PPM) systems. In this paper, the impact of the sampling-period misalignment is studied and evaluated. A new analytical model is established to govern the misalignment between the slot periods of the transmitted and received signals. The ultimate error rate related to this kind of unfavorable misalignment is also derived theoretically. Besides, we propose a new algorithm, which can adjust the onsets of the slot periods dynamically, to mitigate this misalignment and improve the error performance. To justify our new analysis and algorithm, a visible-light communication system using single-pulse PPM is taken as a practical example. Simulation results demonstrate that the effectiveness of our proposed algorithm is very promising. The frame-error rate of the rectified PPM signal is around 10−5 across all of the misalignment values.

Nowadays, the sufficient quality-of-service (QoS) provision for multimedia mobile applications remains a major challenge to any wireless network. Conventional sufficient QoS provision techniques using resource allocation, data scheduling, and cross-layer optimization have been proposed to tackle this problem. Nevertheless, due to the unpredictable nature of wireless channel conditions, the QoS improvements resulting from the aforementioned approaches are often unsatisfactory. In this paper, we would like to further exploit our previous concept, namely user-placement ushering (UPU). A mobile user can be relocated to an optimal, or at least a better place to boost the QoS for a given application. The QoS is measured by the packet-success rate (PSR), which can be further computed from the code-block error-rate (CBER). Novel fast algorithms are devised here to usher (guide) a mobile user to an appropriate spot with sufficient QoS, which is nearby. We address the UPU problem to accommodate the realistic complex indoor environments, where obstacles are present. Our simulation results have demonstrated that our new fast UPU algorithms are capable of finding new appropriate locations satisfying the required QoSs for different wireless network applications.The preliminary results are presented in [1] .

As a new topic attracting research efforts recently, the Wireless Power Transfer based Energy ReDistribution (WPTERD) problem is targeted for finding energy transmission schedules satisfying the nodes’ energy expectations, meanwhile achieving minimum energy loss and shortest time span. In this paper, we propose a two-step approach for WPTERD by dividing it into two sub-problems named WPTERD-Egy and WPTERD-Time, which focus on the optimization in energy loss and in time span respectively. Based on some proved properties, the WPTERD-Egy problem is transformed to a Linear Programming problem and thus an optimal time-length list of the nodes’ energy transmissions achieving minimum energy loss can be obtain easily. For that WPTERD-Time is NP-hard, we propose an approximate algorithm by exploiting the core idea of the well-known least-degree-last graph coloring algorithm, and prove its approximation ratios respectively for 2D and 3D WSNs. Combining the two steps, we construct our Graph Coloring inspired Energy-Time Decoupling (GCEgyTimeD) algorithm for WPTERD. Numerical simulations validate GCEgyTimeD’s ability in returning a schedule with minimum energy loss and with time span within 20% of the optimal value. Simulation results also show that GCEgyTimeD outperforms a representative flow-based algorithm in terms of the tested performance metrics.

Cyber security has become an important problem nowadays as almost everyone is often linked to the Internet for business and entertainment. Conventional cryptographers fail to address timely issues regarding cyber-attacks, such as cyber identity theft. In this work, we propose a novel idea, namely, a bandage-cover cryptographer (BCC), which is completely software-defined and protocol-free. Besides, this new cryptographic approach can enable camouflages to confuse data-mining robots, which are often encountered in the cyber world nowadays. Because all of the existing cryptographers aim to protect the entire data (document and file) altogether, they cannot have camouflagibility to mislead data-mining robots. Conversely, by our proposed novel BCC, one can select arbitrary contexts or parts of the data (related to individual identify and/or private confidential information) under protection. To evaluate such a first-ever cryptographer capable of misleading data-mining robots, we define two new metrics, namely: 1) vulnerability and 2) camouflage rates. The theoretical analyses of vulnerability rate and camouflage rate for our proposed new BCC are also presented in this article to demonstrate the corresponding effectiveness.

In this paper, we propose a novel hierarchical radio-frequency identification (RFID) tag-recognition method based on blind source separation (BSS), graph-based automatic modulation classification (AMC), and direct-sequence spread-spectrum (DSSS). In our proposed method, RFID tags can be modulated using different modulation schemes according to different scenarios (e.g., different users or different tag devices). For each modulation scheme, the direct-sequence spread-spectrum strategy is employed to allow simultaneous transmissions of multiple commands. In the signal separation phase, BSS is employed to separate different transmitted signals. Then in the first hierarchy of the recognition phase, different modulation types are adopted to distinguish different users, the graph-based AMC is built upon the periodicity of the modulated signals: the cyclic spectrum of the received signal is established; the graph representation is then constructed according to the cyclic spectrum. Ultimately, robust features are extracted from the graph representation. In the second hierarchy of the recognition phase, the DSSS scheme is utilized to differentiate the control or sensed data carried by individual tags; the signature sequence set with low cross-correlations can be generated from Kasami sequences. In the third hierarchy of the recognition phase, the information data are thus spread by these signature sequences. In our proposed new RFID framework, multiple tags can transmit signals simultaneously in the same frequency band where each tag signal can still be separated and identified and its carried information can be recovered. Monte Carlo simulation results demonstrate the promising performance of our proposed new RFID scheme.

A novel pulse-shaping scheme using multiwavelets (MWs) for the non-orthogonal multiple-access (NOMA) communication systems is proposed in this letter. Different from the conventional pulse-shaping schemes, this work exploits multiwavelets for constructing pulse-shaping filters so different users’ information can be multiplexed using the power diversity within the same frequency band. Hence, all users’ transmitted signals can be combined through the superposition by various power-allocation coefficients. At the proposed MW-NOMA receiver, after the successive interference cancellation (SIC) is carried out, the corresponding MW matched filters (shaping pulses) will be invoked to extract each individual user’s transmitted signal. A link-level performance evaluation of the proposed new NOMA scheme is also conducted in this letter. The simulation results demonstrate that our proposed new NOMA technique can significantly improve the system capacity and the bit error rate (BER) performance over the conventional NOMA scheme. When the power-allocation ratio is set to 0.4:0.6 in the proposed NOMA system for two users, the remarkable transmitting $E_{ \text {b}}/\mathcal {N}_{0}$ gain of 15 dB can be achieved by both users.

How to combine multiple data-streams for transmission in aggregate is a very interesting problem, especially for the emerging software-defined networks nowadays. The conventional packet-based protocols cannot provide the flexibility for combining data-streams in the ad hoc nature. If the number of data-streams changes over time, the existing packet formats cannot handle the transmission of multiple data-streams effectively. The effectiveness is measured by two performance metrics, namely coding efficiency and data-transmission intermittency. We propose a new software-defined multiplexing code (SDMC) approach, which can combine (multiplex) multiple data-streams easily and is much more effective than the conventional packet-based method. Three SDMC schemes (distributed, hierarchical, and hybrid) are compared theoretically and by simulation. A trade-off between these two performance metrics can be found when one selects one of the three SDMC schemes for combining multiple data-streams. The hierarchical SDMC scheme brings about the highest coding efficiency while the hybrid SDMC scheme suffers from the smallest overall intermittency.

By enabling energy redistribution among nodes in Wireless Sensor Networks (WSNs), the Wireless Power Transfer (WPT) technology provides a new approach to prolong life time of WSNs or a fundamental block for building perceptual WSNs. In this paper, we attempt to deal with the underlying WPT based Energy ReDistribution (WPTERD) problem, which is targeted for finding an energy transmission schedule that satisfies the nodes’ energy expectations, meanwhile achieves minimum energy loss and shortest time span. In this paper, we propose a two-step approach for WPTERD by splitting it into two sub-problems named WPTERD-Egy and WPTERD-Time, which focus on the optimization in energy loss and in time span respectively. Basing on some proved properties, we transform WPTERD-Egy to a Linear Programming problem and thus can obtain an optimal time-length list of the nodes’ energy transmissions leading to minimum energy loss. Finding that WPTERD-Time is NP-hard, we propose an approximate algorithm to solve the WPTERD problem by exploiting the concurrent energy transmission opportunities. The algorithm is named as Greedy Energy ReDistribution Scheduling (GERDS) algorithm, which is proved to finish in limited loops. Simulations validate the algorithm's ability in minimizing energy loss, and meanwhile return near-optimal schedules with time span in range $100{\sim }125$ % of the optimal value. Simulation results also show that GERDS outperforms a typical flow-based algorithm in terms of the tested performance metrics.

In this paper, a novel robust graph-based adequate and concise information representation paradigm is explored. This new signal representation framework can provide a promising alternative for manifesting the essential structure of random signals. A typical application, namely, target detection within sea clutter, can thus be carried out using our proposed new graph-based signal characterization. According to Monte Carlo simulation results, the proposed graph-based signal (target) detection method leads to the outstanding performance, compared to other existing techniques, especially when the signal-to-noise ratio is rather small (0–6 dB). This new graph-based target detector can be expected to be the future backbone technique for identifying and tracking marine vessels using high-resolution radars.

A novel robust pilot-tone spotting (PTS) technique for software defined radio (SDR) is introduced in this letter. This proposed new technique is designed upon the new idea of spectrum clustering. Given the spectrum of a band-limited signal, the pilot appears to be a spectral impulse located at the valley between two signal-spectrum segments. Then, the pilot-tone search is carried out thereby to first characterize two spectral segments around the pilot in the frequency domain and then find the pilot-tone frequency within the middle spectral segment. Based on the new PTS algorithm, the pilot signal embedded in the frequency-modulation (FM) stereo signal can be spotted accurately, even in the very low signal-to-noise-ratio conditions. The software-defined radio platform has been utilized to undertake real-world experiments for performance evaluation. Experimental results show that the performance of the proposed new PTS scheme is very promising.

Advances in several engineering fields have led to a trend toward miniaturization and portability of wearable biosensing devices, which used to be confined to large tools and clinical settings. Various systems to continuously measure electrophysiological activity through electrical and optical methods are one category of such devices. Being wearable and intended for prolonged use, the amount of noise introduced on sensors by movement remains a challenge and requires further optimization. User movement causes motion artifacts that alter the overall quality of the signals obtained, hence corrupting the resulting measurements. This paper introduces a fully wearable optical biosensing system to continuously measure pulse oximetry and heart rate, utilizing a reflectance-based probe. Furthermore, a novel data-dependent motion artifact tailoring algorithm is implemented to eliminate noisy data due to the motion artifact and measure oxygenation level with high accuracy in real time. By taking advantages of current wireless transmission and signal processing technologies, the developed wearable photoplethysmography device successfully captures the measured signals and sends them wirelessly to a mobile device for signal processing in real time. After applying motion artifact tailoring, evaluating accuracy with a continuous clinical device, the blood oxygenation measurements obtained from our system yielded an accuracy of at least 98%, when compared to a range of 93.6%-96.7% observed before from the same initial data. Additionally, heart rate accuracy above 97% was achieved. Motion artifact tailoring and removal in real time, continuous systems will allow wearable devices to be truly wearable and a reliable electrophysiological monitoring and diagnostics tool for everyday use.

System implementation is full of many unknowns and risks, and different solutions emerge according to the various context structure in the process. This study explored how solutions emerge during a system implementation project to understand the interactions among series of events, actions, outcomes, and their impacts on implementation. This study employed the contextualistic approach to analyze the solution emerging process for the system implementation and integration project of a case company which was initiated by the managers of the Accounting and IT departments. The main goal of the project was to solve the issue of too many exceptions in sales-allowance orders by implementing a new workflow system and integrating it with the ERP system. This study identified, in the solution emerging process, the events, the stakeholders and their responses to the events, and the action outcomes and their impacts on the context structure. Based on the analysis, this study discussed the possible strategies to speed up the solution emerging process.

We propose a novel device-free localization technique for human-object tracking indoors using wireless radiofrequency fingerprint. The received signal-strength indicators (RSSIs) are measured by the receiving antennae as the features for machine learning subject to the Random Forest model. In our proposed approach, both transmitters and receivers are fixed within the room, thus making human(s) free from carrying any transceiver. The placement of receivers will impact on the localization accuracy, and therefore we investigate the effect of receiver placement in this work. We will introduce an optimal receiver placement strategy such that the average communication-link distance can be minimized. Our proposed method is verified through simulations supported by a popular channel-propagation software, Feko. The pertinent experimental results demonstrate that the localization accuracy of 77.50% can be attained by our proposed Random Forest learning system for a 20 m×10 m indoor area divided into eight equi-sized zones where sixteen receiving antennae are placed at their optimal locations along the perimeter of the room.

Internet of things (IoT) has emerged as a crucial technology nowadays. To efficiently transmit data across smart devices, short packets and software-defined networks (SDNs) are employed in IoT. In contrast to the prevalent packet-based communications, we recently devised new software-defined multiplexing codes (SDMCs) which enable packetless communications in the future. The SDMC scheme simply relies on simple algorithms to insert delimiters as needed such that no fixed packet boundaries are required anymore. In this paper, the SDMC capacity of information fusion is further extended to electronic logistics. Two popular existing logistic coding approaches have been addressed, namely Dabbawala graphic code (DGC) and baggage information message (BIM) protocol. Then we compare the electronic implementation of DGC and BIM using the conventional packet-based technique with our proposed new packetless method. Our new packetless method can lead to better coding efficiency than the conventional technique when the delimiter-length is not very large; it can also facilitate selective encryption over important partial information in the message while the conventional technique cannot. Our proposed new approach is a very flexible and efficient scheme for future electronic logistics transmitted by software-defined edge-computing devices.

Wireless communication and sensor networks have been widely employed in our daily life. Sensing-range characteristics, inter-nodal connectivity, and communication capacity are critical factors in providing reliable quality-of-service. Placement of sensors or access-points indoors is always critical to the effectiveness of a wireless communication/sensor network. In this paper, we would like to investigate the important classical sensor/access-point placement problem in the discrete-graph domain. Instead of the conventional trial-and-error approach upon rigorous quality-of-service measurements, we study the sensor/access-point coverage-area maximization problem over visibility graphs and then design a new graph-theoretical approach to deal with it. This new approach just takes the bird's eye view of an indoor geometry (floor plan) as the only necessary input information and does not rely on any quality-of-service measurement. Thus, the crucial coverage-area maximization problem for sensor/access-point placement can be tackled algorithmically. Our new approach can be applied for indoor geometries as arbitrary polygons, either convex or nonconvex.

A novel radio-frequency (RF) fingerprint recognition scheme using finite-element-multiwavelets-based cyclic-spectrum graph analysis is introduced in this work. Recognizing or classifying RF signals emitted from different devices without any help of spread/embedded identification codes or sequences is intriguing but challenging nowadays. In our proposed new approach, an RF signal emitted by an electronic device is first decomposed into multiple subbands using finite-element multiwavelets with a certain appropriate multiplicity. Thus, the cyclic spectra of extracted subband signals are formed and then the corresponding graphs can be established thereby. Such graph features may represent the RF fingerprints and be utilized to distinguish (recognize) devices by use of the discrepancies among them. Our simulations are based on realworld RF signals and the results demonstrate that the proposed finite-element-multiwavelets-based scheme can lead to higher correct recognition rates but lower computational complexities than the prevalent GHM-multiwavelets-based scheme and the discrete-Fourier-transform (DFT) method.

A novel cooperative automatic modulation classification (AMC) scheme by use of the credit-based consensus fusion is introduced in this paper. We propose a two-stage cooperative AMC approach to be deployed by a wireless sensor network consisting of a fusion center (FC). The unknown modulation scheme of a target signal is first identified by each individual sensor using graph-based automatic modulation classification. Then all local (individual) decisions are combined by the fusion center to produce the ultimate result according to the associated individual credits which are pertinent to the respective historical classification accuracies. Compared to the existing AMC schemes, which can be implemented only on a single sensing device (receiver), our proposed new cooperative AMC method can increase the overall classification accuracy as the cooperation and credit-based consensus fusion employed in our new method can greatly mitigate the adverse effect of channel distortion and noise impairment often encountered in the existing single-sensing-device modulation classifiers. Meanwhile, the extra communication overhead required by our new method is quite limited and therefore the incurred communication cost is minimum. Monte Carlo simulation results have justified the effectiveness and superiority of our proposed new cooperative AMC method.

This paper proposes a novel indoor device-free localization and tracking approach using the received signal-strength indicators (RSSIs) of WiFi signals. The RSSI feature-vectors simulated by a channel-propagation emulator software are adopted as the training data for our proposed scheme. Prevalent discriminative machine-learning methods are used to predict the locations of a moving human-object. Hidden Markov models (HMMs) are also incorporated with such machine-learning techniques for robust and reliable indoor tracking. In this work, we partition the given indoor geometry into several equi-sized zones and then convert the underlying localization/tracking problem to the classical multi-classification problem. Simulation results demonstrate that the gradient boosting decision-tree (GBDT) classifier in conjunction with the Viterbi algorithm over hidden Markov models leads to the highest localization-accuracies of 83.9% for eight zones and 71.4% for sixteen zones. As a result, our proposed new indoor localization and tracking scheme can be very promising for many indoor device-free surveillance applications in the future.

Digital Terrestrial Television (DTV) has been widely deployed world-wide in recent years. The transmitter identification (Tx-ID) technique specified in modern DTV standards becomes crucial nowadays as the number of DTV transmitters grows with the expanded coverage area. In the ATSC standards, Kasami sequences, an important family of pseudo random sequences, are adopted as the practical Tx-ID sequences since Kasami sequences posess the favorable properties of nearly impulse autocorrelation/cross-correlation functions and large sequence capacities. In this work, we would like to extend the existing study to address the corresponding Tx-ID reception quality in terms of received signal-to-interference ratio (RSIR) to various channel factors such as propagation fading and propagation decay rate along with Kasami correlational properties. Such in-depth RSIR study can help us to determine the capacity (the number) of effective Tx-ID sequences which can be utilized subject to a given transmitter-deployment topology, a given fading channel, and a given Tx-ID sequence-length. Extensive numerical experiments are also presented to illustrate the relationship between the capacity of Tx-ID sequences and aforementioned pertinent factors.

Traditional techniques to identify macromolecular targets for drugs utilize solely the information on a query drug and a putative target. Nonetheless, the mechanisms of action of many drugs depend not only on their binding affinity toward a single protein, but also on the signal transduction through cascades of molecular interactions leading to certain phenotypes. Although using protein-protein interaction networks and drug-perturbed gene expression profiles can facilitate system-level investigations of drug-target interactions, utilizing such large and heterogeneous data poses notable challenges. To improve the state-of-the-art in drug target identification, we developed GraphDTI, a robust machine learning framework integrating the molecular-level information on drugs, proteins, and binding sites with the system-level information on gene expression and protein-protein interactions. In order to properly evaluate the performance of GraphDTI, we compiled a high-quality benchmarking dataset and devised a new cluster-based cross-validation protocol. Encouragingly, GraphDTI not only yields an AUC of 0.996 against the validation dataset, but it also generalizes well to unseen data with an AUC of 0.939, significantly outperforming other predictors. Finally, selected examples of identified drugtarget interactions are validated against the biomedical literature. Numerous applications of GraphDTI include the investigation of drug polypharmacological effects, side effects through offtarget binding, and repositioning opportunities.

In this paper, a novel blind signal-to-noise-plus-interference ratio (SINR) estimator is proposed based on graph sparsity. The graph representation, which can serve as a new promising alternative to time-, frequency-, and transform-domain waveforms, is capable of extracting sufficient and concise information for communication signals. We discover that the sparsity of the ultimate graph converted from the original signal waveform can be utilized to measure the SINR of the received communication signal without any need of training symbols. For this purpose, a new blind SINR estimation algorithm is introduced in this work. According to Monte Carlo simulations in comparison with the well-known blind SINR estimator using higher-order statistics, our proposed new graph-based SINR estimation scheme has demonstrated the excellent performance, especially for relatively small sample size.

Moving objects, especially humans and vehicles, impose very high risks on driverless cars or autonomous driving. Real-time moving-target detection plays a critical role in the future intelligent transportation systems due to public-safety concerns. Recently object tracking and scene analysis using a hybrid of high-resolution RGB images and low-resolution LiDAR point-clouds have been drawing a lot of research interest because fast but precise information rendering can be facilitated thereby. Compared to the conventional object detectors using RGB images, the new target detectors using LiDAR point-clouds can lead to three-dimensional object localization with the crucial depth information RGB images cannot provide, which is very critical to navigation of autonomous vehicles and robots. In this paper, we explore a novel moving-target detection approach which is built upon motion-estimation using RGB images and deep-learning over LiDAR data so that the moving object(s) can be accurately localized in an image frame and the target recognition can be conducted using LiDAR point-clouds. Our proposed new scheme can benefit from both tracking-accuracy brought by RGB images and sufficient training-data together with low time-complexity brought by LiDAR data. Preliminary experimental results have shown that our proposed scheme can automatically identify and track moving human-objects in the scene successively.

Smart assistants have been indispensable mechanisms on electronic devices nowadays. However, as smart devices become prevalent everywhere, eavesdropping attacks could be encountered. Hence, information security is very important. Lately, we proposed a new efficient cryptographic mosaic technique, namely efficient recoverable cryptographic mosaic technique by permutations, for image mosaicing. In this paper, we further extend this mosaic technique for audio signals. To evaluate the effectiveness of this new audio-mosaicing scheme, here we propose a new audio signal-discrepancy measure, namely KullbackLeibler divergence of spectrogram (spec-KLD). We also relate this spec-KLD measure to the well-known signal-to-noise ratio (SNR) by benchmarking the signal destructuring effect. We also present the audio-mosaicing results based on speech recognition by Siri to demonstrate that our proposed approach can greatly (successfully) destroy the speech intelligibility for information hiding. The intelligence loss of speech is also enumerated in terms of spec-KLD and SNR.

3D LiDAR sensor and 2D color camera provide complementary information of a scene. It is essential to fuse both types of data to obtain a holistic representation of the real-world scene. In this paper, we present a method to fuse the data captured by rigidly mounted LiDAR and camera. We first cocalibrate the LiDAR and camera with a calibration target. We then establish 3D-2D correspondences and fuse the LiDAR and camera data through geometric transformations. Experimental results demonstrate that our method is able to fuse the 3D LiDAR point cloud with the 2D color image efficiently and accurately. Our method is also fast and can be used in autonomous driving applications.

This paper introduces a new indoor-localization approach using a deep-learning network for which the received signal-strength indicator (RSSI) is adopted as the radiofrequency fingerprint. In our proposed scheme, the RSSIs which are estimated by a channel-propagation emulator software, are adopted as the input features for deep learning. This approach is measurement-free and thus it is very cost-effective and convenient to users. A multilayer perceptron (MLP) is constructed to predict the location(s) of the mobile object(s). The time evolution of the predicted locations of an object will form the predicted trajectory thereby. Because deep-learning networks require tremendous training data to achieve good prediction accuracy, we propose to partition the indoor geometry of interest (ex., a room) into several zones. Preliminary simulation results demonstrate that the AUC (area under the receiver-operating characteristic curve) can reach up to 0.89 for a room partitioned into eight zones. Our proposed new indoor-localization scheme in this work can be a rare but promising localization technology, which is neither passive nor active as other existing prevalent localization methods.

Proper deployment of roadside units (RSUs) is of crucial importance to vehicular ad hoc networks (VANETs). However, our understandings to the simple one-dimensional RSU Deployment (D1RD) problem with nonuniform profit density is still seriously limited. In this paper, we analyze the D1RD problem and try to design optimal algorithms for it. We first analyze the properties of the optimal solutions of the D1RD problem involving a single RSU, and then extend to multiple RSUs. Next, we propose an efficient technique named Dynamic Limiting (DynLim), which reduces the solution search space size considerably by adjusting search space limits dynamically. Finally, an optimal algorithm named OptDynLim is proposed based on the DynLim technique, and its optimality is proved. Numerical simulations validate the correctness of our analyzes and show that DynLim can usually reduce solution search space size by more than 99%.

An innovative automatic modulation classification via graph-based constellation analysis for ${M}$ -ary QAM signals is presented. In our framework, a unified mesh model for the constellation diagrams of the ${M}$ -QAM signals within the modulation candidate set is first constructed and exploited to transform the received ${M}$ -QAM signal into graph domain. The concise graph representation of the received ${M}$ -QAM signal is established from its constellation according to the positions of the recovered symbols in the mesh model. Then, the modulation feature vector is built from the eigenvector(s) corresponding to the maximum eigenvalue of its adjacency matrix. The modulation type can be identified by measuring the angle between the feature vectors resulting from the training data and the test data. Monte Carlo simulation results and theoretical analysis demonstrate that the proposed method with lower computational complexity can provide superior performance to the existing subtractive clustering technique, and is robust to the residual phase and timing offsets.

Genomic profiles of cancer cells provide valuable information on genetic alterations in cancer. Several recent studies employed these data to predict the response of cancer cell lines to treatment with drugs. Nonetheless, due to the multifactorial phenotypes and intricate mechanisms of cancer, the accurate prediction of the effect of pharmacotherapy on a specific cell line based on the genetic information alone is problematic. High prediction accuracies reported in the literature likely result from significant overlaps among training, validation, and testing sets, making many predictors inapplicable to new data. To address these issues, we developed GraphGR, a graph neural network with sophisticated attention propagation mechanisms to predict the therapeutic effects of kinase inhibitors across various tumors. Emphasizing on the system-level complexity of cancer, GraphGR integrates multiple heterogeneous data, such as biological networks, genomics, inhibitor profiling, and genedisease associations, into a unified graph structure. In order to construct diverse and information-rich cancer-specific networks, we devised a novel graph reduction protocol based on not only the topological information, but also the biological knowledge. The performance of GraphGR, properly cross-validated at the tissue level, is 0.83 in terms of the area under the receiver operating characteristics, which is notably higher than those measured for other approaches on the same data. Finally, several new predictions are validated against the biomedical literature demonstrating that GraphGR generalizes well to unseen data, i.e. it can predict therapeutic effects across a variety of cancer cell lines and inhibitors. GraphGR is freely available to the academic community at https://github.com/pulimeng/GraphGR.

The efficiency of drug development defined as a number of successfully launched new pharmaceuticals normalized by financial investments has significantly declined. Nonetheless, recent advances in high-throughput experimental techniques and computational modeling promise reductions in the costs and development times required to bring new drugs to market. The prediction of toxicity of drug candidates is one of the important components of modern drug discovery. In this work, we describe eToxPred, a new approach to reliably estimate the toxicity and synthetic accessibility of small organic compounds. eToxPred employs machine learning algorithms trained on molecular fingerprints to evaluate drug candidates. The performance is assessed against multiple datasets containing known drugs, potentially hazardous chemicals, natural products, and synthetic bioactive compounds. Encouragingly, eToxPred predicts the synthetic accessibility with the mean square error of only 4% and the toxicity with the accuracy of as high as 72%. eToxPred can be incorporated into protocols to construct custom libraries for virtual screening in order to filter out those drug candidates that are potentially toxic or would be difficult to synthesize. It is freely available as a stand-alone software at https://github.com/pulimeng/etoxpred .

Cognitive radio network (CRN) and non-orthogonal multiple-access (NOMA) play the crucial roles in the fifth generation wireless communication systems. However, the physical-layer security still remains quite challenging. In this letter, a downlink cascaded transmitting zero-forcing-beamforming (ZFBF) technique is proposed to secure communications in a two-cell multiple-input multiple-output (MIMO) NOMA-based CRN. The proposed technique protects the information from illegitimate users (eavesdroppers) within the same and adjacent cells. The restriction on the number of at the transmitter can also be relaxed by applying the signal alignment for the ZFBF technique. This proposed new method can maximize the total secrecy rate of the MIMO-NOMA-based CRNs in the lack of conventional transceiver beamforming techniques. Numerical and simulation evaluations have demonstrated the effectiveness of the proposed method.

Optimal road side unit (RSU) deployment is of crucial importance to vehicular ad hoc networks. The RSU deployment (RD) problem models in the literature usually lack the ability to describe curve-shaped roads and roads nonuniform statistics. We had proposed a more realistic RD problem model taking these features into consideration. However, our understanding of the RD problem with the new model is still seriously limited. In this paper, we try to deepen our understanding of the one-dimensional RD (D1RD) problem with the new model by analyzing its properties and designing optimal and efficient approximate algorithms for it. We first analyze the properties of optimal solutions of the D1RD problem. Next, suspecting that the D1RD problem with $n$ RSUs of different coverage radii is intractable, we propose two greedy-based algorithms (named as Greedy2P3 and Greedy2P3E) and show that Greedy2P3E's approximation ratio is at least $1-(\frac{n-1}{n})^2$ . Then, by exploiting the properties of the D1RD problem, we propose an optimal algorithm named OptGreDyn by combining greedy idea and dynamic programming, and prove its optimality. At last, we proved that, if applied to the RD problem with $n$ RSUs of identical radii, the approximation ratios of Greedy2P3 and Greedy2P3E are at least $\frac{2}{3}$ , and it is tight for Greedy2P3 when $n=3i$ , where $i$ is a positive integer. Simulation results verify the optimality of OptGreDyn, meanwhile show that Greedy2P3 and Greedy2P3E usually return near-optimal solutions with profit more than 98% of the optimal solutions, and they are preferable than the existing approximate algorithms.

Power-line communications (PLCs) endure a severe transmission channel distortion due to Gaussian noise, impulsive noise, and multipath self-interference. Conventional transmission models with white Gaussian noise would be very inappropriate and impair the necessary channel estimation drastically to the PLC receivers. In this study, a statistical model for impulsive noise, namely Middleton Class A model, is employed for Gaussian information sources. Robust channel estimation algorithms based on a novel parameter estimation scheme are established thereby. Superior to the conventional robust-least-square estimation algorithms, the authors' proposed new adaptive robust-least-square estimation methods can take advantage of the statistical parameter estimation for impulsive noise. The studies on the theoretical and practical aspects of their proposed new parameter estimator are also presented.

In order to facilitate the ubiquitous heterogeneous and medium-independent communication networks, transmitter identification is crucial especially as the number of transmitters becomes tremendous in reality. Since the birth of spread-spectrum communications, pseudo-random identification sequences have been widely adopted due to their preferable nearly Dirac-Delta auto-correlation property, which would lead to the advantages in synchronization and multi-user interference mitigation. Maximum-length shift-register sequences (a.k.a. m -sequences) are pseudo-random sequences often adopted for multi-access communications, but categorization or grouping of the patterns (bit substrings) contained in m -sequences has not been investigated to the best of our knowledge. Thus, this paper is dedicated to the new study on the categorization of m -sequences, which would give rise to many applications involving assignment, addressing, and management of identification or spreading sequences. We define new essential parameters, namely single-pattern-searching parameters, and design a new parallel algorithm to spot such inherent parameters associated with each selected underlying pattern. Our proposed new approach will facilitate a highly computationally-efficient solution to find the pattern-attributed (pattern-contained) m -sequences without actually generating any m -sequence for subsequence-matching between an m -sequence and the underlying pattern(s). We further utilize these single-pattern-searching parameters to establish the new analysis of pattern capacity and quality by characterizing multiple pattern attributes within m -sequences. The first goal of this work is to find single-pattern-attributed m -sequences recursively until the final subset of m -sequences is acquired so that each m -sequence contains all underlying patterns. The second goal of this work is to find the shortest sequences containing specified patterns. To achieve this goal, we first construct the shortest binary sequences subject to the underlying patterns and it can be accomplished by solving the generalized traveling salesman problem (GTSP). Then, according to the constructed shortest binary sequences, the number of pattern-contained m -sequences can be determined thereby. Memory- and computational-complexity analyses are also presented to demonstrate that our proposed new scheme is much more computationally efficient than the conventional subsequence-matching method for searching and counting the pattern-attributed m -sequences. On the other hand, our proposed new scheme leads to the same memory-complexity as the conventional method for registering the spotted pattern-attributed m -sequences. Finally, three new metrics for assessing the underlying patterns are proposed. They are attributability, discriminability, and ambiguity. The associated numerical evaluation is also provided in this paper.

This paper presents a novel automatic modulation classification (AMC) method using graph-based constellation analysis, to classify $M$ -ary Generalized Quadrature Amplitude Modulation (GQAM) signals which employ uniform or nonuniform constellations for the first time. In our framework, a unified grid model is first built from the GQAM signal with the maximum constellation size in the modulation candidate set, and exploited to transform the received signal into graph domain. The graph representation of the received signal is established by mapping its recovered symbol points on the I/Q plane into the unified grid model, and then the eigenvalue(s) and eigenvector(s) of its corresponding adjacency matrix are computed. The modulation feature vector is constructed according to the eigenvector(s) corresponding to the maximum eigenvalue(s). The modulation type is eventually identified by searching the minimum angle between training features and test feature. Monte Carlo simulation results demonstrate that the proposed method can effectively classify the GQAM even in low signal-to-noise ratio.

The applications of unmanned aerial vehicles (UAVs) have increased dramatically in the past decade. Meanwhile, close-range UAV detection has been intriguing by many researchers for its great importance in privacy, security, and safety control. Positioning of the UAV controller (hub) is quite challenging but still difficult. In order to combat this emerging problem for public interest, we propose to utilize a software-defined radio (SDR) platform, namely HackRF One, to enable the UAV hub detection and localization. The SDR receiver can acquire the UAV source signals. The theoretical path-loss propagation model is adopted to predict the signal strength attenuation. Thus, the UAV hub location can be estimated using the modified multilateration approach by only three or more SDR receivers.