Labeled data used for training activity recognition classifiers are usually limited in terms of size and diversity. Thus, the learned model may not generalize well when used in real-world use cases. Semi-supervised learning augments labeled examples with unlabeled examples, often resulting in improved performance. However, the semi-supervised methods studied in the activity recognition literatures assume that feature engineering is already done. In this paper, we lift this assumption and present two semi-supervised methods based on convolutional neural networks (CNNs) to learn discriminative hidden features. Our semi-supervised CNNs learn from both labeled and unlabeled data while also performing feature learning on raw sensor data. In experiments on three real world datasets, we show that our CNNs outperform supervised methods and traditional semi-supervised learning methods by up to 18% in mean F1-score (Fm).

This article presents advances in resource allocation (RA) for downlink non-orthogonal multiple access (NOMA) systems, focusing on user pairing (UP) and power allocation (PA) algorithms. The former pairs the users to obtain the high capacity gain by exploiting the channel gain difference between the users, while the later allocates power to users in each cluster to balance system throughput and user fairness. Additionally, the article introduces the concept of cluster fairness and proposes the divideand- next largest difference-based UP algorithm to distribute the capacity gain among the NOMA clusters in a controlled manner. Furthermore, performance comparison between multiple-input multiple-output NOMA (MIMO-NOMA) and MIMO-OMA is conducted when users have pre-defined quality of service. Simulation results are presented, which validate the advantages of NOMA over OMA. Finally, the article provides avenues for further research on RA for downlink NOMA.

Fast digitisers and digital pulse processing have been widely used for spectral application and pulse shape discrimination (PSD) owing to their advantages in terms of compactness, higher trigger rates, offline analysis, etc. Meanwhile, the noise of readout electronics is usually trivial for organic, plastic, or liquid scintillator with PSD ability because of their poor intrinsic energy resolution. However, LaBr3(Ce) has been widely used for its excellent energy resolution and has been proven to have PSD ability for alpha/gamma particles. Therefore, designing a digital acquisition system for such scintillators as LaBr3(Ce) with both optimal energy resolution and promising PSD ability is worthwhile. Several experimental research studies about the choice of digitiser properties for liquid scintillators have already been conducted in terms of the sampling rate and vertical resolution. Quantitative analysis on the influence of waveform digitisers, that is, fast amplifier (optional), sampling rates, and vertical resolution, on both applications is still lacking. The present paper provides quantitative analysis of these factors and, hence, general rules about the optimal design of digitisers for both energy resolution and PSD application according to the noise analysis of time-variant gated charge integration.

In this paper, we compare the individual rate of MIMO-NOMA and MIMO-OMA when users are paired into clusters. A power allocation (PA) strategy is proposed, which ensures that MIMO-NOMA achieves a higher individual rate for each user than MIMO-OMA with arbitrary PA and optimal degrees of freedom split. In addition, a special case with equal degrees of freedom and arbitrary PA for OMA is considered, for which the individual rate superiority of NOMA still holds. Moreover, it is shown that NOMA can attain better fairness through appropriate PA. Finally, simulations are carried out, which validate the developed analytical results.

In this paper, we consider the power allocation (PA) problem in cognitive radio networks (CRNs) employing nonorthogonal multiple access (NOMA) technique. Specifically, we aim to maximize the number of admitted secondary users (SUs) and their throughput, without violating the interference tolerance threshold of the primary users (PUs). This problem is divided into a two-phase PA process: a) maximizing the number of admitted SUs; b) maximizing the minimum throughput among the admitted SUs. To address the first phase, we apply a sequential and iterative PA algorithm, which fully exploits the characteristics of the NOMA-based system. Following this, the second phase is shown to be quasiconvex and is optimally solved via the bisection method. Furthermore, we prove the existence of a unique solution for the second phase and propose another PA algorithm, which is also optimal and significantly reduces the complexity in contrast with the bisection method. Simulation results verify the effectiveness of the proposed two-phase PA scheme.

The evolution of conventional wireless communication networks to the fifth generation (5G) is driven by an explosive increase in the number of wireless mobile devices and services, as well as their demand for all-time and everywhere connectivity, high data rates, low latency, high energy-efficiency and improved quality of service. To address these challenges, 5G relies on key technologies, such as full duplex (FD), device-to-device (D2D) communications, and network densification. In this article, a heterogeneous networking architecture is envisioned, where cells of different sizes and radio access technologies coexist. Specifically, collaboration for spectrum access is explored for both FD- and cognitive-based approaches, and cooperation among devices is discussed in the context of the state-of-the-art D2D assisted communication paradigm. The presented cooperative framework is expected to advance the understandings of the critical technical issues towards dynamic spectrum management for 5G heterogeneous networks.

In this letter, we investigate the energy efficiency (EE) problem in a millimeter wave (mmWave) massive MIMO (mMIMO) system with non-orthogonal multiple access (NOMA). Multiple two-user clusters are formulated according to their channel correlation and gain difference. Following this, we propose a hybrid analog/digital precoding scheme for the low radio frequency (RF) chains structure at the base station (BS). On this basis, we formulate a power allocation problem aiming to maximize the EE under users' quality of service (QoS) requirements and per-cluster power constraint. An iterative algorithm is proposed to obtain the optimal power allocation. Simulation results show that the proposed NOMA achieves superior EE performance than that of conventional OMA.

This article provides an overview of power-domain non-orthogonal multiple access for 5G systems. The basic concepts and benefits are briefly presented, along with current solutions and standardization activities. In addition, limitations and research challenges are discussed.

In this paper, the performance of multiple-input multiple-output non-orthogonal multiple access (MIMO-NOMA) is investigated when multiple users are grouped into a cluster. The superiority of MIMO-NOMA over MIMO orthogonal multiple access (MIMO-OMA) in terms of both sum channel capacity and ergodic sum capacity is proved analytically. Furthermore, it is demonstrated that the more users are admitted to a cluster, the lower is the achieved sum rate, which illustrates the tradeoff between the sum rate and maximum number of admitted users. On this basis, a user admission scheme is proposed, which is optimal in terms of both sum rate and number of admitted users when the signal-to-interference-plus-noise ratio thresholds of the users are equal. When these thresholds are different, the proposed scheme still achieves good performance in balancing both criteria. Moreover, under certain conditions,it maximizes the number of admitted users. In addition, the complexity of the proposed scheme is linear to the number of users per cluster. Simulation results verify the superiority of MIMO-NOMA over MIMO-OMA in terms of both sum rate and user fairness, as well as the effectiveness of the proposed user admission scheme.

In neutrinoless double beta (0{\nu}{\beta}{\beta}) decay experiments, the diversity of topological signatures of different particles provides an important tool to distinguish double beta events from background events and reduce background rates. Aiming at suppressing the single-electron backgrounds which are most challenging, several groups have established Monte Carlo simulation packages to study the topological characteristics of single-electron events and 0{\nu}{\beta}{\beta} events and develop methods to differentiate them. In this paper, applying the knowledge of graph theory, a new topological signature called REF track (Refined Energy-Filtered track) is proposed and proven to be an accurate approximation of the real particle trajectory. Based on the analysis of the energy depositions along the REF track of single-electron events and 0{\nu}{\beta}{\beta} events, the REF energy deposition models for both events are proposed to indicate the significant differences between them. With these differences, this paper presents a new discrimination method, which, in the Monte Carlo simulation, achieved a single-electron rejection factor of 93.8+-0.3 (stat.)% as well as a 0{\nu}{\beta}{\beta} efficiency of 85.6+-0.4 (stat.)% with optimized parameters in CdZnTe.

The key to photoelectric X-ray polarimetry is the determination of the emission direction of photoelectrons. Because of the low mass of an electron, the ionisation trajectory is not straight and the useful information needed for polarimetry is stored mostly in the initial part of the track where less energy is deposited. We present a new algorithm, based on the shortest path problem in graph theory, to reconstruct the 2D electron track from the measured image that is blurred due to transversal diffusion along drift and multiplication in the gas chamber. Compared with previous methods based on moment analysis, this algorithm allows us to identify the photoelectric interaction point more accurately and precisely for complicated tracks resulting from high energy photons or low pressure chambers. This leads to a better position resolution and a higher degree of modulation toward high energy X-rays. The new algorithm is justified using simulations and measurements with the gas pixel detector (GPD), and it should also work for other polarimetric techniques such as a time projection chamber (TPC). As the improvement is restricted in the high energy band, this new algorithm shows limited improvement for the sensitivity of GPD polarimeters, but it may have a larger potential for low-pressure TPC polarimeters.

Laser Wakefield Accelerator (LWFA) is considered as one of the most competitive candidates for the accelerators of the next generation. With the development of high power laser technologies, LWFA has shown its potential of replacing the conventional radio-frequency (RF) accelerators due to its flexibility and adjustability. In this paper, we will study the potential high flux electron beam productions of LWFA driven by petawatt-level laser pulses. In our three dimensional particle-in-cell simulations, an optimal set of parameters gives $\sim 40\ \rm nC$ of charge with $2\ \rm PW$ laser power, thus $\sim 400\ \rm kA$ of instantaneous current if we assume the electron beam duration is 100 fs. This high flux and its secondary radiation are widely applicable in nuclear and QED physics, industrial imaging, medical and biological studies.

An underwater in situ gamma ray spectrometer based on LaBr3 was developed and optimized to monitor marine radioactivity. The intrinsic background mainly from La138 and Ac227 of LaBr3 was well determined by low background measurement and pulse shape discrimination method. A method of self-calibration using three internal contaminant peaks was proposed to eliminate the peak shift during long term monitoring. With experiments under different temperatures, the method was proved to be helpful for maintaining long term stability. To monitor the marine radioactivity, the spectrometer efficiency was calculated via water tank experiment as well as Monte Carlo simulation.

City-identification of videos aims to determine the likelihood of a video belonging to a set of cities. In this paper, we present an approach using only audio, thus we do not use any additional modality such as images, user-tags or geo-tags. In this manner, we show to what extent the city-location of videos correlates to their acoustic information. Success in this task suggests improvements can be made to complement the other modalities. In particular, we present a method to compute and use semantic acoustic features to perform city-identification and the features show semantic evidence of the identification. The semantic evidence is given by a taxonomy of urban sounds and expresses the potential presence of these sounds in the city- soundtracks. We used the MediaEval Placing Task set, which contains Flickr videos labeled by city. In addition, we used the UrbanSound8K set containing audio clips labeled by sound- type. Our method improved the state-of-the-art performance and provides a novel semantic approach to this task

The ability of background discrimination using pulse shape discrimination (PSD) in broad-energy germanium (BEGe) detectors makes them as competitive candidates for neutrinoless double beta decay (0{\nu}\b{eta}\b{eta}) experiments. The measurements of key parameters for detector modeling in a commercial p-type BEGe detector are presented in this paper. Point-like sources were used to investigate the energy resolution and linearity of the detector. A cylindrical volume source was used for the efficiency calibration. With an assembled device for source positioning, a collimated 133Ba point-like source was used to scan the detector and investigate the active volume. A point-like source of 241Am was used to measure the dead layer thicknesses, which are approximately 0.17 mm on the front and 1.18 mm on the side. The described characterization method will play an important role in the 0{\nu}\b{eta}\b{eta} experiments with BEGe detectors at China JinPing underground Laboratory (CJPL) in the future.

For the laser wakefield acceleration, suppression of beam energy spread while keeping sufficient charge is one of the key challenges. In order to achieve this, we propose bichromatic laser ionization injection with combined laser wavelengths of $2.4\rm \mu m$ and $0.8\rm \mu m$ for wakefield excitation and for triggering electron injection via field ionization, respectively. A laser pulse at $2.4\rm \mu m$ wavelength enables one to drive an intense acceleration structure with relatively low laser power. To further reduce the requirement of laser power, we also propose to use carbon dioxide as the working gas medium, where carbon acts as the injection element. Our full three dimensional particle-in-cell simulations show that electron beams at the GeV energy level with both low energy spreads (around one percent) and high charges (several tens of picocoulomb) can be obtained by this scheme with laser parameters achievable in the near future.

Encoding readout methods based on different schemes have been successfully developed and tested with different types of position-sensitive detectors with strip-readout structures. However, how to construct an encoding scheme in a more general and systematic way is still under study. In this paper, we present a graph model for the encoding scheme. With this model, encoding schemes can be studied in a more systematic way. It is shown that by using an encoding readout method, a maximum of n(n-1)/2+1 strips can be processed with n channels if n is odd, while a maximum of (n(n-2))/2+2 strips can be processed with n channels if n is even. Furthermore, based on the model, the encoding scheme construction problem can be translated into a problem in graph theory, the aim of which is to construct an Eulerian trail such that the length of the shortest subcycle is as long as possible. A more general approach to constructing the encoding scheme is found by solving the associated mathematical problem. In addition, an encoding scheme prototype has been constructed, and verified with MRPC detectors.

The China Dark Matter Experiment (CDEX) is a low background experiment at China Jinping Underground Laboratory (CJPL) designed to directly detect dark matter with a high-purity Germanium (HPGe) detector. In the second phase CDEX-10 with a 10 kg Germanium array detector system, the liquid argon (LAr) anti-compton active shielding and cooling system is proposed. For purpose of studying the properties of LAr detector, a prototype with an active volume of 7 liters of liquid argon was built and operated. The photoelectron yields, as a critically important parameter for the prototype detector, have been measured to be 0.051-0.079 p.e./keV for 662 keV Gamma lines at different positions. The good agreement between the experimental and simulation results has provided a quite reasonable understanding and determination of the important parameters such as the Surviving Fraction of the Ar2 excimers, the absorption length for 128 nm photons in liquid argon, the reflectivity of Teflon and so on.

Laser wakefield accelerators have great potential as the basis for next generation compact radiation sources because their accelerating gradients are three orders of magnitude larger than traditional accelerators. However, X-ray radiation from such devices still lacks of tunability, especially the intensity and polarization distribution. Here we propose a tunable polarized radiation source from a helical plasma undulator based on plasma channel guided wakefield accelerator. When a laser pulse is initially incident with a skew angle relative to the channel axis, the laser and accelerated electrons experience collective spiral motions, which leads to elliptically polarized synchrotron-like radiation with flexible tunability on radiation intensity, spectra and polarization. We demonstrate that a radiation source with millimeter size and peak brilliance of $2\times10^{19} photons/s/mm^{2}/mrad^{2}/0.1%$ bandwidth can be made with moderate laser and electron beam parameters. This brilliance is comparable with the third generation of synchrotron radiation facilities running at similar photon energies, suggesting that laser plasma based radiation sources are promising for advanced applications.

The LaBr3:Ce scintillator has been widely studied for nuclear spectroscopy because of its optimal energy resolution (<3%@ 662 keV) and time resolution (~300 ps). Despite these promising properties, the intrinsic radiation background of LaBr3:Ce is a critical issue, and pulse shape discrimination (PSD) has been shown to be an efficient potential method to suppress the alpha background from the 227Ac. In this paper, the charge comparison method (CCM) for alpha and gamma discrimination in LaBr3:Ce is quantitatively analysed and compared with two other typical PSD methods using digital pulse processing. The algorithm parameters and discrimination efficiency are calculated for each method. Moreover, for the CCM, the correlation between the CCM feature value distribution and the total charge (energy) is studied, and a fitting equation for the correlation is inferred and experimentally verified. Using the equations, an energy-dependent threshold can be chosen to optimize the discrimination efficiency. Additionally, the experimental results show a potential application in low-activity high-energy {\gamma} measurement by suppressing the alpha background.

The X-ray Timing and Polarization (XTP) is a mission concept for a future space borne X-ray observatory and is currently selected for early phase study. We present a new design of X-ray polarimeter based on the time projection gas chamber. The polarimeter, placed above the focal plane, has an additional rear window that allows hard X-rays to penetrate (a transmission of nearly 80% at 6 keV) through it and reach the detector on the focal plane. Such a design is to compensate the low detection efficiency of gas detectors, at a low cost of sensitivity, and can maximize the science return of multilayer hard X-ray telescopes without the risk of moving focal plane instruments. The sensitivity in terms of minimum detectable polarization, based on current instrument configuration, is expected to be 3% for a 1mCrab source given an observing time of 10^5 s. We present preliminary test results, including photoelectron tracks and modulation curves, using a test chamber and polarized X-ray sources in the lab.

A prototype of LaBr3:Ce in situ gamma-ray spectrometer for marine environmental monitoring is developed and applied for in situ measurement. A 3-inch LaBr3:Ce scintillator is used in the detector, and a digital pulse process electronics is chosen as the pulse height analyzer. For this prototype, the energy response of the spectrometer is linear and the energy resolution of 662keV is 2.6% (much better than NaI). With the measurement of the prototype in a water tank filled with 137Cs, the detect efficiency for 137Cs is (0.288 0.01)cps/(Bq/L), which is close to the result of Monte Carlo simulation, 0.283cps/(Bq/L). With this measurement, the MDAC for 137Cs in one hour has been calculated to 0.78Bq/L, better than that of NaI(Tl) in-situ gamma spectrometer, which is ~1.0Bq/L.

Cosmic ray muon tomography is a novel technology to detect high-Z material. A prototype of TUMUTY with 73.6 cm x 73.6 cm large scale position sensitive MRPC detectors has been developed and is introduced in this paper. Three test kits have been tested and image is reconstructed using MAP algorithm. The reconstruction results show that the prototype is working well and the objects with complex structure and small size (20 mm) can be imaged on it, while the high-Z material is distinguishable from the low-Z one. This prototype provides a good platform for our further studies of the physical characteristics and the performances of cosmic ray muon tomography.

A method based on laser wakefield acceleration with controlled ionization injection triggered by another frequency-tripled laser is proposed, which can produce electron bunches with low energy spread. As two color pulses co-propagate in the background plasma, the peak amplitude of the combined laser field is modulated in time and space during the laser propagation due to the plasma dispersion. Ionization injection occurs when the peak amplitude exceeds certain threshold. The threshold is exceeded for limited duration periodically at different propagation distances, leading to multiple ionization injections and separated electron bunches. The method is demonstrated through multi-dimensional particle-in-cell simulations. Such electron bunches may be used to generate multi-chromatic X-ray sources for a variety of applications.

We report an overall enhancement of a laser wakefield acceleration (LWFA) using the ionization injection in a mixture of 0.3 % nitrogen gas in 99.7 % helium gas. Upon the interaction of 30 TW, 30 fs laser pulses with a gas jet of the above gas mixture, > 300 MeV electron beams were generated at a helium plasma densities of 3.3-8.5*10^18 cm^{-3}. Compared with the electron self-injection in pure helium gas jet, the ionization injection has led to the generation of electron beams with higher energies, higher charge, lower density threshold for trapping, and a narrower energy spread without dark current (low energy electrons) or multiple bunches. It is foreseen that further optimization of such a scheme is expected to bring the electron beam energy-spread down to 1 %, making them suitable for driving ultra-compact free-electron lasers