A cell-free massive multiple-input multiple-output (MIMO) uplink is considered, where quantize-and-forward (QF) refers to the case where both the channel estimates and the received signals are quantized at the access points (APs) and forwarded to a central processing unit (CPU) whereas in combinequantize- and-forward (CQF), the APs send the quantized version of the combined signal to the CPU. To solve the non-convex sum rate maximization problem, a heuristic sub-optimal scheme is exploited to convert the power allocation problem into a standard geometric programme (GP). We exploit the knowledge of the channel statistics to design the power elements. Employing largescale-fading (LSF) with a deep convolutional neural network (DCNN) enables us to determine a mapping from the LSF coefficients and the optimal power through solving the sum rate maximization problem using the quantized channel. Four possible power control schemes are studied, which we refer to as i) small-scale fading (SSF)-based QF; ii) LSF-based CQF; iii) LSF use-and-then-forget (UatF)-based QF; and iv) LSF deep learning (DL)-based QF, according to where channel estimation is performed and exploited and how the optimization problem is solved. Numerical results show that for the same fronthaul rate, the throughput significantly increases thanks to the mapping obtained using DCNN.
Energy efficiency has become an important aspect of wireless communication, both economically and environmentally. This letter investigates the energy efficiency of downlink AWGN channels by employing multiple decoding policies. The overall energy efficiency of the system is based on the bits-per-joule metric, where energy efficiency contours are used to locate the optimal operating points based on the system requirements. Our novel approach uses a linear power model to define the total power consumed at the base station, encompassing the circuit and processing power, and amplifier efficiency, and ensures that the best energy efficiency value can be achieved whilst satisfying other system targets such as QoS and rate-fairness.
We consider a cell-free massive multiple-input multiple-output (MIMO) system where the channel estimates and the received signals are quantized at the access points (APs) and forwarded to a central processing unit (CPU). Zero-forcing technique is used at the CPU to detect the signals transmitted from all users.. To solve the non-convex sum rate maximization problem, a heuristic sub-optimal scheme is proposed to convert the problem into a geometric programme (GP). Exploiting a deep convolutional neural network (DCNN) allows us to determine both a mapping from the large-scale fading (LSF) coefficients and the optimal power by solving the optimization problem using the quantized channel. Depending on how the optimization problem is solved, different power control schemes are investigated; i) small-scale fading (SSF)-based power control; ii) LSF use-and-then-forget (UatF)-based power control; and iii) LSF deep learning (DL)-based power control. The SSF-based power control scheme needs to be solved for each coherence interval of the SSF, which is practically impossible in real time systems. Numerical results reveal that the proposed LSF-DL-based scheme significantly increases the performance compared to the practical and well-known LSF-UatF-based power control, thanks to the mapping obtained using DCNN.
Energy consumption has become an increasingly important aspect of wireless communications, from both an economical and environmental point of view. New enhancements are being placed on mobile networks to reduce the power consumption of both mobile terminals and base stations. This paper studies the achievable rate region of AWGN broadcast channels under Time-division, Frequency-division and Superposition coding, and locates the optimal energy-efficient rate-pair according to a comparison metric based on the average energy efficiency of the system. In addition to the transmit power, circuit power and signalling power are also incorporated in the energy efficiency function, with simulation results verifying that the Superposition coding scheme achieves the highest energy efficiency in an ideal, but non-realistic scenario, where the signalling power is zero. With moderate signalling power, the Frequency-division scheme is the most energy-efficient, with Superposition coding and Time-division becoming second and third best. Conversely, when the signalling power is high, both Time-division and Frequency-division schemes outperform Superposition coding. On the other hand, the Superposition coding scheme also incorporates rate-fairness into the system, which allows both users to transmit whilst maximising the energy efficiency.
This letter addresses energy-efficient design in multi-user, single-carrier uplink channels by employing multiple decoding policies. The comparison metric used in this study is based on average energy efficiency contours, where an optimal rate vector is obtained based on four system targets: Maximum energy efficiency, a trade-off between maximum energy efficiency and rate fairness, achieving energy efficiency target with maximum sum-rate and achieving energy efficiency target with fairness. The transmit power function is approximated using Taylor series expansion, with simulation results demonstrating the achievability of the optimal rate vector, and negligible performance difference in employing this approximation.
To counteract spoofing attacks, the majority of recent approaches to face spoofing attack detection formulate the problem as a binary classification task in which real data and attack-accesses are both used to train spoofing detectors. Although the classical training framework has been demonstrated to deliver satisfactory results, its robustness to unseen attacks is debatable. Inspired by the recent success of anomaly detection models in face spoofing detection, we propose an ensemble of one-class classifiers fused by a Stacking ensemble method to reduce the generalisation error in the more realistic unseen attack scenario. To be consistent with this scenario, anomalous samples are considered neither for training the component anomaly classifiers nor for the design of the Stacking ensemble. To achieve better face-anti spoofing results, we adopt client-specific information to build both constituent classifiers as well as the Stacking combiner. Besides, we propose a novel 2-stage Genetic Algorithm to further improve the generalisation performance of Stacking ensemble. We evaluate the effectiveness of the proposed systems on publicly available face anti-spoofing databases including Replay-Attack, Replay-Mobile and Rose-Youtu. The experimental results following the unseen attack evaluation protocol confirm the merits of the proposed model.
Recently, impressively growing efforts have been devoted to the challenging task of facial age estimation. The improvements in performance achieved by new algorithms are measured on several benchmarking test databases with different characteristics to check on consistency. While this is a valuable methodology in itself, a significant issue in the most age estimation related studies is that the reported results lack an assessment of intrinsic system uncertainty. Hence, a more in-depth view is required to examine the robustness of age estimation systems in different scenarios. The purpose of this paper is to conduct an evaluative and comparative analysis of different age estimation systems to identify trends, as well as the points of their critical vulnerability. In particular, we investigate four age estimation systems, including the online Microsoft service, two best state-of-the-art approaches advocated in the literature, as well as a novel age estimation algorithm. We analyse the effect of different internal and external factors, including gender, ethnicity, expression, makeup, illumination conditions, quality and resolution of the face images, on the performance of these age estimation systems. The goal of this sensitivity analysis is to provide the biometrics community with the insight and understanding of the critical subject-, camera- and environmental-based factors that affect the overall performance of the age estimation system under study.
—Existing studies in facial age estimation have mostly focused on intra-dataset protocols that assume training and test images captured under similar conditions. However, this is rarely valid in practical applications, where training and test sets usually have different characteristics. In this paper, we advocate a cross-dataset protocol for age estimation benchmarking. In order to improve the cross-dataset age estimation performance, we mitigate the inherent bias caused by the learning algorithm itself. To this end, we propose a novel loss function that is more effective for neural network training. The relative smoothness of the proposed loss function is its advantage with regards to the optimisation process performed by stochastic gradient descent. Its lower gradient, compared with existing loss functions, facilitates the discovery of and convergence to a better optimum, and consequently a better generalisation. The crossdataset experimental results demonstrate the superiority of the proposed method over the state-of-the-art algorithms in terms of accuracy and generalisation capability.
—Label distribution Learning (LDL) is the state-of-the-art approach to deal with a number of real-world applications , such as chronological age estimation from a face image, where there is an inherent similarity among adjacent age labels. LDL takes into account the semantic similarity by assigning a label distribution to each instance. The well-known Kullback–Leibler (KL) divergence is the widely used loss function for the LDL framework. However, the KL divergence does not fully and effectively capture the semantic similarity among age labels, thus leading to the sub-optimal performance. In this paper, we propose a novel loss function based on optimal transport theory for the LDL-based age estimation. A ground metric function plays an important role in the optimal transport formulation. It should be carefully determined based on underlying geometric structure of the label space of the application in-hand. The label space in the age estimation problem has a specific geometric structure, i.e. closer ages have more inherent semantic relationship. Inspired by this, we devise a novel ground metric function, which enables the loss function to increase the influence of highly correlated ages; thus exploiting the semantic similarity among ages more effectively than the existing loss functions. We then use the proposed loss function, namely γ–Wasserstein loss, for training a deep neural network (DNN). This leads to a notoriously computationally expensive and non-convex optimisa-tion problem. Following the standard methodology, we formulate the optimisation function as a convex problem and then use an efficient iterative algorithm to update the parameters of the DNN. Extensive experiments in age estimation on different benchmark datasets validate the effectiveness of the proposed method, which consistently outperforms state-of-the-art approaches.
Good generalisation performance is the fundamental goal of any machine learning algorithm. Using the uniform stability concept, this paper theoretically proves that the choice of loss function impacts on the generalisation performance of a trained deep neural network (DNN). The adopted stability based framework provides an effective tool for comparing the generalisation error bound with respect to the utilised loss function. The main result of our analysis is that using an effective loss function makes stochastic gradient descent more stable which consequently leads to the tighter generalisation error bound, and so better generalisation performance. To validate our analysis, we study learning problems in which the classes are semantically correlated. To capture this semantic similarity of neighbouring classes, we adopt the well-known semantics-preserving learning framework, namely label distribution learning (LDL). We propose two novel loss functions for the LDL framework and theoretically show that they provide stronger stability than the other widely used loss functions adopted for training DNNs. The experimental results on three applications with semantically correlated classes, including facial age estimation, head pose estimation and image aesthetic assessment, validate the theoretical insights gained by our analysis and demonstrate the usefulness of the proposed loss functions in practical applications.
Energy savings are becoming a global trend, hence the importance of energy efficiency (EE) as an alternative performance evaluation metric. This paper proposes an EE based resource allocation method for the broadcast channel (BC), where a linear power model is used to characterize the power consumed at the base station (BS). Having formulated our EE based optimization problem and objective function, we utilize standard convex optimization techniques to show the concavity of the latter, and thus, the existence of a unique globally optimal energy-efficient rate and power allocation. Our EE based resource allocation framework is also extended to incorporate fairness, and provide a minimum user satisfaction in terms of spectral efficiency (SE). We then derive the generic equation of the EE contours and use them to get insights about the EE-SE trade-off over the BC. The performances of the aforementioned resource allocation schemes are compared for different metrics against the number of users and cell radius. Results indicate that the highest EE improvement is achieved by using the unconstrained optimization scheme, which is obtained by significantly reducing the total transmit power. Moreover, the network EE is shown to increase with the number of users and decrease as the cell radius increases.
n this paper we present a new image sensor architecture for fast and accurate compressive sensing (CS) of natural images. Measurement matrices usually employed in compressive sensing CMOS image sensors (CS-CIS) are recursive pseudo-random binary matrices. We have proved that the restricted isometry property (RIP) of these matrices is limited by a low sparsity constant. The quality of these matrices is also affected by the non-idealities of pseudo-random numbers generators (PRNG). To overcome these limitations, we propose a hardware-friendly pseudo-random ternary measurement matrix generated on-chip by means of class III elementary cellular automata (ECA). These ECA present a chaotic behaviour that emulates random CS measurement matrices better than other PRNG. We have combined this new architecture with a block-based CS smoothed-projected Landweber (BCS-SPL) reconstruction algorithm. By means of single value decomposition (SVD) we have adapted this algorithm to perform fast and precise reconstruction while operating with binary and ternary matrices. Simulations are provided to qualify the approach
Energy efficiency has become increasingly important in wireless communications, with significant environmental and financial benefits. This paper studies the achievable capacity region of a single carrier uplink channel consisting of two transmitters and a single receiver, and uses average energy efficiency contours to find the optimal rate pair based on four different targets: Maximum energy efficiency, a trade-off between maximum energy efficiency and rate fairness, achieving energy efficiency target with maximum sum-rate and achieving energy efficiency target with fairness. In addition to the transmit power, circuit power is also accounted for, with the maximum transmit power constrained to a fixed value. Simulation results demonstrate the achievability of the optimal energy-efficient rate pair within the capacity region, and provide the trade-off for energy efficiency, fairness and maximum sum-rate.
Joint sparse representation (JSR) model has recently emerged as a powerful technique with wide variety of applications. In this paper, the JSR model is extended to error concealment (EC) application, being effective to recover the original image from its corrupted version. This model is based on jointly learning a dictionary pair and two mapping matrices that are trained offline from external training images. Given the trained dictionaries and mappings, the restoration is done by transferring the recovery problem into the sparse representation domain with respect to the trained dictionaries, which is further transformed into a common space using the respective mapping matrices. Then, the reconstructed image is obtained by back projection into the spatial domain. In order to improve the accuracy and stability of the proposed JSR-based EC algorithm and avoid unexpected artifacts, the local and non-local priors are seamlessly integrated into the JSR model. The non-local prior is based on the self-similarity within natural images and helps to find an accurate sparse representation by taking a weighted average of similar areas throughout the image. The local prior is based on learning the local structural regularity of the natural images and helps to regularize the sparse representation, exploiting the strong correlation in the small local areas within the image. Compared with the state-of-the-art EC algorithms, the results show that the proposed method has better reconstruction performance in terms of objective and subjective evaluations.