Accurate downlink channel information is crucial to the beamforming design, but it is difficult to obtain in practice. This paper investigates a deep learning-based optimization approach of the downlink beamforming to maximize the system sum rate, wh
en only the uplink channel information is available. Our main contribution is to propose a model-driven learning technique that exploits the structure of the optimal downlink beamforming to design an effective hybrid learning strategy with the aim to maximize the sum rate performance. This is achieved by jointly considering the learning performance of the downlink channel, the power and the sum rate in the training stage. The proposed approach applies to generic cases in which the uplink channel information is available, but its relation to the downlink channel is unknown and does not require an explicit downlink channel estimation. We further extend the developed technique to massive multiple-input multiple-output scenarios and achieve a distributed learning strategy for multicell systems without an inter-cell signalling overhead. Simulation results verify that our proposed method provides the performance close to the state of the art numerical algorithms with perfect downlink channel information and significantly outperforms existing data-driven methods in terms of the sum rate.
This paper studies fast adaptive beamforming optimization for the signal-to-interference-plus-noise ratio balancing problem in a multiuser multiple-input single-output downlink system. Existing deep learning based approaches to predict beamforming re
ly on the assumption that the training and testing channels follow the same distribution which may not hold in practice. As a result, a trained model may lead to performance deterioration when the testing network environment changes. To deal with this task mismatch issue, we propose two offline adaptive algorithms based on deep transfer learning and meta-learning, which are able to achieve fast adaptation with the limited new labelled data when the testing wireless environment changes. Furthermore, we propose an online algorithm to enhance the adaptation capability of the offline meta algorithm in realistic non-stationary environments. Simulation results demonstrate that the proposed adaptive algorithms achieve much better performance than the direct deep learning algorithm without adaptation in new environments. The meta-learning algorithm outperforms the deep transfer learning algorithm and achieves near optimal performance. In addition, compared to the offline meta-learning algorithm, the proposed online meta-learning algorithm shows superior adaption performance in changing environments.
People live in a 3D world. However, existing works on person re-identification (re-id) mostly consider the semantic representation learning in a 2D space, intrinsically limiting the understanding of people. In this work, we address this limitation by
exploring the prior knowledge of the 3D body structure. Specifically, we project 2D images to a 3D space and introduce a novel parameter-efficient Omni-scale Graph Network (OG-Net) to learn the pedestrian representation directly from 3D point clouds. OG-Net effectively exploits the local information provided by sparse 3D points and takes advantage of the structure and appearance information in a coherent manner. With the help of 3D geometry information, we can learn a new type of deep re-id feature free from noisy variants, such as scale and viewpoint. To our knowledge, we are among the first attempts to conduct person re-identification in the 3D space. We demonstrate through extensive experiments that the proposed method (1) eases the matching difficulty in the traditional 2D space, (2) exploits the complementary information of 2D appearance and 3D structure, (3) achieves competitive results with limited parameters on four large-scale person re-id datasets, and (4) has good scalability to unseen datasets. Our code, models and generated 3D human data are publicly available at https://github.com/layumi/person-reid-3d .
Beamforming is evidently a core technology in recent generations of mobile communication networks. Nevertheless, an iterative process is typically required to optimize the parameters, making it ill-placed for real-time implementation due to high comp
lexity and computational delay. Heuristic solutions such as zero-forcing (ZF) are simpler but at the expense of performance loss. Alternatively, deep learning (DL) is well understood to be a generalizing technique that can deliver promising results for a wide range of applications at much lower complexity if it is sufficiently trained. As a consequence, DL may present itself as an attractive solution to beamforming. To exploit DL, this article introduces general data- and model-driven beamforming neural networks (BNNs), presents various possible learning strategies, and also discusses complexity reduction for the DL-based BNNs. We also offer enhancement methods such as training-set augmentation and transfer learning in order to improve the generality of BNNs, accompanied by computer simulation results and testbed results showing the performance of such BNN solutions.
This paper investigates the optimal transmit beamforming design of simultaneous wireless information and power transfer (SWIPT) in the multiuser multiple-input-single-output (MISO) downlink with specific absorption rate (SAR) constraints. We consider
the power splitting technique for SWIPT, where each receiver divides the received signal into two parts: one for information decoding and the other for energy harvesting with a practical non-linear rectification model. The problem of interest is to maximize as much as possible the received signal-to-interference-plus-noise ratio (SINR) and the energy harvested for all receivers, while satisfying the transmit power and the SAR constraints by optimizing the transmit beamforming at the transmitter and the power splitting ratios at different receivers. The optimal beamforming and power splitting solutions are obtained with the aid of semidefinite programming and bisection search. Low-complexity fixed beamforming and hybrid beamforming techniques are also studied. Furthermore, we study the effect of imperfect channel information and radiation matrices, and design robust beamforming to guarantee the worst-case performance. Simulation results demonstrate that our proposed algorithms can effectively deal with the radio exposure constraints and significantly outperform the conventional transmission scheme with power backoff.
Squeeze-and-Excitation (SE) block presents a channel attention mechanism for modeling global context via explicitly capturing dependencies across channels. However, we are still far from understanding how the SE block works. In this work, we first re
visit the SE block, and then present a detailed empirical study of the relationship between global context and attention distribution, based on which we propose a simple yet effective module, called Linear Context Transform (LCT) block. We divide all channels into different groups and normalize the globally aggregated context features within each channel group, reducing the disturbance from irrelevant channels. Through linear transform of the normalized context features, we model global context for each channel independently. The LCT block is extremely lightweight and easy to be plugged into different backbone models while with negligible parameters and computational burden increase. Extensive experiments show that the LCT block outperforms the SE block in image classification task on the ImageNet and object detection/segmentation on the COCO dataset with different backbone models. Moreover, LCT yields consistent performance gains over existing state-of-the-art detection architectures, e.g., 1.5$sim$1.7% AP$^{bbox}$ and 1.0$sim$1.2% AP$^{mask}$ improvements on the COCO benchmark, irrespective of different baseline models of varied capacities. We hope our simple yet effective approach will shed some light on future research of attention-based models.
Domain adaptation is an important technique to alleviate performance degradation caused by domain shift, e.g., when training and test data come from different domains. Most existing deep adaptation methods focus on reducing domain shift by matching m
arginal feature distributions through deep transformations on the input features, due to the unavailability of target domain labels. We show that domain shift may still exist via label distribution shift at the classifier, thus deteriorating model performances. To alleviate this issue, we propose an approximate joint distribution matching scheme by exploiting prediction uncertainty. Specifically, we use a Bayesian neural network to quantify prediction uncertainty of a classifier. By imposing distribution matching on both features and labels (via uncertainty), label distribution mismatching in source and target data is effectively alleviated, encouraging the classifier to produce consistent predictions across domains. We also propose a few techniques to improve our method by adaptively reweighting domain adaptation loss to achieve nontrivial distribution matching and stable training. Comparisons with state of the art unsupervised domain adaptation methods on three popular benchmark datasets demonstrate the superiority of our approach, especially on the effectiveness of alleviating negative transfer.
Beamforming is an effective means to improve the quality of the received signals in multiuser multiple-input-single-output (MISO) systems. Traditionally, finding the optimal beamforming solution relies on iterative algorithms, which introduces high c
omputational delay and is thus not suitable for real-time implementation. In this paper, we propose a deep learning framework for the optimization of downlink beamforming. In particular, the solution is obtained based on convolutional neural networks and exploitation of expert knowledge, such as the uplink-downlink duality and the known structure of optimal solutions. Using this framework, we construct three beamforming neural networks (BNNs) for three typical optimization problems, i.e., the signal-to-interference-plus-noise ratio (SINR) balancing problem, the power minimization problem, and the sum rate maximization problem. For the former two problems the BNNs adopt the supervised learning approach, while for the sum rate maximization problem a hybrid method of supervised and unsupervised learning is employed. Simulation results show that the BNNs can achieve near-optimal solutions to the SINR balancing and power minimization problems, and a performance close to that of the weighted minimum mean squared error algorithm for the sum rate maximization problem, while in all cases enjoy significantly reduced computational complexity. In summary, this work paves the way for fast realization of optimal beamforming in multiuser MISO systems.
This paper studies the performance of cache-enabled dense small cell networks consisting of multi-antenna sub-6 GHz and millimeter-wave base stations. Different from the existing works which only consider a single antenna at each base station, the op
timal content placement is unknown when the base stations have multiple antennas. We first derive the successful content delivery probability by accounting for the key channel features at sub-6 GHz and mmWave frequencies. The maximization of the successful content delivery probability is a challenging problem. To tackle it, we first propose a constrained cross-entropy algorithm which achieves the near-optimal solution with moderate complexity. We then develop another simple yet effective heuristic probabilistic content placement scheme, termed two-stair algorithm, which strikes a balance between caching the most popular contents and achieving content diversity. Numerical results demonstrate the superior performance of the constrained cross-entropy method and that the two-stair algorithm yields significantly better performance than only caching the most popular contents. The comparisons between the sub-6 GHz and mmWave systems reveal an interesting tradeoff between caching capacity and density for the mmWave system to achieve similar performance as the sub-6 GHz system.
We consider a full-duplex decode-and-forward system, where the wirelessly powered relay employs the time-switching protocol to receive power from the source and then transmit information to the destination. It is assumed that the relay node is equipp
ed with two sets of antennas to enable full-duplex communications. Three different interference mitigation schemes are studied, namely, 1) optimal 2) zero-forcing and 3) maximum ratio combining/maximum ratio transmission. We develop new outage probability expressions to investigate delay-constrained transmission throughput of these schemes. Our analysis show interesting performance comparisons of the considered precoding schemes for different system and link parameters.
