Huawei Technologies (Canada)

There is considerable pressure to define the key requirements of 5G, develop 5G standards, and perform technology trials as quickly as possible. Normally, these activities are best done in series but there is a desire to complete these tasks in parallel so that commercial deployments of 5G can begin by 2020. 5G will not be an incremental improvement over its predecessors; it aims to be a revolutionary leap forward in terms of data rates, latency, massive connectivity, network reliability, and energy efficiency. These capabilities are targeted at realizing high-speed connectivity, the Internet of Things, augmented virtual reality, the tactile internet, and so on. The requirements of 5G are expected to be met by new spectrum in the microwave bands (3.3-4.2 GHz), and utilizing large bandwidths available in mm-wave bands, increasing spatial degrees of freedom via large antenna arrays and 3-D MIMO, network densification, and new waveforms that provide scalability and flexibility to meet the varying demands of 5G services. Unlike the one size fits all 4G core networks, the 5G core network must be flexible and adaptable and is expected to simultaneously provide optimized support for the diverse 5G use case categories. In this paper, we provide an overview of 5G research, standardization trials, and deployment challenges. Due to the enormous scope of 5G systems, it is necessary to provide some direction in a tutorial article, and in this overview, the focus is largely user centric, rather than device centric. In addition to surveying the state of play in the area, we identify leading technologies, evaluating their strengths and weaknesses, and outline the key challenges ahead, with research test beds delivering promising performance but pre-commercial trials lagging behind the desired 5G targets.

Abstract The fifth generation (5G) wireless communication networks are being deployed worldwide from 2020 and more capabilities are in the process of being standardized, such as mass connectivity, ultra-reliability, and guaranteed low latency. However, 5G will not meet all requirements of the future in 2030 and beyond, and sixth generation (6G) wireless communication networks are expected to provide global coverage, enhanced spectral/energy/cost efficiency, better intelligence level and security, etc. To meet these requirements, 6G networks will rely on new enabling technologies, i.e., air interface and transmission technologies and novel network architecture, such as waveform design, multiple access, channel coding schemes, multi-antenna technologies, network slicing, cell-free architecture, and cloud/fog/edge computing. Our vision on 6G is that it will have four new paradigm shifts. First, to satisfy the requirement of global coverage, 6G will not be limited to terrestrial communication networks, which will need to be complemented with non-terrestrial networks such as satellite and unmanned aerial vehicle (UAV) communication networks, thus achieving a space-air-ground-sea integrated communication network. Second, all spectra will be fully explored to further increase data rates and connection density, including the sub-6 GHz, millimeter wave (mmWave), terahertz (THz), and optical frequency bands. Third, facing the big datasets generated by the use of extremely heterogeneous networks, diverse communication scenarios, large numbers of antennas, wide bandwidths, and new service requirements, 6G networks will enable a new range of smart applications with the aid of artificial intelligence (AI) and big data technologies. Fourth, network security will have to be strengthened when developing 6G networks. This article provides a comprehensive survey of recent advances and future trends in these four aspects. Clearly, 6G with additional technical requirements beyond those of 5G will enable faster and further communications to the extent that the boundary between physical and cyber worlds disappears.

Fifth generation (5G) mobile communication systems have entered the stage of commercial deployment, providing users with new services, improved user experiences as well as a host of novel opportunities to various industries. However, 5G still faces many challenges. To address these challenges, international industrial, academic, and standards organizations have commenced research on sixth generation (6G) wireless communication systems. A series of white papers and survey papers have been published, which aim to define 6G in terms of requirements, application scenarios, key technologies, etc. Although ITU-R has been working on the 6G vision and it is expected to reach a consensus on what 6G will be by mid-2023, the related global discussions are still wide open and the existing literature has identified numerous open issues. This paper first provides a comprehensive portrayal of the 6G vision, technical requirements, and application scenarios, covering the current common understanding of 6G. Then, a critical appraisal of the 6G network architecture and key technologies is presented. Furthermore, existing testbeds and advanced 6G verification platforms are detailed for the first time. In addition, future research directions and open challenges are identified to stimulate the on-going global debate. Finally, lessons learned to date concerning 6G networks are discussed.

Artificial neural networks thrive in solving the classification problem for a particular rigid task, acquiring knowledge through generalized learning behaviour from a distinct training phase. The resulting network resembles a static entity of knowledge, with endeavours to extend this knowledge without targeting the original task resulting in a catastrophic forgetting. Continual learning shifts this paradigm towards networks that can continually accumulate knowledge over different tasks without the need to retrain from scratch. We focus on task incremental classification, where tasks arrive sequentially and are delineated by clear boundaries. Our main contributions concern: (1) a taxonomy and extensive overview of the state-of-the-art; (2) a novel framework to continually determine the stability-plasticity trade-off of the continual learner; (3) a comprehensive experimental comparison of 11 state-of-the-art continual learning methods; and (4) baselines. We empirically scrutinize method strengths and weaknesses on three benchmarks, considering Tiny Imagenet and large-scale unbalanced iNaturalist and a sequence of recognition datasets. We study the influence of model capacity, weight decay and dropout regularization, and the order in which the tasks are presented, and qualitatively compare methods in terms of required memory, computation time, and storage.

Multicarrier CDMA is a multiplexing approach in which modulated QAM symbols are spread over multiple OFDMA tones by using a generally complex spreading sequence. Effectively, a QAM symbol is repeated over multiple tones. Low density signature (LDS) is a version of CDMA with low density spreading sequence allowing us to take advantage of a near optimal ML receiver with practically feasible complexity. In this paper, we propose a new multiple access scheme so called sparse code multiple access (SCMA) which still enjoys the low complexity reception technique but with better performance compared to LDS. In SCMA, the procedure of bit to QAM symbol mapping and spreading are combined together and incoming bits are directly mapped to a multidimensional codeword of an SCMA codebook set. Each layer or user has its dedicated codebook. Shaping gain of a multidimensional constellation is the main source of the performance improvement in comparison to the simple repetition of QAM symbols in LDS. In general, SCMA codebook design is an optimization problem. A systematic sub-optimal approach is proposed here for SCMA codebook design.

MTC are expected to play an essential role within future 5G systems. In the FP7 project METIS, MTC has been further classified into mMTC and uMTC. While mMTC is about wireless connectivity to tens of billions of machinetype terminals, uMTC is about availability, low latency, and high reliability. The main challenge in mMTC is scalable and efficient connectivity for a massive number of devices sending very short packets, which is not done adequately in cellular systems designed for human-type communications. Furthermore, mMTC solutions need to enable wide area coverage and deep indoor penetration while having low cost and being energy-efficient. In this article, we introduce the PHY and MAC layer solutions developed within METIS to address this challenge.

Non-IID data present a tough challenge for federated learning. In this paper, we explore a novel idea of facilitating pairwise collaborations between clients with similar data. We propose FedAMP, a new method employing federated attentive message passing to facilitate similar clients to collaborate more. We establish the convergence of FedAMP for both convex and non-convex models, and propose a heuristic method to further improve the performance of FedAMP when clients adopt deep neural networks as personalized models. Our extensive experiments on benchmark data sets demonstrate the superior performance of the proposed methods.

Multicarrier CDMA is a multiple access scheme in which modulated QAM symbols are spread over OFDMA tones by using a generally complex spreading sequence. Effectively, a QAM symbol is repeated over multiple tones. Low density signature (LDS) is a version of CDMA with low density spreading sequences allowing us to take advantage of a near optimal message passing algorithm (MPA) receiver with practically feasible complexity. Sparse code multiple access (SCMA) is a multi-dimensional codebook-based non-orthogonal spreading technique. In SCMA, the procedure of bit to QAM symbol mapping and spreading are combined together and incoming bits are directly mapped to multi-dimensional codewords of SCMA codebook sets. Each layer has its dedicated codebook. Shaping gain of a multi-dimensional constellation is one of the main sources of the performance improvement in comparison to the simple repetition of QAM symbols in LDS. Meanwhile, like LDS, SCMA enjoys the low complexity reception techniques due to the sparsity of SCMA codewords. In this paper a systematic approach is proposed to design SCMA codebooks mainly based on the design principles of lattice constellations. Simulation results are presented to show the performance gain of SCMA compared to LDS and OFDMA.

Recently data-driven image inpainting methods have made inspiring progress, impacting fundamental image editing tasks such as object removal and damaged image repairing. These methods are more effective than classic approaches, however, due to memory limitations they can only handle low-resolution inputs, typically smaller than 1K. Meanwhile, the resolution of photos captured with mobile devices increases up to 8K. Naive up-sampling of the low-resolution inpainted result can merely yield a large yet blurry result. Whereas, adding a high-frequency residual image onto the large blurry image can generate a sharp result, rich in details and textures. Motivated by this, we propose a Contextual Residual Aggregation (CRA) mechanism that can produce high-frequency residuals for missing contents by weighted aggregating residuals from contextual patches, thus only requiring a low-resolution prediction from the network. Since convolutional layers of the neural network only need to operate on low-resolution inputs and outputs, the cost of memory and computing power is thus well suppressed. Moreover, the need for high-resolution training datasets is alleviated. In our experiments, we train the proposed model on small images with resolutions 512 × 512 and perform inference on high-resolution images, achieving compelling inpainting quality. Our model can inpaint images as large as 8K with considerable hole sizes, which is intractable with previous learning-based approaches. We further elaborate on the light-weight design of the network architecture, achieving real-time performance on 2K images on a GTX 1080 Ti GPU. Codes are available at: https://github. com/Ascend-Huawei/Ascend-Canada/tree/ master/Models/Research_HiFIll_Model.

A spectrally-localized waveform is proposed based on filtered orthogonal frequency division multiplexing (f-OFDM). By allowing the filter length to exceed the cyclic prefix (CP) length of OFDM and designing the filter appropriately, the proposed f-OFDM waveform can achieve a desirable frequency localization for bandwidths as narrow as a few tens of subcarriers, while keeping the inter-symbol interference/inter-carrier interference (ISI/ICI) within an acceptable limit. Enabled by the proposed f-OFDM, an asynchronous filtered orthogonal frequency division multiple access (f-OFDMA)/filtered discrete-Fourier transform-spread OFDMA (f-DFT-S-OFDMA) scheme is introduced, which uses the spectrum shaping filter at each transmitter for side lobe leakage elimination and a bank of filters at the receiver for inter-user interference rejection. Per-user downsampling and short fast Fourier transform (FFT) are used at the receiver to ensure a reasonable complexity of implementation. The proposed scheme removes the inter-user time-synchronization overhead required in the synchronous OFDMA/DFT-S-OFDMA. The performance of the asynchronous f-OFDMA is evaluated and compared with that of the universal-filtered OFDM (UF-OFDM), proposed in [1], [2].

The integration of communications with different scales, diverse radio access technologies, and various network resources renders next-generation wireless networks (NGWNs) highly heterogeneous and dynamic. Emerging use cases and applications, such as machine to machine communications, autonomous driving, and factory automation, have stringent requirements in terms of reliability, latency, throughput, and so on. Such requirements pose new challenges to architecture design, network management, and resource orchestration in NGWNs. Starting from illustrating these challenges, this paper aims at providing a good understanding of the overall architecture of NGWNs and three specific research problems under this architecture. First, we introduce a network-slicing based architecture and explain why and where artificial intelligence (AI) should be incorporated into this architecture. Second, the motivation, research challenges, existing works, and potential future directions related to applying AI-based approaches in three research problems are described in detail, i.e., flexible radio access network slicing, automated radio access technology selection, and mobile edge caching and content delivery. In summary, this paper highlights the benefits and potentials of AI-based approaches in the research of NGWNs.

Multimodallarge language models (MLLMs) have emerged as a prominent area of interest within the research community, given their proficiency in handling and reasoning with non-textual data, including images and videos. This study seeks to extend the application of MLLMs to the realm of autonomous driving by introducing DriveGPT4, a novel interpretable end-to-end autonomous driving system based on LLMs. Capable of processing multi-frame video inputs and textual queries, DriveGPT4 facilitates the interpretation of vehicle actions, offers pertinent reasoning, and effectively addresses a diverse range of questions posed by users. Furthermore, DriveGPT4 predicts low-level vehicle control signals in an end-to-end fashion. These advanced capabilities are achieved through the utilization of a bespoke visual instruction tuning dataset, specifically tailored for autonomous driving applications, in conjunction with a mix-finetuning training strategy. DriveGPT4 represents the pioneering effort to leverage LLMs for the development of an interpretable end-to-end autonomous driving solution. Evaluations conducted on the BDD-X dataset showcase the superior qualitative and quantitative performance of DriveGPT4. Additionally, the fine-tuning of domain-specific data enables DriveGPT4 to yield close or even improved results in terms of autonomous driving grounding when contrasted with GPT4-V.

We consider the interference management problem in a multicell MIMO heterogeneous network. Within each cell there is a large number of distributed micro/pico base stations (BSs) that can be potentially coordinated for joint transmission. To reduce coordination overhead, we consider user-centric BS clustering so that each user is served by only a small number of (potentially overlapping) BSs. Thus, given the channel state information, our objective is to jointly design the BS clustering and the linear beamformers for all BSs in the network. In this paper, we formulate this problem from a {sparse optimization} perspective, and propose an efficient algorithm that is based on iteratively solving a sequence of group LASSO problems. A novel feature of the proposed algorithm is that it performs BS clustering and beamformer design jointly rather than separately as is done in the existing approaches for partial coordinated transmission. Moreover, the cluster size can be controlled by adjusting a single penalty parameter in the nonsmooth regularized utility function. The convergence of the proposed algorithm (to a stationary solution) is guaranteed, and its effectiveness is demonstrated via extensive simulation.

Fifth generation (5G) wireless networks are expected to support very diverse applications and terminals. Massive connectivity with a large number of devices is an important requirement for 5G networks. Current LTE system is not able to efficiently support massive connectivity, especially on the uplink (UL). Among the issues that arise due to massive connectivity is the cost of signaling overhead and latency. In this paper, an uplink contention-based sparse code multiple access (SCMA) design is proposed as a solution. First, the system design aspects of the proposed multiple-access scheme are described. The SCMA parameters can be adjusted to provide different levels of overloading, thus suitable to meet the diverse traffic connectivity requirements. In addition, the system-level evaluations of a small packet application scenario are provided for contention-based UL SCMA. SCMA is compared to OFDMA in terms of connectivity and drop rate under a tight latency requirement. The simulation results demonstrate that contention-based SCMA can provide around 2.8 times gain over contention-based OFDMA in terms of supported active users. The uplink contention-based SCMA scheme can be a promising technology for 5G wireless networks for data transmission with low signaling overhead, low delay, and support of massive connectivity.

The underlying waveform has always been a shaping factor for each generation of the cellular networks, such as orthogonal frequency division multiplexing (OFDM) for the 4th generation cellular networks (4G). To meet the diversified and pronounced expectations upon the upcoming 5G cellular networks, here we present an enabler for flexible waveform configuration, named as filtered-OFDM (f-OFDM). With the conventional OFDM, a unified numerology is applied across the bandwidth provided, balancing among the channel characteristics and the service requirements, and the spectrum efficiency is limited by the compromise we made. In contrast, with f-OFDM, the assigned bandwidth is split up into several subbands, and different types of services are accommodated in different subbands with the most suitable waveform and numerology, leading to an improved spectrum utilization. After outlining the general framework of f-OFDM, several important design aspects are also discussed, including filter design and guard tone arrangement. In addition, an extensive comparison among the existing 5G waveform candidates is also included to illustrate the advantages of f-OFDM. Our simulations indicate that, in a specific scenario with four distinct types of services, f-OFDM provides up to 46% of throughput gains over the conventional OFDM scheme.

Split learning (SL) is a collaborative learning framework, which can train an artificial intelligence (AI) model between a device and an edge server by splitting the AI model into a device-side model and a server-side model at a cut layer. The existing SL approach conducts the training process sequentially across devices, which incurs significant training latency especially when the number of devices is large. In this paper, we design a novel SL scheme to reduce the training latency, named <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">C</u> luster-based <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">P</u> arallel <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SL</u> (CPSL) which conducts model training in a “first-parallel-then-sequential” manner. Specifically, the CPSL is to partition devices into several clusters, parallelly train device-side models in each cluster and aggregate them, and then sequentially train the whole AI model across clusters, thereby parallelizing the training process and reducing training latency. Furthermore, we propose a resource management algorithm to minimize the training latency of CPSL considering device heterogeneity and network dynamics in wireless networks. This is achieved by stochastically optimizing the cut layer selection, device clustering, and radio spectrum allocation. The proposed two-timescale algorithm can jointly make the cut layer selection decision in a large timescale and device clustering and radio spectrum allocation decisions in a small timescale. Extensive simulation results on non-independent and identically distributed data demonstrate that the proposed solution can greatly reduce the training latency as compared with the existing SL benchmarks, while adapting to network dynamics.

Autonomous robotic systems and self driving cars rely on accurate perception of their surroundings as the safety of the passengers and pedestrians is the top priority. Semantic segmentation is one of the essential components of road scene perception that provides semantic information of the surrounding environment. Recently, several methods have been introduced for 3D LiDAR semantic segmentation. While they can lead to improved performance, they are either afflicted by high computational complexity, therefore are inefficient, or they lack fine details of smaller object instances. To alleviate these problems, we propose (AF) <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> -S3Net, an end-to-end encoder-decoder CNN network for 3D LiDAR semantic segmentation. We present a novel multibranch attentive feature fusion module in the encoder and a unique adaptive feature selection module with feature map re-weighting in the decoder. Our (AF) <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> -S3Net fuses the voxel-based learning and point-based learning methods into a unified framework to effectively process the potentially large 3D scene. Our experimental results show that the proposed method outperforms the state-of-the-art approaches on the large-scale nuScenes-lidarseg and SemanticKITTI benchmark, ranking 1 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">st</sup> on both competitive public leaderboard competitions upon publication.

Existing deep learning-enabled semantic communication systems often rely on shared background knowledge between the transmitter and receiver that includes empirical data and their associated semantic information. In practice, the semantic information is defined by the pragmatic task of the receiver and cannot be known to the transmitter. The actual observable data at the transmitter can also have non-identical distribution with the empirical data in the shared background knowledge library. To address these practical issues, this paper proposes a new neural network-based semantic communication system for image transmission, where the task is unaware at the transmitter and the data environment is dynamic. The system consists of two main parts, namely the semantic coding (SC) network and the data adaptation (DA) network. The SC network learns how to extract and transmit the semantic information using a receiver-leading training process. By using the domain adaptation technique from transfer learning, the DA network learns how to convert the data observed into a similar form of the empirical data that the SC network can process without re-training. Numerical experiments show that the proposed method can be adaptive to observable datasets while keeping high performance in terms of both data recovery and task execution.

In this work, we introduce β-expansion, a notion borrowed from number theory, as a theoretical framework to study fast construction of polar codes based on a recursive structure of universal partial order (UPO) and polarization weight (PW) algorithm. We show that polar codes can be recursively constructed from UPO by continuously solving several polynomial equations at each recursive step. From these polynomial equations, we can extract an interval for β, such that ranking the synthetic channels through a closed- form β-expansion preserves the property of nested frozen sets, which is a desired feature for low- complex construction. In an example of AWGN channels, we show that this interval for β converges to a constant close to 1.1892 when the code block-length trends to infinity. Both asymptotic analysis and simulation results validate our theoretical claims.

Autonomous driving has achieved significant milestones in research and development over the last two decades. There is increasing interest in the field as the deployment of autonomous vehicles (AVs) promises safer and more ecologically friendly transportation systems. With the rapid progress in computationally powerful artificial intelligence (AI) techniques, AVs can sense their environment with high precision, make safe real-time decisions, and operate reliably without human intervention. However, intelligent decision-making in such vehicles is not generally understandable by humans in the current state of the art, and such deficiency hinders this technology from being socially acceptable. Hence, aside from making safe real-time decisions, AVs must also explain their AI-guided decision-making process in order to be regulatory-compliant across many jurisdictions. Our study sheds comprehensive light on the development of explainable artificial intelligence (XAI) approaches for AVs. In particular, we make the following contributions. First, we provide a thorough overview of the state-of-the-art and emerging approaches for XAI-based autonomous driving. We then propose a conceptual framework considering the essential elements for explainable end-to-end autonomous driving. Finally, we present XAI-based prospective directions and emerging paradigms for future directions that hold promise for enhancing transparency, trustworthiness, and societal acceptance of AVs.