In this paper, we explore a cooperative integrated sensing and communication (ISAC) framework that utilizes orthogonal frequency division multiplexing (OFDM) waveforms. Under the control of a central processing unit (CPU), multiple access points (APs) collaboratively perform multistatic sensing while providing communication service in a cell-free multiple-input multiple-output (MIMO) system. Achieving high sensing accuracy requires the collection of global sensing information at the CPU, which can lead to significant fronthaul signaling overhead due to the feedback of the sensing signals from each AP. To tackle this issue, we propose a collaborative processing scheme in which the APs locally compress and quantize the received sensing signals before forwarding them to the CPU. The CPU then aggregates the information from all APs to estimate the location and velocity of the targets. We develop a distributed vector-quantized variational autoencoder (D-VQVAE) to enable an end-to-end implementation of this scheme. D-VQVAE consists of distributed encoders at the APs to locally encode the received sensing signals, codebooks for quantizing the encoded results, and a decoder at the CPU for location and velocity estimation. It effectively reduces the amount of data transmitted from each AP to the CPU while maintaining a high sensing accuracy.We employ a collaborative learning-assisted scheme to train D-VQVAE in an end-to-end manner. Simulation results show that the proposed D-VQVAE network outperforms the baseline schemes in sensing accuracy and reduces fronthaul signaling overhead by 99% when compared with the centralized sensing approach.

In this paper, we present a cooperative integrated sensing and communication (ISAC) framework in cell-free multiple-input multiple-output (MIMO) systems, where multiple access points (APs), under the control of a central processing unit (CPU), collaboratively perform target sensing by using the reflected echo signals. Most of the existing works first estimate the sensing parameters (e.g., range, angle, relative velocity) observed by each AP and then use these estimated parameters for sensing tasks such as localization and velocity estimation. However, this approach may suffer from performance degradation due to errors in the estimated parameters. We propose a deep neural network (DNN)-based scheme to jointly process the echo signals received across the distributed APs and directly estimate the location and velocity of the targets. The proposed scheme bypasses the sensing parameter estimation stage and enhances the sensing performance. Simulation results show that our proposed scheme significantly reduces the localization and velocity estimation error when compared with a state-of-the-art approach.

Integrated sensing and communication (ISAC) is one of the usage scenarios for the sixth generation (6G) wireless networks. In this paper, we study cooperative ISAC in cell-free multiple-input multiple-output (MIMO) systems, where multiple MIMO access points (APs) collaboratively provide communication services and perform multistatic sensing. We formulate an optimization problem for the ISAC beamforming design, which maximizes the achievable sum-rate while guaranteeing the sensing signal-to-noise ratio (SNR) requirement and total power constraint. Learning-based techniques are regarded as a promising approach for addressing such a nonconvex optimization problem. By taking the topology of cell-free MIMO systems into consideration, we propose a heterogeneous graph neural network (GNN), namely SACGNN, for ISAC beamforming design. The proposed SACGNN framework models the cell-free MIMO system for cooperative ISAC as a heterogeneous graph and employs a transformer-based heterogeneous message passing scheme to capture the important information of sensing and communication channels and propagate the information through the graph network. Simulation results demonstrate the performance gain of the proposed SACGNN framework over a conventional null-space projection based scheme and a deep neural network (DNN)-based baseline scheme.

Wireless traffic prediction is indispensable for network planning and resource management. Due to different population distributions and user behavior, there exist strong spatial-temporal variations in wireless traffic across different regions. Most of the conventional traffic prediction approaches can only tackle a particular spatial-temporal pattern and cannot capture such variations in wireless traffic. This motivates us to develop an adaptive approach which can tackle spatial-temporal variations and predict wireless traffic in different regions. In this paper, we formulate an adaptive traffic prediction problem from a probabilistic inference perspective and develop a variational spatial-temporal Bayesian meta-learning (VST-BML) algorithm. We model the traffic prediction in different regions as different prediction tasks. The proposed VST-BML algorithm can learn the common spatial-temporal features shared by all prediction tasks, and adaptively infer the task-specific parameters to tackle spatial-temporal variations. We evaluate the performance of our proposed VST-BML algorithm using a real-world traffic dataset. Experimental results show that the proposed algorithm can quickly adapt to different prediction tasks by using only a small number of data samples and provide accurate traffic prediction in different regions. When compared with five baseline methods, the proposed algorithm can reduce the root- mean-square error (RMSE) and mean absolute error (MAE) by 53.0% and 48.4%, respectively.

Integrated sensing and communications (ISAC) has emerged as a promising technology for predictive beamforming design in vehicle-to-infrastructure (V2I) networks. Most of the existing works use a two-step approach for predictive beamforming design. The first step is to estimate the state parameters of a vehicle (e.g., angle of arrival (AoA), channel state information (CSI)) from the received sensing signal samples at the road side unit (RSU). The second step is to determine the beamforming vector based on the estimated parameters. However, estimation errors may be introduced in the first step which impacts the subsequent beamforming design and leads to degradation in the achievable rate. In this work, by using deep learning, we propose an ISAC-enabled end-to-end predictive beamforming (E2E-PB) approach to obtain the beamforming vector directly from the reflected signal samples. The proposed approach does not require an intermediate state parameters estimation step. We develop an attention-based long short-term memory (LSTM) network to capture the temporal correlation in the reflected signal samples and determine the beamformer. The network is trained in an unsupervised manner to maximize the achievable rate. We compare our proposed E2E-PB approach with two state-of-the-art schemes, namely, the extended Kalman filtering framework and a deep learning based two-step approach. The results show that our proposed E2E-PB approach obtains a higher achievable rate than the other two baseline schemes, and has close performance when compared with the optimal beamforming design with perfect CSI.

Predictive analysis on cellular traffic is important for the control and monitoring of wireless networks. Cellular traffic prediction is a challenging problem due to the non- stationarity and dynamic spatial-temporal correlation of the traffic. In this paper, we address the problem of accurate traffic prediction in a base station by proposing a deep neural network called RAConv. Its structure includes residual network, attention mechanism, and deep convolutional network. In the proposed architecture, a deep 3D residual convolutional network (ResConv3D) with three residual blocks are employed to learn the local spatial-temporal features. An attention-aided convolutional long short-term memory network (AConvLSTM) is then used to capture the long-term spatial-temporal dependencies. The use of the attention modules enable the network to focus on the most important spatial-temporal information. We evaluate the performance of the proposed RAConv network using a dataset provided by a Canadian wireless service provider. We consider the traffic prediction on two time scales (i.e., hourly and daily), which exhibit different spatial-temporal dependency patterns. Experimental results show that the proposed RAConv network can achieve accurate prediction under both time scales. Results also show that our proposed network provides a lower root-mean-square error (RMSE) than the conventional ConvLSTM baseline scheme.

Cell-free millimeter-wave (mmWave) multiple-input multiple-output (MIMO) communications have been proposed as promising enablers for the next generation wireless networks. In this paper, we study the user association and hybrid beamforming in cell-free mmWave systems without full channel state information (CSI) acquisition. We consider a cloud radio access network (C-RAN), where multiple remote radio heads (RRHs) are distributed to communicate with users via analog beamforming, and connected to a centralized baseband unit (BBU) through fronthaul links which executes digital beamforming. We aim to jointly design user association, hybrid beamforming, and fronthaul compression with the aid of uplink training. A train-and-design framework is developed to achieve this goal. In particular, we first propose a two-stage uplink training approach to assist RRH-level design, during which the analog beamforming and user association are obtained. After that, digital beamforming and fronthaul compression are optimized at BBU based on the training results. Two performance metrics are considered in this paper, i.e. weighted sum-rate maximization and max-min fairness. Simulation results demonstrate the effectiveness of the proposed train-and-design framework for both sum-rate maximization and max-min fairness performance metrics. It is shown that the proposed algorithms can achieve comparable performance to the full-digital beamformer.

Mixed-numerology has been proposed in the Third Generation Partnership Project (3GPP) standard for the fifth generation (5G) wireless networks, where flexible subcarrier spacing (SCS) can be applied to support uses cases with different quality-of-service (QoS) requirements. In this paper, we study the joint design of resource block allocation and beamforming with mixed-numerology for enhanced mobile broadband (eMBB) and ultra-reliable low-latency communications (URLLC) use cases. We consider multiple multi-antenna base stations (BSs) cooperatively provide services to the users. By using beamforming, inter-user interference can be mitigated and a resource block can be utilized by more than one user. Short packet transmission is considered for URLLC users to satisfy their low-latency requirements. We formulate a mixed-integer nonlinear programming problem to maximize the aggregate throughput of eMBB users while guaranteeing the throughput, reliability, and latency requirements of URLLC users. We propose a low-complexity algorithm, which leverages fractional programming and successive convex approximation (SCA), to obtain the solutions. Simulation results show that our proposed algorithm can improve the aggregate eMBB throughput by 30% compared with the fixed-numerology based approach.

This paper investigates hybrid beamforming for cooperative multi-user millimeter-wave (mmWave) multiple-input multiple-output (MIMO) networks. We aim to jointly design the user scheduling and hybrid beamforming to maximize the sum-rate subject to the transmit power of each base station. Due to the non-convexity of constant modulus of phase shifters and objective function, the problem is mathematically intractable. We propose a low-complexity two-step scheme, in which user scheduling and analog beamforming are first obtained to maximize the sum-beamforming-gain, followed by digital beamforming calculation based on weighted minimum-mean-square-error (wMMSE) approach. We further extend the hybrid beamforming design to dynamic sub-array architecture, where a novel antenna selection algorithm is developed. Simulation results demonstrate the effectiveness of the proposed algorithms, which can outperform other state-of-the-art approaches.

Hybrid precoding combined with large-scale antenna arrays is considered as a key enabling technology for millimeter wave (mmWave) communications for its advantages in both reducing the number of power-hungry radio frequency (RF) chains and providing for spatial multiplexing. In this paper, we consider a dynamically sub-connected hybrid architecture with hardware-efficient low-resolution phase shifters (PSs) for a wide-band mmWave multi-user multi-input single-output orthogonal frequency division multiplexing (MU-MISO-OFDM) system. In this architecture, each RF chain is adaptively connected to a non-overlapping subarray corresponding to channel state information (CSI). Thus, multiple-antenna diversity can be fully utilized to mitigate the performance loss caused by the use of low-resolution PSs. Aiming at maximize the average sum-rate of the considered mmWave MU-MISO-OFDM system, we develop an iterative algorithm based on penalty dual decomposition (PDD) methods. Simulation results demonstrate the advantages of the considered dynamically sub-connected hybrid architecture.

This paper considers the cloud radio access network (C-RAN) based millimeter-wave (mmWave) cell-free communications, where multiple remote radio heads (RRHs) are distributed to provide reliable communication links to users via analog beamforming and connected to centralized baseband unit (BBU) which carries out digital signal processing. We aim to jointly design the user association and analog/digital hybrid beamforming along with fronthaul compression to maximize the minimum signal to interference-plus-noise ratio (SINR) among users while satisfying the fronthaul capacity constraints. To solve this difficult combinatory problem, we propose to first obtain the user association and analog beamforing to maximize the minimum beamforming gain among users. Then, given the effective baseband channel, the digital beamformer and quantization noise covariance matrix still cannot be calculated directly due to the non-convexities of objective function and fronthaul constraint. To efficiently solve this problem, we transform the objective function into convex terms based on fractional programming method and iteratively calculate the digital beamformer and quantization noise covariance matrix until convergence is achieved. Simulation results show that the proposed algorithm can achieve comparable performance to the full-digital beamforming.

Millimeter wave (mmWave) communications with large-scale antenna arrays and hardware-efficient analog/digital hybrid beamforming have been widely considered as one of the key technologies to enable very high data rate in the fifth generation (5G) applications. Meanwhile, physical layer security (PLS) in mmWave wiretap systems and secure hybrid beamformer designs have drawn increasing attention to safeguard 5G-and-beyond networks. However, in existing literatures, infinite or high-resolution phase shifters (PSs) are often assumed to implement fine-tunable analog beamformers, which are impractical due to high hardware cost and power consumption. In this paper, we consider the problem of hybrid beamformers design with practicallow-resolutionPSs for secure transmission in mmWave wiretap multi-input multi-output (MIMO) systems. We aim to develop secure hybrid beamforming algorithms to maximize the secrecy rate according to different availabilities of eavesdropper's channel state information (CSI). Particularly, when eavesdropper's CSI is available, the proposed algorithm first determines the secure analog beamformers by an iterative algorithm, then finds the digital beamformers which can further enhance the security. If eavesdropper's CSI is unknown, we develop an artificial noise (AN)-based secure hybrid beamforming approach. Simulation results demonstrate that our proposed algorithms can provide significant secrecy performance improvement.

Analog beamformer with large-scale antenna arrays has been widely considered in millimeter wave (mmWave) communication systems because of its superiority in hardware cost and energy consumption compared with traditional fully digital beamforming schemes. In this paper, we introduce an efficient dynamic subarray analog beamforming architecture with low-resolution phase shifters (PSs) for mmWave multiuser multipleinput single-output (MU-MISO) systems. In an effort to mitigate the performance loss due to the use of low- resolution PSs, each user can dynamically select a non-overlap subarray from total transmit antennas and use corresponding subarray analog beamformer to transmit signals. This dynamic subarray analog beamforming architecture can utilize the multi- antenna/multiuser diversities by dynamically adapting to channel state information (CSI) of users. An efficient dynamic subarray analog beamformer design algorithm is also presented, which aims at maximizing the sum-rate of the MU-MISO system. Simulation results demonstrate that the proposed dynamic analog beamforming solution can significantly outperform the conventional fixed-subarray schemes.

This paper investigates analog beamforming with large-scale antenna arrays for single-group multicast transmission. We focus on the max-min fair (MMF) problem and aim to design the analog beamformer with infinite and finite resolution phase shifters (PSs), respectively, to maximize the minimum signal-to-noise ratio (SNR) over all users subject to a transmit power constraint. However, the constant magnitude and infinite/finite phase constraints imposed by PSs frustrate the access of an optimal solution of analog beamformer. We thus formulate a sub-optimal MMF problem alternatively and propose a low-complexity algorithm, which iteratively determines each element of analog beamformer to conditionally maximize the minimum SNR among users. The computational complexities of our proposed algorithms are linear in the number of antennas. Simulation results illustrate that our proposed analog beamformer design can achieve satisfactory performance which is close to the full-digital case and outperform the other state-of-the-art schemes.

Economical and energy-efficient analog/digital hybrid beamforming has been widely considered as a promising approach for millimeter wave (mmWave) multiple-input multiple-output (MIMO) systems. While most hybrid beamforming techniques consider a fully-connected structure with a large number of phase shifters (PSs), the partially-connected structure has drawn more attention recently since it requires much less PSs and can further improve energy-efficiency. However, the impractical assumption of infinite or high resolution of PSs in existing solutions frustrates the real-world deployment of hybrid beamforming designs, and low- resolution PSs are typically adopted to reduce the hardware complexity and power consumption. In an effort to achieve maximum hardware efficiency, this paper focuses on the partially-connected architecture with one-bit (binary) PSs and considers the problem of joint hybrid precoder and combiner design for such mmWave MIMO systems. We propose to successively design the analog beamformers associated with each pair of sub- array, aiming at conditionally maximizing the spectral efficiency. A novel binary analog precoder and combiner optimization algorithm is proposed under a rank-1 approximation of the interference-included equivalent channel with polynomial complexity in the number of antennas. Then, the digital precoder and combiner are computed based on the obtained effective baseband channel to further enhance the spectral efficiency. Simulation results demonstrate the advantages of proposed hardware-efficiency hybrid precoder and combiner design.

Acting like a legitimate user via sending identical pilot signals, the pilot spoofing attack launched by an active eavesdropper can disrupt the reception of the legitimate receiver and, more importantly, cause severe information leakage. Although such an attack can be detected by the recently presented random-training-aided (RTA) pilot spoofing detector with high accuracy and low complexity, the assumption that the active eavesdropper remains silent during the random phase limits the usage of RTA algorithm in practice. In addition, a natural question to ask is how to secure the data transmission after spoofing detection. Motivated by these two aspects, in this paper we first investigate the robustness of RTA pilot spoofing detector and illustrate that it can provide high detection accuracy even when the eavesdropper is active during the random phase. Then, we further propose a zero-forcing (ZF)-based secure transmission approach to protect the legitimate transmission from eavesdropping in case of the missed detection of the active eavesdropper. Simulation studies demonstrate the robustness of the RTA pilot spoofing detector and the satisfactory performance of the proposed secure transmission strategy.

Millimeter wave (mmWave) communications have been considered as a key technology for future 5G wireless networks. In order to overcome the severe propagation loss of mmWave channel, multiple-input multiple-output (MIMO) systems with analog/digital hybrid precoding and combining transceiver architecture have been widely considered in mmWave systems. However, physical layer security (PLS) in mmWave MIMO systems and the secure hybrid beamformer design have not been well investigated. In this paper, we consider the problem of hybrid precoder and combiner design for secure transmission in mmWave MIMO systems in order to protect the legitimate transmission from eavesdropping. When eavesdropper's channel state information (CSI) is known, we first propose a joint analog precoder and combiner design algorithm which can prevent the information leakage to the eavesdropper. Then, the digital precoder and combiner are computed based on the obtained effective baseband channel to further maximize the secrecy rate. Next, if prior knowledge of the eavesdropper's CSI is unavailable, we develop an artificial noise (AN)-based hybrid beamforming approach, which can jam eavesdropper's reception while maintaining the quality-of-service (QoS) of intended receiver at the pre-specified level. Simulation results demonstrate that our proposed algorithms offer significant secrecy performance improvement compared with other hybrid beamforming algorithms.

Millimeter wave (mmWave) communications have been considered as a key technology for future 5G wireless networks since it can provide orders‐of‐magnitude wider bandwidth than current cellular bands. To overcome the severe propagation loss of the mmWave channel, an economic and energy‐efficient analogue/digital hybrid precoding and combining transceiver architecture is widely used in mmWave massive multiple‐input multiple‐output (MIMO) systems. The digital precoding/combining layer offers more freedom than pure analogue one and enables multi‐stream transmission. In this study, the authors consider the problem of codebook‐based joint hybrid precoder and combiner design for multi‐stream transmission in mmWave MIMO systems. The authors propose to jointly select an analogue precoder and combiner pair for each data stream successively, which can maximise the channel gain as well as suppress the interference between different data streams. Then, the digital precoder and combiner are computed based on the obtained effective baseband channel to further mitigate the interference and maximise the sum‐rate. Both fully‐connected and partially‐connected hybrid beamforming structures are investigated. Simulation results demonstrate that the proposed algorithms exhibit prominent advantages in combating interference between different data streams and offer satisfactory performance improvements compared with the existing codebook‐based hybrid beamforming schemes.