In this article, a framework of artificial intelligence (AI)-native cross-module optimized physical layer with cooperative control agents is proposed, which involves optimization across global AI/machine learning (ML) modules of the physical layer with innovative design of multiple enhancement mechanisms and control strategies. Specifically, it achieves simultaneous optimization across global modules of uplink AI/ML-based joint source-channel coding with modulation, and downlink AI/ML-based modulation with precoding and corresponding data detection, reducing traditional inter-module information barriers to facilitate end-to-end optimization toward global objectives. Moreover, multiple enhancement mechanisms are also proposed, including i) an AI/ML-based cross-layer modulation approach with theoretical analysis for downlink transmission that breaks the isolation of inter-layer features to expand the solution space for determining improved constellation, ii) a utility-oriented precoder construction method that shifts the role of the AI/ML-based CSI feedback decoder from recovering the original CSI to directly generating precoding matrices aiming to improve end-to-end performance, and iii) incorporating modulation into AI/ML-based CSI feedback to bypass bit-level bottlenecks that introduce quantization errors, non-differentiable gradients, and limitations in constellation solution spaces. Furthermore, AI/ML-based control agents for optimized transmission schemes are proposed that leverage AI/ML to perform model switching according to channel state, thereby enabling integrated control for global throughput optimization. Finally, simulation results demonstrate the superiority of the proposed solutions in terms of block error rate and throughput. These extensive simulations employ more practical assumptions that are aligned with the requirements of the 3rd Generation Partnership Project (3GPP), which hopefully provides valuable insights for future 3GPP standardization discussions.

In addressing the trade-off between prediction model accuracy and computational cost in the context of path tracking control, this paper proposes a learning-based model predictive control (LB-MPC) strategy for autonomous buses. A three-degree-of-freedom (DOF) single-track vehicle dynamic model is established, and an in-depth analysis is conducted on its step response error with respect to variations in vehicle speed, pedal position, and front wheel steering angle compared to the IPG TruckMaker model. Methods for constructing error datasets and receding horizon updates are designed, and a Gaussian process regression (GPR) is employed to establish an error fitting model for real-time error compensation and correction of the nominal single-track model. The error correction model is utilized as the prediction model, and a path tracking cost function is designed to formulate a quadratic programming (QP) optimization problem, proposing an LB-MPC path tracking control architecture. Through joint simulations using the IPG TruckMaker & Simulink platform and real bus experiment, the real-time performance and effectiveness of the proposed GPR error correction model and LB-MPC path tracking control strategy are verified. Results demonstrate that compared to traditional MPC path tracking control strategy, the proposed LB-MPC strategy reduces the average path tracking error by 79.00%.

3D mapping is a fundamental task in robot applications. The traditional mapping methods mainly rely on spatial discretization, in which the amount of data that needs to be stored is large, and the representation ability is limited. As a continuous probability model, the Gaussian Mixture Model (GMM) has a small memory footprint and high-fidelity representation ability. Thus the GMM map is superior to discrete map representations in basic robot tasks such as navigation and localization. The general method of building GMM maps is the iterative Expectation-Maximization (EM) algorithm with K-means initialization. The EM and K-means algorithms are computation-intensive, making it challenging to meet real-time 30 fps mapping requirements on the embedded robot systems. This paper proposes a Gaussian mixture model mapping and navigation engine (GAME) on embedded FPGA to accelerate the mapping process. To achieve fully pipelined with minimal hardware resource cost, we design a unified dataflow and hardware architecture for both K-means and EM for GMM. We analyze different quantization strategies for higher parallelism and find a low-bit quantization method with mixed 8/16-bit data representation, bringing negligible loss in accuracy. Combining the unified dataflow and the mixed-bit data quantization, GAME enables real-time GMM mapping and navigation on embedded robots. The experimental results on ZCU102 show that our proposed hardware-software co-optimization framework on FPGA can run over 60× faster than on a GeForce 1080Ti GPU and over 490× faster than on an Nvidia Jetson TX2, and achieves 59 fps.

Considering the time duplex division (TDD)-based cellular network, this paper investigates the uplink/downlink interference under the framework of centralized-radio access network (C-RAN). According to different uplink/downlink configurations within three neighboring cells, two fundamental cases are discussed, which can be further classified into several scenarios. For the sake of tackling this kind of interference, the interference alignment (IA) transceivers are studied case-by-case. Specifically, by applying the concept of reference vectors, to which direction the interference is to be aligned, we propose a simply methodology to align the interference at the un-intended receivers. Then with respect to the antenna settings, the heuristic process of IA design is presented and discussed. Finally, simulations verify the proposed IA transceivers greatly outperform its benchmark, when there is strong uplink/downlink inter-cell interference.

In this paper, we study the regenerative multi-way relaying (MWR) network where single-antenna users exchange data with each other via a multi-antenna relay. By considering the broadcast phase as a particular space-time channel, we jointly design the relay transmitter and user receivers, aiming to minimize the total mean square error (T-MMSE). As a consequence, an effective multiple-input multiple-output (MIMO) channel with full rank has been established between the relay and each user. Due to the coupled transceiver structure, we propose an iterative algorithm to obtain a converged transceiver. Finally, simulation results evidence that the jointly designed transceiver performs well, even in the overloaded scenario.

We consider the multi-way relaying (MWR) network, where each user communicate with the other users through a relay. The single-antenna users transmit to the multi-antenna relay simultaneously. Then the relay randomly precodes the received or estimated signal, and broadcasts it back to all the users. Focusing on a specific user, we set up the equivalent space-time channel from the other users to it. Afterwards, for both amplify-and-forward and estimate-and-forward MWR, we design the ordered minimum mean square error (MMSE) receivers with successive interference cancellation (SIC) at users. In particular, three types of metrics determining the order of SIC are derived by approximating the interference-plus-noise term as Gaussian. Simulation results on error performance evidence that our proposed receivers are able to harvest the inherent temporal diversity gains within the MWR networks. Moreover, the proposed receivers perform even well in the overloaded scenarios.

Consider a wireless network, in which multiple users exchange data with each other via a multi-antenna relay. The users transmit to the relay simultaneously. Then the relay regenerates the transmitted signal and precodes it before broadcasting to the users. In particular, we propose a minimum mean square error (MMSE) based precoder, which takes the relay detection errors into account. We indicate that this precoder can improve the bit error performance when uplink channel is in a bad condition. Afterwards, we deploy the ordered MMSE successive interference cancellation (MMSE-SIC) receiver, which exploits temporal diversity gains over the multiple downlink slots. Simulation results show that the ordered MMSE-SIC receiver can achieve sufficient diversity gains over traditional MMSE receiver.

The performance of multihop links is studied in this contribution by both analysis and simulations, when communicating over Generalized-K (KG) fading channels. The performance metrics considered include symbol error rate (SER), outage probability, level crossing rate (LCR) and average outage duration (AOD). First, the expressions for both the SER and outage probability are derived by approximating the probability density function (PDF) of the end-to-end signal-to-noise ratio (SNR) using an equivalent end-to-end PDF. We show that this equivalent end-to-end PDF is accurate for analyzing the outage probability. Then, the second-order statistics of LCR and AOD of multihop links are analyzed. Finally, the performance of multihop links is investigated either by simulations or by evaluation of the expressions derived. Our performance results show that the analytical expressions obtained can be well justified by the simulation results. The studies show that the KGchannel model as well as the expressions derived in this paper are highly efficient for predicting the various types of performance metrics and statistics for design of multihop communication links.

Cooperative diversity has the ability of achieving multiplexing and diversity gains by taking advantage of the features and potentials of spatial distribution. Based on the asymptotic outage probability, we define the Cooperative Gain (CG) to evaluate the benefits of cooperation over direct transmission under high Signal-to-Noise Ratio (SNR). With the presence of Rayleigh fading and path loss, our analytical results illustrate the appropriate Cooperative Region (CR), in which the source prefers to coordinately work with the relay rather than transmit directly. Finally, we provide visible CRs and numerical results on the exact outage performance by Monte Carlo simulations.