In modern wireless systems, the process of acquiring channel state information (CSI) has become increasingly important and challenging due to the conflicting needs of expanding antenna arrays and achieving high spectrum efficiency. Recent solutions leverage joint source and channel coding (JSCC) realized by an auto-encoder structure to enable a hybrid optimization of CSI compression and transmission, which shows great potential to improve the effectiveness of CSI feedback. However, the gap between research and realization is still big. This article proposes a comprehensive JSCC method for CSI feedback, including a technique solution for eliminating the impacts of residential fading and radio frequency ($\mathbf{R F}$) imperfections, and a protocol design to support the realization of the method in practical cellular systems. Furthermore, a hardware proof-ofconcept (PoC) platform was developed to validate the JSCC-based CSI feedback, and the results demonstrate the superiority of the method in realistic environments.

The non-linear distortion caused by non-ideal radio frequency (RF) components especially the power amplifier (PA) limits the applications of higher order modulation and degrades power utilization efficiency. To improve the achievable rate in modern systems, it becomes critical to overcome the non-linear distortion so that we can maximize the opportunity of using higher order modulation such as 256QAM, 1024QAM and even 4096QAM at high transmission power. In this paper, we introduce an artificial intelligence (AI)-enabled non-linearity compensation scheme (AI-NC) to avoid the "model deficit" problem. The introduced AI-NC adapts the Echo State Network (ESN) to enable fast online training without additional training overhead. Furthermore, it is general enough to be used for any types of power amplifiers (PAs) with different non-linearity characteristics and different channel environments. It can also be used for the communication system using multiple antennas and supporting multiple users simultaneously. Simulation results and hardware-based tests show that the proposed AI-NC can drastically improve the link performance and/or coverage of higher order modulations in practice.

The conflict between increasing demands for high accuracy positioning information and poor performance of conventional schemes in non-line-of-sight (NLOS) scenarios motivates the study to find new promising approaches. Artificial intelligence or machine learning (AIIML) based schemes have shown great potential to reach this goal. However, one key point is to design suitable data input for enabling AIIML for positioning in practice. And considering the increasing bandwidth, sampling rate, and number of antennas in wireless cellular systems, the commonly used raw channel information (e.g., channel impulse response CIR), which could be very high-dimensional, may not be appropriate for the input of AI models and result in high model complexity and demand for massive training samples. Therefore, in this paper, the log signature transform based method, which utilizes iterated integrals of paths, is proposed for AI-based positioning with novel time-augmented path construction and feature derivation. The proposed method can drastically reduce the input feature dimensionality and consequently reduce the AI model complexity. Moreover, it can have better accuracy for limited training samples or/and apply simpler AIIML models such as multi-layer perceptron (MLP). The experimental results show that around 98 % reduction in both input dimensionality and computational complexity can be achieved without compromising the positioning accuracy.

This paper discusses a new 5G-Advanced duplex scheme called sub-band full-duplex (SBFD) and the details of implementing it for 5G TDD spectrum. The SBFD requires an advanced self interference cancellation (SIC) which is designed to handle extremely high level of cross link interference. Novel SIC techniques were designed for base stations (BSs) operating on 3.4 GHz with massive MIMO and on 26 GHz with multi-beam capability. By utilizing electromagnetic walls in conjunction with electromagnetic bandgap, an extremely high level of isolation between transmit and receive antennas is achieved. Residual interference is further removed using fractional nonlinear interference estimation in baseband. Two Proofs-of-Concepts (PoCs) have been implemented for feasibility verification. The first PoC was for a massive MIMO BS operating with 32 cross polarized antenna ports on 3.4 GHz and a bandwidth of 100 MHz. The second PoC was for a multi-beam BS with 256 antenna elements operating on 26 GHz and a bandwidth of 400 MHz. It was observed that the self-interference in SBFD can be suppressed in real time to a level of less than 0.7 dB over the RX noise floor for both 3.4 GHz and 26 GHz in an outdoor environment.

Full duplex (FD) communication is considered as a promising technology in the development of 5G-advanced and 6G systems as it theoretically doubles the capability of channel. However, this improvement relying on a simultaneously bidirectional manner of communication induces intrinsic self-interference (SI) demanding to be fully cancelled, which is generally intractable. This challenge is minimized in the scenario of integrated access and backhaul (IAB) networks operated in millimeter wave (mmW) band, as its transceivers are stationary and the complexity of SI is greatly reduced by beamforming technique. As a matter of fact, FD can be a pioneering technique to unlock the full potential mmW-based IAB network by releasing its suffering of the half-duplex inefficiency. This article presents the design principle and validation of a practical SI cancellation (SIC) technique in the case of high transmission power class in mmW-based IAB networks. The proposed technique sequentially eliminates SI from the spatial, RF, and digital domains that reduces the SI down to the noise floor. To validate the technique, a prototype system is developed in accordance with 5G IAB specifications and field tests are conducted. Results suggest that the designed mmW SIC transceiver significantly reduces residual-interference to noise ratio (R-INR) to 2.1 dB or less. Additionally, a system-level simulation is conducted in line with 5G IAB evaluation methodology, which explores the potential performance gain of the proposed technique in presence of cross-link interference (CLI). Results indicate that the method could yield a cell throughput gain of about 84%, compared to the current time division duplex deployment.

The ever-increasing roll-out of 5G integrated access and backhaul (IAB) networks in millimeter wave (mmW) band and its current hard-constraint of half duplex operation motivate the study of full duplex (FD) communication in a high power beamforming system. In an mmW IAB network using FD, the intrinsic self-interference inherits the nature of mmW high attenuation and cross link interference can be quite manageable via sharp beamforming between fixed nodes. As a matter of fact, mmW IAB could be a pioneering use case of FD in near future. In this article, a practical self-interference cancellation (SIC) transceiver design is presented targeting for “down to noise floor” SIC in the case of high transmission power class. We developed an mmW prototype testbed system in accordance with 5G IAB specifications. Our experimental results via prototype verifies the feasibility of the designed mmW SIC transceiver and confirms the potential of 2 times throughput gain from FD compared with conventional half duplex in an IAB scenario.

This paper emphasize the importance of the inter-sector self-interference (SI) cancellation to realize the full duplex (FD) for the practical three-sectored cell deployment scenario. We measure the inter-sector SI, the interference coming from the neighboring sector in the same site, using our FD enabled prototype testbed, and then we conduct the system-level simulation to validate the the impact of the inter-sector SI along with the intra-sector SI and cross link interference (CLI) in mmWave integrated access and backhaul (IAB) networks. From our extensive simulation results, the FD operation compared with the half duplex (HD) operation achieves throughput gain of 70% with inter-sector SI cancellation (SIC), while the FD operation achieves throughput gain of 20% without inter-sector SIC. Conclusively, the inter-sector SI should be effectively cancelled out to guarantee the meaningful FD gain over HD.

Full duplex (FD) communication is an enabling technology with simultaneous uplink and downlink transmission over the same spectrum, which could not only virtually double the spectrum but also shorten the latency of bi-directional communications. One of the key challenges to bring full duplex into reality is how to design a practically implementable self-interference cancellation (SIC). This paper promotes a practical joint-design of SIC capable of >120dB SIC gain to make in-band FD practically viable. It consists of novel integrated SIC antenna, multi-tap tunable RF SIC and non-linear digital SIC. Further, a prototype system implemented using in-band FD with 5G NR commercial level hardware components is presented, which not only verifies in-band FD is practically implementable with the proposed joint SIC, but also achieves highest SIC results, i.e. 122.5dB SIC capability with 32dBm transmit power, to our best knowledge. This result confirms the appealing potential of in-band FD and conclusively, the in-band FD communication with the developed effective SIC turns out to be a promising enabler for future business thriving.

Positioning for non-line-of-sight (NLOS) scenarios is a long-live challenging task, since the conventional approaches for positioning are mostly LOS-dependent, and perform poorly in NLOS scenarios. Thanks to the powerful computing and learning abilities of artificial intelligence (AI), the present study on machine learning (ML) based approaches shows promising potential to conquer the challenge in theory. Nonetheless, one critical aspect for using AI model is the generalizing ability, which tells the inference level in practice. This problem is more vital for ML based positioning, since the channel information (CI) for actual usage may not be aligned with that for training. One typical case is the noisy or incomplete CI, which may degrade the inference performance and even leads the trained model unusable. In this paper, a novel hybrid machine learning (HML) approach is introduced by exploiting both supervised and unsupervised learning models developed with denoising and inpainting abilities to enable accurate positioning in NLOS scenarios. The simulation shows the proposed approach can have 10 times higher accuracy than conventional approaches.

In this paper, an advanced duplex scheme, referred to as cross division duplex (XDD), is proposed to use flexible downlink (DL) and uplink (UL) allocation in time division duplex (TDD) carrier. By implementing self-interference cancellation (SIC) functions at the base station, the proposed XDD enables to simultaneously operate UL and DL on different frequency subband in the same channel. To do so, both TDD's ability to efficiently handle asymmetric UL and DL traffic and frequency division duplexing's coverage and latency advantage can be achieved together. To verify the feasibility of XDD, a Proof-of-Concept is developed and self-interference can be removed almost perfectly with proper implementation of SIC without guardband between DL and UL subband. From evaluation results, UL coverage, as well as latency, was improved using a longer UL duty period in XDD. Based on XDD, in 5G-Advanced, a significantly higher resource utilization than the existing 5G will be possible.

To support the numerous small data package transmission from the prosperously growing number of connected devices in industrial internet of things (IIoT) networks, the transmission schemes specifically designed for the small data package are eager to have due to inefficiency of using current scheduling-based schemes. Three types of transmission schemes are being promoted in the 5G standardization and its evaluation to optimally suit the properties of small data transmission. The Early Data Transmission (EDT) and Fast Data Transmission (FDT) incorporates the data package into the random access procedure in Msg.3 and Msg.1, respectively; while the Pure Data Transmission (PDT) is designed to initiate the data transmission even without trying actually access the network. The TRIDENT schemes are proved to be commercial success by being included in the current Release(s) of LTE & 5G communication network standard and being evolved in the future Releases.

The increasing small data package transmission from the prosperously growing number of deployed networks, connected devices and multi-media applications has triggered the study of new technologies to have optimized spectrum utilization. Non-orthogonal multiple access (NoMA) has been proposed in both academic and industry as a promising technology to achieve such purpose by providing the ability to serve massive user equipment (UE) and exploit the properties of small data transmission. In addition to the intra-cell multi-user interference handling, an evolved operation of a NoMA scheme with bit-level interleaving and symbol level grid mapping is proposed in this paper to solve its major issue in practical communication systems: inter-cell interference (ICI). Theoretical analysis and simulation evaluations are provided and the results show that the proposed method can effectively mitigate the inter-cell interference, and hardware test-bed also proves robust BLER performance could be achieved under multi-cell scenarios.

To supporting the systematic requirements of higher spectrum efficiency and user deployment density in future Industrial Internet of Things (IIoT), traditional orthogonal multiple access schemes, such as orthogonal frequency-division multiplex, are quite limiting. Nonorthogonal multiple access (NoMA) has been recognized as one of the enabling technologies for IIoT in 5G network. In this paper, we have proposed an alternative NoMA scheme called interleave-grid multiple access (IGMA). Depending on interleaving and grid-mapping process, IGMA is capable to provide reliable block error rate performance, marvelous user multiplexing capability, as well as robustness against intercell interference. IGMA can apply various detection and decoding techniques at receiver sides to improve the detection performance with acceptable complexity. Both of link-level and system-level results show the promising benefits of IGMA, in particular that nearly seven times user multiplexing gain compared with orthogonal frequency-division multiple access has been observed in the system-level simulation. In addition, a hardware test bed has been implemented to verify the IGMA performance and the testing results proved the strong competitiveness of IGMA over orthogonal frequency-division multiple access in terms of block error rate performance in user overloading scenarios.

In this paper, a low PAPR interleaving based non-orthogonal multiple access (NoMA) scheme is introduced. By utilizing the property of DFT spread OFDM (DFT-s-OFDM), the peak-to-average power ratio (PAPR) can be reduced compared with existing NoMA schemes without sacrificing achievable data rates. Meanwhile, a low-complexity detection algorithm is also proposed on top of the elementary signal estimator (ESE) with the joint Gaussian distribution assumption, which can maintain good detection performance with moderate detection complexity. Simulation results, including PAPR and BLER performance, show that the proposed scheme can provide good performance with lower PAPR, which is suitable for mMTC scenario for 5G.