Feasibility and applications of in-band full-duplex radios for future wireless networks

Abstract

Due to the continuous increase of the demands for the wireless network’s capacity, in-band full-duplex (IBFD) has recently become a key research topic due to its potential to double spectral efficiency, reduce latency, enhance emerging applications, etc., by transmitting and receiving simultaneously over the same channel. Meanwhile, many studies in the literature experimentally demonstrated the feasibility of IBFD radios, which leads to the belief that it is possible to introduce IBFD in the standard of the next-generation networks. Therefore, in this thesis, we timely study the feasibility of IBFD and investigate its advantages for emerging applications in future networks.

In the first part, we investigate the interference suppression methods to maximize the IBFD gain by minimizing the effects of self-interference (SI) and co-channel interference (CCI). To this end, we first study a 3-step self-interference cancellation (SIC) scheme. We focus on the time domain-based analog canceller and nonlinear digital canceller, explaining their rationale, demonstrating their effectiveness, and finding the optimal design by minimizing the residual effects. To break the limitation of conventional electrical radio frequency (RF) cancellers, we study the photonic-assisted canceller (PAC) and propose a new design, namely a fiber array-based canceller. We propose a new low-complexity tuning algorithm for the PAC. The effectiveness of the proposed fiber array canceller is demonstrated via simulations. Furthermore, we construct a prototype of the fiber array canceller with two taps and carry out experiments in real-world environments. Results show that the 3-step cancellation scheme can bring the SI close to the receiver's noise floor. Then, we consider the multiple-input multiple-output (MIMO) scenarios, proposing to employ hybrid RF-digital beamforming to reduce the implementation cost and studying its effects on the SIC design. Additionally, we propose a user allocation algorithm to reduce the CCI from the physical layer. A heterogeneous industrial Internet of Things (IIoT) scenario is considered, while the proposed algorithm can be generalized by modifying the parameters to fit any other network.

In the second part, we study the beamforming schemes for IBFD multi-cell multi-user (IBFD-MCMU) networks. The transceiver hardware impairments (HWIs) and channel uncertainty are considered for robustness. We first enhance zero-forcing (ZF) and maximum ratio transmission and combining (MRTC) beamforming to be compatible with IBFD-MCMU networks in the presence of multi-antenna users. Then, we study beamforming for SIC, which is challenging for MCMU networks due to the limited antennas but complex interference. We propose a minimum mean-squared error (MMSE)-based scheme to enhance the SIC performance while minimizing its effects on the sum rate. Furthermore, we investigate a robust joint power allocation and beamforming (JPABF) scheme, which approaches the performance of existing optimal designs with reduced complexity. Their performance is evaluated and compared through 3GPP-based simulations.

In the third part, we investigate the advantages of applying IBFD radios for physical layer security (PLS). We focus on a channel frequency response (CFR)-based secret key generation (SKG) scheme in MIMO systems. We formulate the intrinsic imperfections of IBFD radios (e.g., SIC overheads and noise due to imperfect SIC) and derive their effects on the probing errors. Then we derive closed-form expressions for the secret key capacity (SKC) of the SKG scheme in the presence of a passive eavesdropper. We analyze the asymptotic behavior of the SKC in the high-SNR regime and reveal the fundamental limits for IBFD and half-duplex (HD) radios. Based on the asymptotic SKC, numerical results illustrate that effective analog self-interference cancellation (ASIC) is the basis for IBFD to gain benefits over HD. Additionally, we investigate essential processing for the CFR-based SKG scheme and verify its effectiveness via simulations and the National Institute of Standards and Technology (NIST) test.

In the fourth part, we consider a typical application of IBFD radios: integrated sensing and communication (ISAC). To provide reliable services in high-mobility scenarios, we introduce orthogonal time frequency space (OTFS) modulation and develop a novel framework for OTFS-ISAC. We give the channel representation in different domains and reveal the limitations and disadvantages of existing ISAC frameworks for OTFS waveforms and propose a novel radar sensing method, including a conventional MUSIC algorithm for angle estimation and a delay-time domain-based range and velocity estimator. Additionally, we study the communication design based on the estimated radar sensing parameters. To enable reliable IBFD radios in high-mobility scenarios, a SIC scheme compatible with OTFS and rapidly-changing channels is proposed, which is lacking in the literature. Numerical results demonstrate that the proposed ISAC waveform and associated estimation algorithm can provide both reliable communications and accurate radar sensing with reduced latency, improved spectral efficiency, etc.