This paper studies the sum spectral/energy efficiency (SE/EE) of a full-duplex (FD) cell-free (CF) massive multiple-input multiple-output (MIMO) system with low-resolution analog-to-digital converters (ADCs) at the access points (APs) and the downlink (DL) users. We characterize the joint impact of the residual self-interference, inter-AP interference, uplink (UL)-to-DL interference, pilot contamination and quantization noise (QN) by deriving closed-form solutions for the UL/DL SE. It is shown that the UL SE loss due to the QN can be compensated by increasing the receiving antenna arrays. By contrast, in the DL, increasing the transmit antennas is ineffective in compensating for the DL SE loss. Furthermore, the analysis shows that the operating region of the FD CF massive MIMO is reduced as the ADCs’ resolution declines. We investigate the EE/SE trade-offs.

This paper studies the performance of a half-duplex (HD) two-way relay (TWR) in power line communication (PLC) over a log-normal fading channel with impulsive noise. We consider the two common relaying protocols, i.e., amplify-and-forward (AF) and decode-and-forward (DF). For the DF relaying, we apply physical-layer network coding (PNC) and analog-network coding (ANC) to the PLC TWR. We derive analytic expressions for the average sum capacity and the outage probability of the system. Monte Carlo simulations are provided throughout to validate our analysis. Analytic results show a tight approximation to the simulation results. We compare the one-way relay (OWR) to the TWR and show that the HD spectral efficiency loss incurred by the OWR can be sufficiently mitigated in PLC. However, the outage probability of the TWR is inferior to that of the OWR. To enhance the outage performance of the PLC TWR, we implement a hybrid PLC/wireless (HPW) system where all nodes are equipped with PLC and wireless capabilities. Data transmission occurs over the two parallel links. The diversity in the transmission allows the TWR to improve its outage performance in the AF and DF protocols. The impact of the impulsive noise, inherent to PLC channels, is also highlighted in the simulation results. It is shown that the impulsive noise severely impairs system performance.

This letter studies physical layer security (PLS) in non-orthogonal multiple access (NOMA) power line communication (PLC) networks. A source node transmits information to two nodes of near and far users with different power allocations. We consider two different cases of eavesdropping: internal and external. The PLC channel and noise are modeled by correlated log-normal fading and Bernoulli-Gaussian random process, respectively. We derive closed-form expressions for the secrecy outage probability (SOP) of both the internal and external cases by applying Gauss-Chebyshev quadrature, which are verified by Monte Carlo simulations. We observe that higher impulsive noise degrades the SOP while higher channel correlation can improve the SOP. In addition, the analytic results reveal the impact of power allocation on the SOP performance.

This paper proposes a heterogeneous network (Het-Net) topology where the backhaul links are supported by massive MIMO systems with full-duplex (FD) modes. Communication is achieved in two phases. In the first phase, we equip macro cell (MC) base station (BS) with massive receive antennas and few transmit antennas and all small cell (SC) BSs with massive receive antennas and a single transmit antenna. In the second phase, a circulator switches the massive receive antennas into transmit antennas and transmit antennas into receive antennas. Under the assumption of imperfect channel state information and large antenna regime, closed-form expressions have been derived for the uplink/downlink achievable sum-rate. From the results, we check that the strength of self-interference (SI) and SC-to-SC interference which occur due to FD operation depends largely on the number of SCs. By increasing the number of SCs, achievable sum-rates also increase but requires more antennas to overcome the deleterious effects of SI and SC-to-SC interference. Based on our analytical results, a hybrid FD/half-duplex system is proposed according to the interferences and the number of antennas. Furthermore, using our system model, it is shown that the transmit power of the MC BS and the SC BSs can be scaled down proportionally to the massive number of receive antennas (first phase) and the massive number of transmit antennas (second phase) and still achieve a required rate.

This paper studies the spectral/energy efficiency (SE/EE) of a heterogeneous network with the backhaul enabled by low-resolution analog-to-digital converters (ADCs) quantized full-duplex massive multiple-input multiple-output (MIMO) over Rician channels. Backhaul communication is completed over two phases. During the first phase, the macro-cell (MC) base station (BS) deploys massive receive antennas and a few transmit antennas; the small-cell (SC) BSs employ large-scale receive antennas and a single transmit antenna. For the second phase, the roles of the transmit and receive antennas are switched. Due to the low-resolution ADCs, we account for quantization noise (QN). We characterize the joint impact of the number of antennas, self-interference, SC-to-SC interference, QN, and Rician K-factor. For the first phase, the SE is enhanced with the massive receive antennas and the loss due to QN is limited. For the second phase, the desired signal and QN have the same order. Therefore, the SE saturates with the massive transmit antennas. As the Rician K-factor increases, the SE converges. Power scaling laws are derived to demonstrate that the transmit power can be scaled down proportionally to the massive antennas. We investigate the EE/SE trade-offs. The envelope of the EE/SE region grows with increase in the Rician K-factor.

This paper analyzes the spectral/energy efficiency (SE/EE) of a full-duplex (FD) cell-free (CF) massive multiple-input multiple-output (mMIMO) over Rician fading channels. Due to the FD radios, the access points (APs) suffer from self-interference (SI) and inter-AP interference (IAI) while the downlink (DL) users’ signals are corrupted by the uplink (UL) users’ transmissions. We consider the case, where low-resolution analog-to-digital converters (ADCs) are utilized at the APs and DL users, which introduces the quantization noise (QN). The combined effects of Rician κ-factor, residual SI/IAI, UL-to-DL interference, multi-user interference, and QN on the UL/DL SEs are characterized. The UL SE is degraded by the increase in DL power, whereas the growth in UL power deteriorates the DL SE. The effects of the residual SI/IAI and UL-to-DL interference are worsened by the low-resolution ADCs. We optimize the UL/DL transmit powers to maximize the overall sum SE of the network. It is observed that the proposed power allocation algorithm brings substantial sum SE gain. A trade-off analysis between the EE and SE, as a function of the ADCs’ resolution, shows that the entire envelope of the operating region of FD CF mMIMO is enhanced in Rician channels.