Improving Spectral-efficiency And Energy-Efficiency In In-band Full-duplex Systems

As the mobile devices have been widely used,the demand of data rates in mobile internet has been growing rapidly,which is a huge challenge of wireless communication systems.Recently,in-band full-duplex(IBFD)technique,which attracts much attention,has been proposed to enhance the spectral-efficiency(SE).Full-duplex(FD)communications allow transmitting and receiving over the same frequency band simultaneously.Therefore,it has the potential to double the link gain.The main challenge of FD communications is the self interference(SI)caused by the FD device itself.There are three kinds of SI cancellation methods,known as propagation cancellation,analog cancellation,and digital cancellation.By jointly utilizing the three kinds of methods,the strength of the residual self interference(RSI)can be suppressed to a sufficiently low level.In this paper,a practical FD system is considered,where only the base station(BS)has the ability of FD communication while the mobile device still works in the conventional half-duplex(HD)mode due to the hardware limitation.In such a system,an uplink user and a downlink user can be paired to communicate over the same frequency band simultaneously.As a result,there are two kinds of interference:RSI at the BS affecting the uplink transmission and the co-channel interference(CCI)at the user side affecting the downlink transmission.We first aims at maximizing the system throughput of frequency-division FD cellular net-works.Opportunistic interference cancellation(OIC)technique is utilized to cancel the CCI.The throughput maximization problem is a joint power control,channel allocation,and user pairing problem,which is non-convex and NP-hard due to the binary variables.Thereby,a suboptimal heuristic algorithm with low computational complexity is proposed.Numerical results demon-strate that user diversity gain,FD gain,and OIC gain can be achieved by the proposed algorithm,respectively.Moreover,the performance of FD communications depends on the power of RSI.Then,we further investigate the system throughput in time-division cellular networks.The throughput maximization problem is a continuous non-convex problem,which can not be solved optimally.A sub-optimal algorithm based on the concave-convex procedure(CCCP)is proposed.In order to further reduce the complexity,a two-step heuristic algorithm without iteration is also proposed.Similarly to the frequency-division FD network,numerical results show that both user diversity gain and FD gain can be achieved and the performance of FD communications depends on the power of RSI.In the next,the power efficiency of FD networks is investigated.The feasible power region for co-channel uplink and downlink users without channel state information(CSI)is proposed in a closed-form way.The minimum overall transmit power is solved numerically.In the high signal-to-noise(SNR)case,the minimum overall transmit power can derived in a closed-form expression.A theorem to compare the power efficiency between FD and HD communications is further derived.Finally,the tradeoff between energy-efficiency(EE)and spectral-efficiency is addressed.We consider two different RSI models:constant RSI model and linear RSI model.First,the necessary conditions for a FD transceiver to achieve better EE-SE tradeoff than a HD one are derived for both RSI models.Then,for the constant RSI model,a closed-form EE-SE expression is obtained in the scenario of single pair of users.We further extend our result and prove that EE is a quasi-concave function of SE in the scenario of multiple user pairs.Accordingly,an optimal algorithm to achieve the maximum EE is developed.For the linear RSI model,the EE-SE relation is difficult to deal with and we develop a heuristic algorithm by decoupling the problem into two sub-problems:power control and resource allocation.Our analysis and algorithms are finally verified by comprehensive numerical results.