Heterogeneous Traffic Multiplexing in Next Generation Cellular Networks

The vision shaping the upcoming sixth-generation (6G) wireless cellular networks has recently gained considerable attention from researchers in academia and industry. 6G networks are expected to fulfill the limitations of the fifth-generation (5G) networks and support a wide range of new applications and services beyond those supported by 5G, namely, enhanced mobile broadband (eMBB), ultra-reliable and low latency communications (URLLC) and massive machine-type communications (mMTC). Further, these emerging networks are thus mandated to support new emerging applications that concurrently demand multiple quality of service (QoS) requirements of data rate, reliability, latency, and connectivity. Due to the fundamental trade-off of such extremely diverse QoS requirements, the coexistence of these emerging applications has been identified as a major challenge in 6G networks and their predecessors. This dissertation aims at addressing the coexistence problem, specifically URLLC and eMBB traffic, by developing spectrally efficient multiplexing and scheduling solutions. By considering different key enabling technologies, this dissertation provides unique research contributions to the coexistence problem that led to effective designs. In particular, coupling URLLC and eMBB through the Third Generation Partnership Project (3GPP) superposition/puncturing scheme naturally arises as a promising option due to the latter's tolerance in terms of latency and reliability. Moreover, reconfigurable intelligent surface (RIS) has been proposed as a potential low-cost and energy-efficient technology that can control the wireless propagation environment providing endless benefits in supporting coexisting 6G services. Regarding the superposition scheme, this thesis investigates the joint scheduling of eMBB and URLLC traffic while minimizing the eMBB rate loss, considering URLLC reliability and the eMBB QoS. In the context of puncturing, this thesis studied the interplay between the RIS configuration, URLLC reliability and eMBB rate by proposing proactive RIS configurations to guarantee the URLLC latency requirements. Although simulation results demonstrate that adopting the proposed scheme can further boost eMBB and URLLC traffic performance, the computational complexity of optimizing the RIS phase shifts is challenging. To this end, this thesis proposes two low-complexity methods for optimizing the RIS phase shift matrix. The first solution proposes reducing the number of optimization variables configuring the RIS to the number of users. The second algorithm is based on a closed-form expression for the RIS phase shift matrix. Finally, a new puncturing strategy is proposed to mitigate the impact on the eMBB transmission. The key idea of the proposed scheme is to puncture the eMBB data that has maximum symbol similarities with the URLLC leading to reducing the contaminated eMBB symbols. We study the performance of the proposed schemes in terms of the eMBB spectral efficiency, URLLC reliability and low complexity. We show analytically and through simulations the efficacy of the proposed schemes over their existing counterparts