Celik Abdulkadir
Computer, Electrical, and Mathematical Sciences & Engineering (CEMSE) Division, King Abdullah University of Science and Technology (KAUST), Thuwal 23955-6900, Saudi Arabia.
Sensors (Basel). 2023 Oct 4;23(19):8245. doi: 10.3390/s23198245.
Non-orthogonal multiple access (NOMA) has emerged as a promising solution to support multiple devices on the same network resources, improving spectral efficiency and enabling massive connectivity required by ever-increasing Internet of Things devices. However, traditional NOMA schemes operate in a grant-based fashion and require channel-state information and power control, which hinders its implementation for massive machine-type communications. Accordingly, this paper proposes synchronous grant-free NOMA (GF-NOMA) frameworks that effectively integrate user equipment (UE) clustering and low-complexity power control to facilitate the power-reception disparity required by the power-domain NOMA. Although single-level GF-NOMA (SGF-NOMA) designates an identical transmit power for all UEs, multi-level GF-NOMA (MGF-NOMA) groups UEs into partitions based on the sounding reference signals strength and assigns partitions with different identical power levels. Based on the objective of interest (e.g., max-sum or max-min rate), the proposed UE clustering scheme iteratively admits UEs to form clusters whose size is dynamically determined based on the number of UEs and available resource blocks (RBs). Once the UEs are acknowledged with power levels and allocated RBs through random-access response (RAR) messages, UEs can transmit anytime without grant acquisition. Numerical results show that the proposed GF-NOMA frameworks can compute clusters in the order of milliseconds for hundreds of UEs. The MGF-NOMA can reach up to 96-99% of the optimal benchmark max-sum rate, and the SGF-NOMA reaches 87% of the optimal benchmark max-sum rate at the same power consumption. Since the MGF-NOMA and optimal benchmark enforce the strongest and weakest channel UEs to transmit at maximum and minimum transmit powers, respectively, the SGF-NOMA also offers a significantly higher energy consumption fairness and network lifetime as all UEs consume equal transmit powers. Although the MGF-NOMA delivers an inferior max-min rate performance, the SGF-NOMA is shown to reach 3e6 MbpJ energy efficiency compared to the 1e7 MbpJ benchmark.
非正交多址接入（NOMA）已成为一种很有前景的解决方案，可在相同网络资源上支持多个设备，提高频谱效率，并实现日益增长的物联网设备所需的大规模连接。然而，传统的NOMA方案以基于授权的方式运行，需要信道状态信息和功率控制，这阻碍了其在大规模机器类型通信中的应用。因此，本文提出了同步免授权NOMA（GF-NOMA）框架，该框架有效地集成了用户设备（UE）聚类和低复杂度功率控制，以促进功率域NOMA所需的功率接收差异。虽然单级GF-NOMA（SGF-NOMA）为所有UE指定相同的发射功率，但多级GF-NOMA（MGF-NOMA）根据探测参考信号强度将UE分组为不同的分区，并为不同的分区分配不同的相同功率电平。基于感兴趣的目标（例如，最大和速率或最大最小速率），所提出的UE聚类方案迭代地接纳UE以形成簇，其大小根据UE的数量和可用资源块（RB）动态确定。一旦UE通过随机接入响应（RAR）消息被确认功率电平并分配了RB，UE就可以随时发送而无需获取授权。数值结果表明，所提出的GF-NOMA框架可以在毫秒级内为数百个UE计算簇。MGF-NOMA在相同功耗下可以达到最优基准最大和速率的96%-99%，SGF-NOMA可以达到最优基准最大和速率的87%。由于MGF-NOMA和最优基准分别强制信道最强和最弱的UE以最大和最小发射功率进行传输，SGF-NOMA也提供了显著更高的能耗公平性和网络寿命，因为所有UE消耗相等的发射功率。虽然MGF-NOMA的最大最小速率性能较差，但与1e7 MbpJ的基准相比，SGF-NOMA的能效达到了3e6 MbpJ。
