Design of Power Allocation and Device Grouping in PD-NOMA Communication Systems

Abstract

Power domain non-orthogonal multiple access (PD-NOMA) is considered an effective means of enabling massive connectivity and improving spectral efficiency in next-generation (6G) wireless communication networks, particularly for the massive Internet of Things (mIoT). This paper optimizes power allocation and device grouping to address energy efficiency and resource scarcity challenges. We investigate optimal power allocation under two distinct successive interference cancellation (SIC) decoding orders at the receiver, incorporating constraints on limited transmit power, quality of service (QoS), and SIC decoding requirements. Using Karush-Kuhn-Tucker (KKT) conditions, we derive rigorous criteria for grouping IoT devices on shared subchannels and determine their optimal power coefficients. Analytical results prove that channel-gain-ascending SIC maximizes downlink subchannel capacity, while channel-gain-descending SIC optimizes uplink performance. Furthermore, we propose a resource-efficient device grouping scheme that reduces the required number of orthogonal subchannels compared to conventional orthogonal multiple access (OMA). Simulations demonstrate that the proposed scheme not only conserves the number of orthogonal subchannels compared to the OMA system, but also exhibits superior overall system capacity. This work directly contributes to energy-efficient resource management in sustainable IoT communications by minimizing spectral resources and enhancing network longevity.