The integration between orthogonal frequency division multiple access (OFDMA), non-orthogonal multiple access (NOMA), and cognitive radio (CR) technology has been recently configured as promising candidate to support the massive high-speed connectivity requirements of beyond fifth generation (B5G) systems. This combination is referred to in the literature as a hybrid OFDMA–NOMA CR network. In such a system, each primary user’s (PU) idle channel, i.e., sub-band, is opportunistically utilized to serve multiple secondary users (SUs) by exploiting power-domain NOMA. This study soughs to explore the resource allocation technique for a hybrid OFDMA–NOMA CR system to further elaborate the potential capabilities of this system. Specifically, a power minimization resource allocation framework is developed, in which the objective is to serve the contending CR users, i.e., SUs, with minimum possible total transmission power, while achieving a set of relevant NOMA and CR constraints. These constraints include the dynamic nature of PU activities, rate demand, probability of success and imposed CR power mask constraints. However, the developed power minimization framework is non-convex in nature. Therefore, we employ the sequential convex approximation (SCA) along with a second-order cone approach to deal with the non-convexity issues of the developed optimization problem, and thus evaluating the corresponding optimization parameters. Compared to the conventional orthogonal resource allocation techniques, simulation results reveal that our proposed resource allocation technique for the hybrid OFDMA–NOMA CR system has a considerable performance enhancement in terms of the overall network throughput and total required transmit power.

正交频分多址 (OFDMA)、非正交多址 (NOMA) 和认知无线电 (CR) 技术之间的集成最近已被配置为支持超五代 (B5G) 的大规模高速连接需求的有希望的候选技术) 系统。这种组合在文献中被称为混合 OFDMA-NOMA CR 网络。在这样的系统中，通过利用功率域 NOMA，每个主用户 (PU) 的空闲信道，即子带，被机会性地用于为多个次要用户 (SU) 提供服务。本研究试图探索混合 OFDMA-NOMA CR 系统的资源分配技术，以进一步阐述该系统的潜在功能。具体而言，开发了功率最小化资源分配框架，其中目标是为竞争的 CR 用户（即 SU）提供尽可能小的总传输功率，同时实现一组相关的 NOMA 和 CR 约束。这些约束包括 PU 活动的动态特性、速率需求、成功概率和强加的 CR 功率掩码约束。然而，开发的功率最小化框架本质上是非凸的。因此，我们采用顺序凸逼近 (SCA) 和二阶锥方法来处理已开发优化问题的非凸问题，从而评估相应的优化参数。与传统的正交资源分配技术相比，