Efficient resource allocation for hybrid nonorthogonal multiple access based heterogeneous networks beyond fifth‐generation

Abstract

Fifth‐generation (5G) and future beyond fifth‐generation (B5G) networks will require immense capacity due to the high rise in the number of multimedia applications and mobile devices. The heterogeneous networks (HetNets) in 5G and B5G can increase the heterogeneity and network throughput because macro base station‐only (MBS‐only) networks cannot satisfy the drastic increase in capacity demands. Hybrid nonorthogonal multiple access (H‐NOMA) can accommodate the increasing number of multimedia applications and mobile devices in 5G and B5G. H‐NOMA in B5G can also improve spectral efficiency. Before this, the researchers had not considered user clustering with H‐NOMA in HetNets. This article investigates the user clustering with downlink H‐NOMA in HetNet to maximize the network throughput in 5G and B5G. The formulated mathematical problem optimizes the considered key performance indicators (KPIs), that is, users admission in clusters, users association with base stations, power allocation to users, and network throughput. At the same time, it also satisfies the minimum transmit power and data rate requirements of users. The formulated problem is a mixed‐integer nonlinear programming (MINLP) problem. We have proposed an ϵ$$ \epsilon $$‐optimal algorithm, that is, outer approximation algorithm (OAA) to solve the MINLP problem because the complexity of the optimal exhaustive search algorithm (ESA) increases exponentially with an increase in the number of users. H‐NOMA with user clustering in HetNet is evaluated with extensive simulations to show its effectiveness regarding network throughput in 5G and B5G. We have also considered the proposed framework in the MBS‐only network and compared its performance with HetNet. The results verify that the proposed framework in HetNet performs better than the MBS‐only network. The complexity of the ϵ$$ \epsilon $$‐optimal algorithm is calculated that gives ϵ$$ \epsilon $$‐optimal results within ϵ=10prefix−3$$ \epsilon =1{0}^{-3} $$. We have also made a complexity analysis of the proposed OAA and ESA and concluded that the complexity of the proposed OAA algorithm is less than the ESA.

Deep reinforcement learning‐based joint task offloading and resource allocation in multipath transmission vehicular networks

“…Li proposes a mobile edge computing offloading algorithm that considers the reliability of task transmission and task completion time 22 . As the task offloading efficiency of MEC is in high dependence on the transmission resources of the wireless network, various joint task offloading and resource allocation (JTORA) algorithms have been proposed 23,24 . Yu and Zhang proposed task offloading schemes integrating sub‐carrier allocation optimization 25,26 .…”

“…22 As the task offloading efficiency of MEC is in high dependence on the transmission resources of the wireless network, various joint task offloading and resource allocation (JTORA) algorithms have been proposed. 23,24 Yu and Zhang proposed task offloading schemes integrating sub-carrier allocation optimization. 25,26 Wang and Hu consider the impact of wireless transmission power on task offloading in MEC.…”

Satisfying strict deadlines for cellular Internet of Things through hybrid multiple access

“…Yet, the advantage of NOMA in terms of system capacity depends on the diversity of user channel conditions and the number of connected users. In order to address various needs of emerging applications, a Hybrid MA scheme consisting of adaptively switching between OMA and NOMA in the time domain is considered by Third Generation Partnership Project (3GPP) 15‐17 …”

“…In order to address various needs of emerging applications, a Hybrid MA scheme consisting of adaptively switching between OMA and NOMA in the time domain is considered by Third Generation Partnership Project (3GPP). [15][16][17] The focus of this study is to maximize the timely throughput using Hybrid MA for cellular IoT applications. Although the potential of Hybrid MA is widely studied in the scope of information freshness, [18][19][20][21] to the best of our knowledge, there is a lack of study on Hybrid MA in the scope of timely throughput for cellular IoT applications.…”

Satisfying Strict Deadlines for Cellular Internet of Things through Hybrid Multiple Access

“…Yet, the advantage of NOMA in terms of system capacity depends on diversity of user channel conditions and number of connected users. In order to address various needs of emerging applications, a Hybrid MA scheme consisting of adaptively switching between OMA and NOMA in the time domain is considered by 3rd Generation Partnership Project (3GPP) 15,16,17 , which is a collaborative project to develop globally applicable specifications for mobile systems.…”