The fifth generation (5G) technology standard, utilizing the Internet of Things, promises enhanced communication systems. However, the efficiency expected from such systems entails significant requirements, such as higher data rates and flexibility of the lowest 5G layer. Meeting these requirements in subsequent wireless communication systems is highly dependent on the use of waveforms capable of efficiently enabling multiple access. In other words, proper waveforms determine the effective handling of diverse traffic within a given band. In this study, four candidate multicarrier waveforms, namely filtered orthogonal frequency division multiplexing, filter bank multicarrier, universal filtered multicarrier, and orthogonal frequency division multiplexing, which is currently used in 4G systems, are compared based on multiple parameters. MATLAB simulation results indicate that the waveforms significantly improved spectrum localization and provided appropriate spectrum fragmentation. As these waveforms can mix diverse traffic specifications, they handle the problem of time-frequency synchronization effectively. Therefore, these new waveforms exhibit significant potential in terms of orthogonality and synchronicity and can support numerous users without dropping signals. In addition, they support all applications and scenarios related to multiple-input and multiple-output. The obtained simulation results confirm the suitability of such waveforms for 5G applications and systems.

Network slicing and resource allocation play pivotal roles in software-defined network (SDN)/network function virtualization (NFV)-assisted 5G networks. In 5G communications, the traffic rate is high, necessitating high data rates and low latency. Deep learning is a potential solution for overcoming these constraints. Secure slicing avoids resource wastage; however, DDoS attackers can exploit the sliced network. Therefore, we focused on secure slicing with resource allocation under massive network traffic. Traffic-aware scheduling is proposed for secure slicing and resource allocation over SDN/NFV-enabled 5G networks. In this approach (T-S3RA), user devices are authenticated using Boolean logic with a password-based key derivation function. The traffic is scheduled in 5G access points, and secure network slicing and resource allocation are implemented using deep learning models such as SliceNet and HopFieldNet, respectively. To predict DDoS attackers, we computed the Renyi entropy for packet classification. Experiments were conducted using a network simulator with 250 nodes in the network topology. Performance was evaluated using metrics such as throughput, latency, packet transmission ratio, packet loss ratio, slice capacity, bandwidth consumption, and slice acceptance ratio. T-S3RA was implemented in three 5G use cases with different requirements, including massive machine-type communication, ultra-reliable low-latency communication, and enhanced mobile broadband.