The convergence of Satellite Mobile Networks (SMNs) with Terrestrial Wireless Networks (TWNs) has emerged as a significant area of research. Such integration is crucial for overcoming the current challenges of static spectrum assignment faced by earlier wireless network generations, including inefficient energy utilisation, insufficient bandwidth, increased latency, reduced reliability, limited connectivity, and restricted capacity. To address these issues, the potential of Cognitive Radio Networks (CRNs) has been exploited in Fifth Generation (5G), facilitating seamless interoperability among networks to sustain wireless communications. This paper presents a novel Vertical Hypothesis Uncertainty (VHU) method for optimising system throughput in CRN, leveraging the spectrum sensing false alarm, Pfa and null detection, Pnd hypotheses. This approach integrates the Filtered Orthogonal Frequency Division Multiplexing (F-OFDM) with Spectrum Sensing (SS) within a Satellite-Terrestrial Network (STN) domain. The performance of the proposed VHU was tested against the Hybrid Filter Detection with Inverse Covariance (HFDIC) and traditional spectrum sensing concepts. Results at –10 dB Signal-to-Noise Ratio (SNR) show that the VHU method remarkably outperforms HFDIC, achieving a 7.9% improvement in a fixed channel and a 15.84% improvement in a dynamic channel under perfect channel conditions.

The multi-carrier transmission based on orthogonal frequency-division multiplexing (OFDM) footing is invented to mitigate problems of multipath echoes and dispersion effect incurred by earlier multiple channel access techniques. Some of these multi-carrier techniques that took their genealogy from OFDM successes are the cyclic prefix-OFDM (CP-OFDM) and filtered-OFDM (F-OFDM), and many more. Analysis carried out in this research work focused on determining the transmission efficiency (TE) requirement for OFDM, CP-OFDM, and F-OFDM. The simulation results are achieved using desired transmission signal power (TSP), out-of-bound emission (OoBE), and peak-to-average power ratio (PAPR) as performance measures. In the aspect of desired signal power, the F-OFDM system achieved 3.1 dBm with a lesser processing time of about 16.7 µs, thereby outperforming CP-OFDM and OFDM systems, which obtained 2.3 dBm in 33.33 µs and 2. dBm in 66.7 µs, respectively. Other findings further showed that the F-OFDM have discredited CP-OFDM and OFDM in all scenarios for commercial deployment in wireless communications. These commercial multi-carrier techniques (CMT) of OFDM, CP-OFDM, and F-OFDM enabled a successful drive in meeting the long-term evolution (LTE), LTE-advanced (LTE-A), and 5G wireless networks requirement, respectively.

Wireless systems are often affected by distortions at the physical layer, which hinder reliable performance. A key challenge is the limited ability to suppress interference at the receiver, preventing the achievement of the asynchronous and orthogonal signal waveforms required for efficient utilization of Fifth Generation New Radio (5G NR) resources. Poor asynchronicity and orthogonality primarily result from Adjacent Channel Interference (ACI), Inter-Symbol Interference (ISI), and Inter-Channel Interference (ICI). This work addresses these challenges using the Feedback Filtered Orthogonal Frequency Division Multiplexing (FF-OFDM) channel access technique. Simulation results at 60 kHz Subcarrier Spacing (SCS) show that Hann--Kaiser-based FF-OFDM achieves superior mitigation of ICI (-47.0 dBm) and ISI (-39.0 dBm) compared to Hann-Chebyshev-based FF-OFDM (-45.2 dBm and -37.2 dBm, respectively). Conversely, Hann-Chebyshev-based FF-OFDM provides better ACI suppression (-27.5 dBm) than Hann-Kaiser-based FF-OFDM (-25.4 dBm). Similar trends were observed at 120 kHz SCS, where Hann-Kaiser-based FF-OFDM consistently produced more orthogonal signal waveforms, while Hann--Chebyshev-based FF-OFDM yielded safer asynchronous signal waveforms. The conventional and modified F-OFDM models were used as benchmarks for evaluating and validating these results.

Classifying the Additive White Gaussian Noise (AWGN) and the Primary User (PU) transmission signal has been a crucial area of research interest. For effective data transmission, the Secondary User (SU) detection threshold needs to be continuously adjusted to distinguish between the AWGN and PU signals at a very low Signal-to-Noise Ratio (SNR). The research analytically compared a proposed Vertical Hypothesis (VH) against the Horizontal Hypothesis (HH) of Spectrum Sensing (SS) within a Cognitive Radio (CR) framework. The developed model aims to increase SU sensitivity and specificity to AWGN and PU signal powers. The performance indices used are the composite hypotheses of spectrum detection Pd, missed detection Pmd, to obtain a false alarm Pfa, and blind detection, Pbd. The results obtained clearly showed that the vertical approach, when varied, achieved a threshold of about 0.96 blind detection and 0.04 false alarm at -10 dB channel gain between PU and SU, which is an improvement over the IEEE guideline threshold of 0.9 spectrum detection and 0.1 false alarm, respectively. The vertical approach proved more efficient at optimizing radio spectrum utilization, potentially improving transmission quality in Fifth Generation (5G) and Sixth Generation (6G) wireless systems.