The 4G network implements Orthogonal Frequency Division Multiplexing (OFDM) as its multiplexing method. Using the said method, the network is expected to be able to cope with severe channel conditions without the use of complex equalization filters. However, very high Peak to Average Power Ratio (PAPR) and Out-of-Band (OOB) emissions are still being experienced where the system’s throughput and spectrum efficiency are significantly reduced. Maxis, one of the major telco operators in Malaysia, reported that they experienced a 33% in speed reduction for its 4G network throughout the year 2019. The currently dedicated spectrum for 4G is reaching its limits and there is a large as well as growing demand for wireless access and applications. The strategies for more intelligent use of the spectrum are urgently required. Thus, in this research, the overall objective is to assemble a new augmented multiplexing method and an enhanced spectrum management technique. The newly proposed multiplexing method for the future 5G is called the Filter Bank Multicarrier (FBMC) with a Lowpass Windowed Finite Impulse Response (FIR) filter (LWF-FBMC). The improved interference management procedure is identified as enhanced dynamic spectrum allocation (E-DSA) that incorporates a cooperative type of Game Theory (GT) called the Nash bargaining solution. These two proposed solutions are very much hoped to be able to maximize the throughput requirements of 5G. In addition, it is expected to address the interference mitigation aspects. This study utilized three types of Software Defined Radios (SDRs) for designing and analyzing the configurations. The SDRs used were the LabVIEW Communications System Design Suite (LV Comm), GNU Radio, and MATLAB software. For hardware implementation, the National Instrument’s Universal Software Radio Peripheral reconfigurable I/O (NI USRP RIO) was used. In the methodology, the construction of the newly proposed LWF-FBMC was done using the LV Comm. The power spectral density (PSD) analyses between OFDM and LWF-FBMC were carried out to compare their level of OOB emissions by analyzing their power spectral densities. In developing the E-DSA, the effectiveness of the DSA for 4G was first analyzed using GNU Radio. The algorithm was then enhanced to be E-DSA and tested in an urban 5G heterogeneous network scenario that involves both macrocell and microcell users. The simulated configurations were then integrated with the NI USRP RIO transceiver to compare their power spectral densities in real-time. For the results, it was shown that the LWF-FBMC can achieve higher spectral efficiency than the OFDM by 15.4%. The newly configured E-DSA scheme can improve the 5G network’s throughput by 104% when compared with the DSA for 4G network. The NI USRP RIO’s results also indicated that the spectral efficiency of LWF-FBMC is higher by 50% than that of OFDM. However, for the simulation-hardware integration part, only the multiplexing methods were compared without analyzing the effects of implementing E-DSA due to the limited time and resources. For future works, it is best to examine this as well. In conclusion, with these improvements, it can be said that both methods, the LWF-FBMC and the E-DSA can help in alleviating the interference for the up-and-coming 5G system.

Energy harvesting wireless networks are one of the most researched topics in this decade. With a radio frequency (RF) energy harvester (EH) embedded, the sensors can operate for extended periods. Thus, providing sustainable solutions for managing massive numbers of sensor nodes (SN). There are different scheduling methods for information decoding (ID) and energy harvesting (EH) in the literature. These scheduling techniques can be classified into time switching (TS) and power splitting (PS). TS alternates between EH and ID on a temporal basis, receiving data just half of the time. In contrast, PS splits the received signal in theory between ID and EH circuitries without regard for the power requirement differences of the two circuitries. This thesis aims to develop an energy harvesting system with a dynamic power allocation transmitter and dynamic power splitting receiver between information decoding and energy harvesting circuitries for WSN to increase the energy harvester output. The presented receiver architecture integrates input signals from several RF sources and divides them between the EH and ID circuits. In addition, by moving the ID load into a separate circuit, the split design enhances the harvestable power at the RF energy harvester. DPA-SWIPT is implemented at the transmitter, where the ES is transmitted using an unmodulated high-power CW signal centred on the carrier frequency, while the IS is transmitted using a low-power signal around the carrier frequency. In TS and PS, the ES is conveyed on a modulated wave, resulting in increased interference with external networks. In contrast, in DPA-SWIPT, the high-power ES is constrained to a narrow band at the carrier frequency, resulting in less interference with bordering networks. Various system parameters were discussed, including the EH circuit's voltage multiplier output and ID data rates. As a result, the split receiver design demonstrated a considerable increase of more than 15 dBm in harvestable power level compared to the combined receiver. Moreover, this thesis also aims to improve the peak to average power ratio (PAPR). Hence, the PAPR of several wavelet methods are investigated, with the wavelet-based modulation technique outperforming fast Fourier transform (FFT) orthogonal frequency division multiplexing (OFDM) substantially where Haar wavelet scored a gain of almost 5dBm. As a result, it is stated that wavelet modulation is an excellent contender for implementation at the SN, where energy efficiency is critical. Ultimately, the increase in the input power level of the EH circuitry coupled with enhancing the energy efficiency of the WSN nodes marks an important milestone toward achieving fully autonomous WSNs.