dc.description.abstract | One of the applications of 5th-Generation Mobile Communication Technology (5G) is ultra-reliable and low latency communications (URLLC), which focuses on features such as low latency and high reliability. If one wants to realize URLLC service in the unlicensed spectrum, he/she may face many challenges, such as the interference from heterogeneous Wi-Fi networks. Based on our previous researches, which proposed the channel reservation mechanism (CRM) over unlicensed spectrum, this thesis attempts to utilize a full-duplex radio repeater (FDRR) platform to achieve channel reservation by continuously emitting the Busy Tone (BT) in unlicensed channel in order to occupy that channel, meanwhile these FDRRs are able to transmit control/data signals to each other without the disturb from Wi-Fi networks. The experiment is used to validate the feasibility of developing the proposed CRM for realizing the URLLC services in unlicensed spectrum.
This thesis uses a number of FDRRs to play the roles of gNodeB (gNB) and user equipment (UE) in the 5G system, aiming to achieve channel reservation in unlicensed spectrum. All FDRRs emit ZC sequences of the same length but with different root indices as BTs, one index for one FDRR, and they use matched filters to effectively distinguish the source FDRRs by matching known ZC sequences. As a result, the gNB and UEs can transmit their BTs and receive the other BTs from other devices simultaneously. At the same time, the channel accessed by Wi-Fi devices will be temporarily occupied by the proposed CRM. Through Iperf (PCC/PCS) testing, the throughput of the Wi-Fi channel is used as an evidence to show that whether the channel is used by Wi-Fi devices or not, thus validating the capability of channel reservation.
During the design phase, we utilize the Quartus tools to program in Verilog and then program the FPGA within the FDRR. Also, we employ the Keil_V5 to write firmware code for the MCU to process sophisticated calculation, aiming to conserve FPGA resources. Finally, we utilize MATLAB to confirm the correlation between the IQ signals extracted from the FDRR via JTAG and then identify the received ZC sequences. Experiments have been established to validate that multiple FDRRs are able to communicate with each other without the disturb from homogeneous and heterogeneous interferences during field tests, which in turn also confirms the ability of channel reservation. | en_US |