| The rapid growth of mobile Internet has promoted the development of device to device(D2D) communication technologies. D2 D system is a distributed system that has no control center, where the wireless devices spontaneously build networks to make direct communications between devices. The Flash Lin Q technology is developed by Qualcomm for D2 D communication. In this protocol, nodes do not immediately retract after detecting the interference signal. Instead, node determines to retract or not depending on the co ndition that the SIR meets a threshold. Therefore, the system gains a very high spectral spatial reuse.In the stage of link scheduling of Flash Lin Q, the link priority is randomly assigned, which does not take full advantage of link diversity gain to imp rove system throughput and results in the increasing of link scheduling delay to some extent. DO-Fast protocol described in this thesis uses two improved methods to solve this problem. The first method is able to sort links according to the channel quality, and give high quality links more scheduling opportunities. The second method is using priority scheduling mechanism for links polling between groups and within- group, ensuring users’ fairness from large-scale and small-scale range. In this thesis, the overall design of DO-Fast protocol is described, frame structure is analyzed, the cost of the protocol is estimated, and the implementation process is given. Finally, the theoretical analysis is verified to enhance the role of the protocol with respect to the original protocol on network performance.According to the description of DO-Fast protocol, the thesis has designed a realizable scheme of paging algorithm based on FPGA software defined radio platform. In the scheme, OFDM baseband, OFDMA baseband and MAC layer system paging algorithm design are needed to be done. The design of OFDM system and the OFDMA system references IEEE 802.11 a standard. The main idea of the MAC layer paging algorithm is sorting and grouping links according to the quality of channe l in descending, to prepare for the link scheduling in business connection slot. DO-Fast is a time synchronization protocol, where the whole network nodes require precise synchronization. FPGA with parallel computing capabilities and certain processing delay is very suitable to realize synchronization protocol. FPGA development board used herein is Nutaq Perseus601 x and the RF transceiver module is Microsystems Radio420. These two devices is powerful, and can easily communicate with the PC through Nutaq development platform. Therefore, these devices ful y meet the design requirements.This thesis has completed the design of OFDM and OFDMA physical layer baseband system, overall block diagram design of the MAC layer of DO-Fast protocol. Programming of physical layer, programming of peer discovery algorithm in MAC layer and programming of paging algorithm in MAC layer have been completed in Verilog HDL language. Actual test of physical layer has been implemented. And the result is correct. The programs of peer discovery algorithm and paging algorithm in MAC layer have passed through the simulation. Due to the large project workload, the actual testing of the program of agreement paging algorithm has not been completed. |
PDF Full Text Request