Font Size: a A A

Research On Relay Selection Schemes In Large-scale Wireless Networks

Posted on:2017-03-09Degree:DoctorType:Dissertation
Country:ChinaCandidate:Y F C OuFull Text:PDF
GTID:1108330488457183Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
As a powerful tool to enhance the transmission performance of the networks, cooperative relaying technique attracts considerable attention of the researchers from universities and organizations. Users in a cooperative network can utilize the antennas of the relay nodes to form a virtual MIMO structure and thus obtain multiple independent fading channels to resist the effect of fading and interference. Considering this, the technique is very suitable to the scenarios where multiple antennas are not available in a single device. In a network with multiple relays, the relay selection protocol can allocate the node resource, balance the demand of users, and improve the network capacity. As a result of this, relay selection becomes one of the key research subjects in cooperative communication after single and multiple relay transmission protocols.However, a majority of the existing relay selection schemes only focus on the discussion in conventional scenarios in which fewer nodes are deployed. As the network scale and the number of relay node grow larger, those classical schemes are gradually unapplicable due to their selection goals, consumptions and complexities. Therefore, it is rather important to design new relay selection schemes according to the needs and constraints of the large-scale networks. Based on this, we concentrate on the relay selection schemes for large-scale networks in this dissertation. The main contributions are listed as follows:1, To meet the waiting-time demand of the user in the large-scale Cognitive Radio Net-works (CRN), a Stackelberg game based relay allocation scheme is proposed. Instead of introducing an extraneous secondary network, the proposed scheme divides the users to tem-porary primary and secondary groups to form a self-generated CRN, in which the temporary secondary users can serve as relays to win unused time fractions. The corresponding coop-erating strategies are determined with the utilization of Stackelberg game, and the existence and uniqueness of the equilibriums are proved as well. The user-relay allocation decision is given by Hungarian method. Finally, the validity of the proposed scheme in waiting-time reduction is confirmed through simulations.2, Consider the limitation of the cell-size in a cellular network, future networks may consists of uncountable numbers of random distributed relays, making it impossible for users and relays to collect the complete information from each other. In such a scenario, a slotted ALOHA based random relay selection scheme is proposed. According to the relation of the relay number distribution in a region and the area of that region, the distribution of the relays that satisfies certain demand of the user is determined first. These relays randomly access the channel through the slotted ALOHA protocol to compete for the selection. The failure probability and the expected transmission rate of the scheme are analyzed. Compared with the ideal selection that requires complete relay information, the proposed scheme is implemented in a blind manner while maintaining almost optimal performance.3, To further improve the performance and reduce the network information request of the random selection, a greedy mechanism based random relay selection scheme is developed. Without the need for the relay density information, the user only broadcasts its minimum transmission request for once. A greedy selection mechanism is introduced with which a user can wait for an even better relay when a suitable one is successfully stored. Moreover, since the number of compete relays may reduce during the random access, the adaptive access probability of a relay is determined for the highest success probability. Given the state transition diagram, the average number of the selection slots and the failure probability of the scheme are analyzed. Simulation results shows that the proposed scheme outperforms other random access selection schemes.4, Due to the fact that random selection cannot guarantee the optimal relay, its perfor-mance is always sub-optimal. To remedy this, a splitting algorithm based optimal relay selection scheme is proposed. By introducing the concept of stochastic geometry, the one-to-one relation between the distribution of the available relay number and the transmission request is established. Furthermore, the selection phase of the CReS scheme is divided in-to two sub-phases according to different distributions of the relay number in boundless and bounded regions. A switch rule is proposed when several relays collide in the first sub-phase, followed by the second sub-phase to resolve the collision with the splitting algorithm. The optimal solution that requires minimum selection slots for all sub-phases is derived and the expression of average number of selection slots is given. In addition, a sub-optimal solution based on the curve approximation is also explored to reduce the computational complexi-ty. By simulations, the proposed scheme shows a better applicability and performance than conventional information-based or random schemes.
Keywords/Search Tags:wireless communication, large-scale networks, relay selection, Stackelberg game, stochastic geometry
PDF Full Text Request
Related items