Recently, the space-time block codes (STBCs) were suggested to use in wireless relayingudnetworks (WRNs), denoted as distributed-STBCs (D-STBCs). This is to exploit effectively theudspatial diversity and hence improve the link reliability. In addition, this usage may increaseudthe network’s spectrum efficiency as it allows concurrent transmission from the relayingudnodes. However, these networks encounter numerous issues that limit their wide practicaluduse. This thesis addresses three critical issues of WRNs, and proposes solutions for each partudindividually.udIn Part I, WRNs are considered under imperfect synchronization. In the literature, most researchudtends to assume perfect synchronization among the cooperative relays. Unfortunately,udthis level of synchronization is almost impossible to achieve in real communication networks,udand this introduces a significant performance degradation if imperfect synchronization isudpresent in the network. This part includes mathematical models that are derived for WRNs,udeither one-way or two-way, under imperfect synchronisation conditions. Unlike existingudmodels, this model provides a simple method of evaluating the problem for variant networkudconfigurations. In addition, this model considers the WRNs with N relays, each is equippedudwith Ra antennas, where N, Ra ∈ N+. With respect to current literature, the contributions ofudthis part are : (1) both the existing PIC and SIC based detectors, which were proposed forudspecific network configurations instances, are extended here to work with the general model.ud(2) an enhanced interference cancellation based detector (EIC) is proposed. These proposeduddetectors shows significant performance improvement compared to the conventional detectorudunder imperfect synchronisation conditions. In addition, the proposed EIC detector providesudbetter improvement due to the designed interference cancellation process. It reducesudthe reliance on low-performance symbols and it benefits from interference components ofudcurrently-detected symbols using a modified maximum likelihood (ML) scheme. Accordingly,udan extra performance improvement is achieved, particularly in the first iteration.udPart II considers the issue of designing D-STBCs for WRNs with an arbitrary number ofudrelays. It has been shown that the reliability of WRNs increases by adding more relays as a result of more communication paths becoming available. Unlike most existing D-STBCs, this part proposes two high rate coding schemes to accommodate an arbitrary number ofudrelays, while retaining low decoding complexity at the destination. The first scheme, full-rateuddistributed space-time block coded-joint transmit/receive antenna diversity (D-STBC-JTRD),udis proposed for AF WRNs. Its code rate is independent of the number of relays and hence noudcode rate loss is incurred as the relays number increases. In addition, this scheme deploysudthe same encoding matrices at every relay; this eliminates the need for additional networkudoverhead to coordinate the code generation by the relays. In other words, there is no needudto interrupt the transmission if a relay has been up/down. The second scheme aims to finduda flexible trade-off between reliability and code-rate that can be offered by DF networks.udTowards this end, a method to construct a D-STBC that is combined with spatial modulationud(SM), denoted as D-STBC-SM, is proposed. This method is not restricted to a specific numberudof relays and can be constructed as necessary. In addition, a novel adaptive transmissionudprotocol that uses the constructed codes, is proposed to achieve higher space diversity gain,udeven with relays equipped with a single antenna. Unlike most existing schemes, this protocoludoffers a throughput that increases as the number of relays increases. Moreover, the offeredudthroughput is achieved using the same total average transmit energy, as only N0 of the Nudavailable participating relays are active at any given time.udIn Part III, the multi-user interference of WRNs is considered. Two transmission protocolsudwith an interference cancellation scheme are proposed: the concurrentS−R−D-PICR,D protocoludfor DF WRNs and the concurrentS−R−D-PICD protocol for AF WRNs. Unlike existingudprotocols, these protocols allow the concurrent transmission in both phases of the transmission.udThus, high spectral efficiency is offered while maintaining low decoding complexity.udThis low decoding complexity is maintained due to the adaptation of the partial interferenceudcancellation group decoding (PICGD) approach for WRNs, which was initially proposed byudGuo, et al., for point-to-point (P2P) communication link. For a WRN consisting of J usersudequipped each with Ja-antenna, a single half duplex (HD) Ra-antenna relay, and M-antennauddestination, the concurrentS−R−D-PICR,D protocol achieves the interference-free diversityudgain (i.e., Ra × min {Ja,M}) without imposing any conditions on a node’s antenna number.udThe interference-free is the diversity gain achieved, assuming that each user in the networkudis transmitting solely without experiencing any interference from other users, hence it isudconsidered as the natural upper bound of the diversity gain in multi-user WRNs. Similar toudmost exiting protocols, this protocol requires the CSI of the users-relay links at the relay. Inudcontrast, the concurrentS−R−D-PICR,D protocol achieves a diversity gain of Ja ×M, given that Ra > 8, while the CSI is required only at the destination. Although the diversity’s upper bound is not achieved, this protocol uses a simple relay as no CSI or encoding is requiredudat the relay. In addition, and unlike the existing protocols, the achievable diversity gain isuddetermined by both Ja and M and it is not sacrificed while J is increased. This part alsoudestablishes sufficient conditions for an STBC to achieve the prior mentioned diversity gains,udwhen the PICGD approach is employed by multi-users WRNs.
展开▼
