The Modeling Of Large-Scale Fading Correlation In Wireless Networks Based On Outdoor Measurement Validation

Cooperative communication technique offer single-antenna users with the benefits of spatial diversity gain by allowing them sharing their antennas, bandwidth and power, and collaboratively transmitting and receiving their information. It also enhances network coverage performance and reduces total power consumption. As a novel transmission scheme, it shows great potentiality in almost all kinds of wireless communication networks, such as wireless sensor networks, wireless mesh network and cellular communication systems, and is gaining more and more attention all around the world.The wireless channel suffers from fading, which is an underlying limitation of wireless networks. In cooperative communication systems, a signal is transmitted to its destination through multiple multi-hop paths, as a result, the communication performance is determined by different links. Shadow fading, caused by major obstructions between transmitter and receiver, is a major composition of large-scale fading. If two links are geographically close to each other, the shadow fading on them are very likely to be correlated. This kind of correlation will strongly affect the performance of coverage, connectivity and cooperative diversity gain, and also influence the design of cooperative partner selection and routing scheme. The main purpose of this dissertation is to study the correlation of shadow fading between different links and model such correlation both by theoretical analysis and experimental measurements.First of all, the large-scale fading of wireless links is briefly introduced and its impact on network connectivity and coverage is studied. An indoor wireless channel measurement was carried out and the analysis based on the measurement data confirms the existence of shadow fading correlation between different links. After that, two relevant statistical models for shadow correlation are explored in a four-node multi-hop wireless network, while the results show that neither can be applied directly in such kind of network.A modeling method for shadow fading correlation based on spatial loss field is then introduced. Starting with the assumption that shadowing losses on the links are the results of a spatial loss field, the statistical model of this field is presented. The analytical relationship between shadow fading and the field is given so that the correlation of shadow fading between any links can be quantified. After all these works, a statistical joint large-scale channel model for arbitrary multiple links in a wireless network is proposed and verified to be reasonable. Finally, with the purpose of validating the proposed model, an experimental measurement for multi-hop wireless channels is carefully designed and carried out in an outdoor environment. From the measurement results, the parameters of the proposed model are extracted and the realistic shadowing correlations are provided for pairs of links with identical relative geometries. The agreement of these realistic correlations and the ones calculated by the model confirms the availability of the proposed model.