Demonstration of Spatial Interweave Cognitive Radio
Master thesis
Permanent lenke
http://hdl.handle.net/11250/2370100Utgivelsesdato
2010Metadata
Vis full innførselSamlinger
Sammendrag
In the last few decades the number of wireless communication systems has grown exponentially making the electromagnetic spectrum more and more crowded. One approach to solve the problem is through Cognitive Radio. This thesis deals with the implementation of one technique within CR, Spatial Interweaving. Well known techniques exploit either idle time or fre- quency domain, spatial interweaving is aimed at the spatial domain. Through adaptation of Eurecom s Open Air Interface (OAI) platform, this thesis implements the spatial interweaving approach. The OAI already im- plements Long Term Evolution (LTE) Release 8, and this is extended and exploited for spatial interweaving. The contribution to the OAI to enable this is significant and enables research within other areas than the focus of this thesis. Although real time demonstrations were not performed, well founded channel models are used for simulation. The simulation environment is closely related to that employed with the hardware RF. Simulations show that a secondary system can coexist a long with a pri- mary system when the Signal-to-Noise Ratio (SNR) is above 5 - 10 dB, de- pending on the channel model. It is achieved by the secondary system apply- ing a spatial beam forming vector, created from reciprocal channel estimates. This is in the case where all channel links experience the same path losses and both systems transmit at the same power. There are many variables that will affect the performance, of which most are part of characterizing the channel. The simulation complexity would increase exponentially with the number of variables. In this thesis limited cases are simulated to show the extremes. Although this demonstration did not make it to a real time implementa- tion, simulations show that a secondary system can coexist with a primary system, through the spatial interweaving approach. More simulations should be made to get the full picture.