Research On Cognitive Channel Coding And Its Achievable Rate Region

With the rapid development of wireless communication technology, radio frequency resource is becoming much scarcer. However, most of the authorized spectrum has not been fully utilized. Cognitive radio (CR) technology can improve spectrum efficiency, and ease the condition of spectrum scarcity. CR has become a hot topic in recent wireless communications research.Cognitive channel is a novel wireless communication model based on cognitive radio theory. It is an extension of interference channel, and classifies users into primary user (non-cognitive user) and secondary user (cognitive radio). In order to improve spectrum utilization efficiency, secondary users are allowed to utilize primary user's spectrum, or to collaboratively code with primary user, without impairing the service quality of primary user. In the perspective of information theory, there are already some results on its achievable rate region and outer bound, but the exact capacity region of a general cognitive channel is still unknown.This thesis focuses on a two-user overlay cognitive channel model. Coding modes and theoretical methods with which to get an achievable rate region of cognitive channel model is firstly summarized, including GP coding, rate splitting, dirty paper coding, superposition coding, and an opportunistic jointly typical coding method. Based on this, a new achievable rate region in discrete memoryless channel is obtained by modifying the model with rate splitting. Then the channel alters to Gaussian condition and some numerical results are calculated. A comparison between our result and some existing results shows that an extension is achieved, whose reason lies in cognitive user's collaborative coding with noncognitive user. What's more, a new model of cognitive channel with feedback is proposed. Its rate property is preliminarily analyzed, showing that feedback can improve cognitive channel's rate, for there exists a variation of multiple access channel in this modified cognitive channel. At last, this thesis discusses some important issues in cognitive channel, such as the acquisition of unusual side-information, the introduction of degree of freedom and interference alignment and collaborative coding, etc.Since the opportunistic jointly typical coding adapted in the proof of achievable rate requires the length of a codeword approach to infinity to achieve a target information rate, the final result can only be achieved in some ideal conditions. Thus, we only get a result in the perspective of information theory. The coding methods we used in the proof cannot be adapted in practice directly. From this point of view, it is difficult to simulate the coding process. Only some inequality constraints and numerical results of achievable rate region of cognitive channel are presented.