Research On Anti-Interference Schemes In Interference Channels

Interference channel (IC) is the elementary model in multiuser communications. Sinceinterference is the most fundamental factor in IC, it becomes the key problem in multiusercommunication system design that how to resist interference effectively. IC can be divided into twoclasses according to whether there is side information at the transmitter side. In this paper, theanti-interference methods and their performance are researched both for those two classes.Cognitive radio co-exist channel is a new kind of IC with side information. Based on Gaussianrandom codes, the theoretical performance limit of this channel described by the minimal value ofthe relay factor is derived. However, due to the lack of knowledge about the performance limit ofstructured codes in this channel, present researches on the implementation of coding schemes are allbased on the theoretical results, while never consider whether the theoretical result can be achievedwith structured codes. Therefore, it is in urgent need to derive the performance limit of structuredcodes in this channel.Among various kinds of structured codes, nested lattice codes is known to be one of the bestand is naturally suitable for cognitive radio co-exist channel. Therefore, in this paper, we implementboth for primary and secondary user the nested lattice codes that can achieve their correspondingchannel capacity respectively, then derive the minimal value of relay factor under this circumstance.Comparing with the theoretical one, the derived minimal value is inferior.The performance of nested lattice codes in cognitive radio co-exist channel is quite differentfrom that in many other coding problems where nested lattice codes act perfectly. Analysis resultsshow it is mainly because that the primary user can never adapt its decoding scheme to the changein channel environment caused by the secondary user’s appearance. However, further researchshows that even the primary user has the ability of adaption, the system performance can still neverachieve the theoretical one completely. Analysis indicate that comparing to Gaussian random codes,nested lattice codes will introduce more constraints to relay factor which makes nested lattice codesinferior to Gaussian random codes on performance in this channel.K-user Gaussian interference channel widely models the multiuser communication systemsexist in our lives. Interference alignment (IA) has shaken orthogonal schemes from dominate to itsfoundations in this channel. Theoretically, IA can achieve at most 1/2 degrees of freedom per userin this channel. Unfortunately, implement conventional IA requires many conditions that are notfeasible in reality.Chordal distance is considered to be one of the best measures in Grassmann manifold. Take chordal distance to measure the effect of IA can relax its constraint on alignment. At the other hand,combined with user selection, opportunistic beamforming (OBF) techniques possess advantages inboth high performance and low complexity. In this paper, we propose both a new user selectioncriterion named chordal distance differentials maximization and a new post-processing method atthe receiver side named smart post-processing matrix selection. With the two mentioned, the idea ofIA and the technique of OBF is combined and a robust opportunistic interference alignment (ROIA)scheme in 3-user MIMO IC is formed. The ROIA method not only permits high robustness ininstantaneous performance of the system, but also highly improves the channels capacity. Comparedto these two aspects, the increase in complexity is acceptable.