Study On Capacity And Transmission Strategies For Interference Channels

With the widespread application of wireless communication technologies, the number of the users in wireless communication networks is increasing fast. And the users have higher requirement for the speed and reliability of communication. In order to meet these requirements, in many cases, some certain frequencies are reused within a certain scope, which makes some users communicate in the same frequency at the same time and interfered with each other. At present, the interference has become a universal phenomenon and a main factor restricting the efficiency of the wireless communication system. In order to study its influence on the performance of wireless communication system, to provide theoretical support, the guidance of performance bounds, and the design consideration of efficient information transmission strategies for the actual wireless communication system, the information theory researchers proposed various interference channel models and studied their capacities and transmission strategies that has become a hot research topic in today’s information theory.At present, capacities and transmission strategies for interference channels can fall into three classes. The first is capacities and transmission strategies for interference channels with point-to-point codes, which is the universal coding scheme used in the current communication system. The second is capacities and transmission strategies for interference channels with superposition codes, which is still at the experimental stage. The third is capacities and transmission strategies for interference channels with multi-user codes involved in interference alignment, which is still in the theory stage. However, all of them can’t determine the capacity for the general two-user interference channel. And we don’t know the gap of their performance.In this dissertation, capacities and transmission strategies for interference channels using point-to-point codes and superposition codes were analysed and studied. The gap of their performance is studied, which provides theoretical guidance to select the appropriate coding scheme, and provides theoretical foundation to design efficient infromation transmission strategies for the actual communication system. The analysis of their relationship can promote their understanding, simplification, and calculation, to determine their capacities and the capacity for the general two-user interference channel, and to extend to the multi-user interference channel or multi-user networks. However, there are many achievable rate regions for the two-user interference channel with point-to-point codes, and there is not the best one in all of them. To accomplish the whole work, the key problem is to determine the best achievable rate region for the two-user interference channel with point-to-point codes first. This dissertation includes the following five research contents:to establish the best achievable rate region for the two-user interference channel with point-to-point codes, to analyse the relationship between point-to-point codes and superposition codes, to compare the performance for two-user Gaussian interference channel using point-to-point codes and superposition codes, to expand the inner bounds with point-to-point codes and superposition codes to multi-user interference channels, and to design efficient information transmission strategies for the interference channel with point-to-point codes. the main research results are as follows in detail.For the two-user discrete memoryless interference channel, we proposed an achievable rate region for the channel with point-to-point codes. It combined the existing basic transmission strategy using point-to-point codes and contains the existing inner bound with point-to-point codes. It has become the best achievable rate region for two-user interference channel with point-to-point codes. Then, we applied it to the Gaussian interference channel. From two aspects of theory and the numerical results, we compared the rate region and the sum rate of it with those of some better region with point-to-point codes, which confirmed the point that it is the best achievable rate region for the two-user interference channel with point-to-point codes.The HK rate region and the CMG rate region is the best achievable rate region of two-user interference channels, their calculation and equivalence are difficult to understand. This dissertation put forward two methods to prove the equivalence of both the HK and the CMG rate region. One is that through the same coding method, we proved that successive decoding methods can reach the HK and the CMG rate regions. The analysis of the decoding order in successive decoding not only proves the equivalence of the two regions, but also illustrates the relationship between simultaneous superposition coding and sequential superposition coding. The second method reduces the expressions of the HK and the CMG rate regions from the perspective of removing the redundant expressions, which proved the equivalence between two regions.We compared the rate region and the sum rate of the best achievable rate region of two-user interference channel using point-to-point codes with those of the HK rate region, the best region for the two-user interference channel, is obtained by the superposition codes. From two aspects of whether to use time shared variables, we compared their performance, which provide a basis for selecting codes in practical communication system.Some capacity results of the two-user interference channel are very difficult to be extended to multi-user interference channel. For the cyclic interference channel, a class of special multi-user interference channel, where each user only interferes with one adjacent user, we proved that the rate region obtained by using successive decoding method is equivalent to that of the HK and the CMG rate region extended to the channel.Under the assumption of the two-user Gaussian interference channel with a point-to-point Gaussian code per user, a decoding strategy was proposed to achieve the maximal throughput according to channel gains and power parameters. The proposed decoding strategy was described in two equivalent forms of channel parameters, which shows that the best decoding strategy is not determined by the intensity of the channel gains but depends on both channel gains and power parameters.