Research On Non-ideal Cooperative Communication System And Its Joint Iterative Decoding Performance

With the rapid development of wireless communication technologies, research on various cooperative communication techniques attracts great attention. This technique offers a mobile terminal with a single antenna as much benefit on diversity as that of a MIMO system. The basic idea is that, in the multiuser environment, a mobile terminal with one antenna can take the advantages from the antennas of other users to subsidarily help the transmission of information-bearing symbols by a proper way, which would eventually result in a virtual MIMO system.Regarding the research areas of cooperative communications, much current research is focused on the turbo/LDPC-based ideal relay coded cooperation under the assumption of error-free decoding for the incoming signal from the source at the relay. Actually such a hypothesis is never possible due the noisy channel between the source and the relay, which would inevitably cause the error propagation for the second encoding at the relay. Thus, it is really a challenge research topic of both practical and theoretical value to design an effective coded coperation to get more benefits from the coding and diversity gains in a non-ideal relay cooperation.The main contributiuons of the thesis are basically summaried as follows:1. Fully understand Shannon channel capacity and noisy channel coding theorems. More importantly, we have studied several fundmental cooperative communication channels and their channel capacity from the viewpoints of random coding.2. Under the point-to-point communication environment, we studied the encoding and decoding principles of turbo and LDPC codes as well as their performance over the noisy channels.3. The author proposes ideal and non-ideal turbo-based relay cooperation schemes, where a convolution code is employed to encode the information bits at the source, and a turbo decoder is adopted to iteratively decode the two incoming signals from the source and relay at the destination. Theoretical analysis and the numerical simulation show that the performance of coded cooperation is much better than that of a simple point-to-point communication without cooperation under the identical condition of the same transmitted power. Futhermore, we carry out the research on the non-ideal coded cooperation and several subsidiary methods are also proposed to deal with the non-ideal scenarios.4. Investigate in-depth the ideal and non-ideal irregular systematic LDPC-coded relay cooperations. We introduced a new approach of effectively joint iterative decoding and novel pre-decoding. In the encoding stage of the cooperation, two distinct irregular systematic LDPC codes are applied by the source and relay node, respectively, which results in lower correlation between the two generated parity-check sequences. This encoding strategy is quite similar as that of a turbo encoder using a random interleaver. In the decoding stage of the cooperation, the joint iterative decoding is effectively performed by means of iterative deocding that is quite similar as that of iterative turbo decoding based on the extrinsic information exchanged between the two parallel constituent decoders. Information transmission between the two subdecoders is completed during each decoding iteration. Although the non-ideal channel between source and relay node may bring some loss to the overall cooperation system, however the proposed new strategy clearly enhances the system performance by effectively combining the benefits of coding and spatial diversity into one synergy using the introduced joint iterative decoding technique.