Research On Video Transmission Method Based On OFDM Channel Characteristics

In recent years, how to transmit video signals is a hot and challenging problem. The research in this thesis mainly focused on the video transmission scheme based on channel characteristics in the applications of both unicast and multicast. The main works and innovations are as follows:1. An adaptive layered video transmission based on channel characteristics is proposed. The scheme provides a video transmission method which gives different-importance data diversity protection based on practical channel characteristics in order to improve the reconstructed video quality. In our proposed scheme, the sender employs the scalable video coding as the source coding to obtain the layered video data. Next, the layered video data adopts different modulation schemes, and then subcarriers and power are allocated to layered video bits with the received channel characteristics. In this unicast case, our proposed scheme achieves superior performance, and then the proposed scheme is generalized to the multicast case. The experimental results show that our proposed scheme provides the different receivers with different video quality matched with their own channel conditions, which realizes the adaptive layered video transmission.2. A channel-aware video Softcast scheme is proposed. The scheme provides a method that addresses the video quality degradation phenomenon in the practical orthogonal frequency division multiplexing(OFDM) channel, which can overcome the disadvantages of raw OFDM in current Softcast scheme in order to improve the video quality. In this proposed scheme, the encoder applies discrete cosine transform(DCT), power allocation, hadamard transform, and then the resulted packets are delivered into OFDM channel. In OFDM, the subcarriers with better channel conditions are allocated to transmit video signals with the channel characteristics fed back from the receiver, while the rest subcarriers are discarded; On the decoding side, modified linear least square estimator(MLLSE) is first applied to reduce the influence of noise, and then zero forcing decision feedback equalization(ZF-DFE) is used as the decoder. The experimental results show that the proposed scheme can accommodate the heterogonous receivers, and realizes the graceful degradation of video quality.3. A hybrid digital-analog video transmission scheme with channel characteristics is proposed. The proposed scheme combines the advantages of digital coding and analog coding. Based on the channel characteristics, the subcarrier allocation is used to realize the transmission of hybrid digital analog scheme, and power is allocated using water-filling algorithm realizing the further performance improvement. The simulation results show the advantages of our proposed scheme in both unicast and multicast.