Research On Key Techniques Of FBMC Based ROF System

With the increasing demand for wireless bandwidth in communication, wireless mobile communications need to have characteristics of high-speed, broadband and high-frequency. OFDM-ROF means that Filter using rectangular pulse Orthogonal frequency-division multiplexing (OFDM) radio is combined with optical carrier (Radio Over Fiber, ROF). It is gaining widespread attention because of its high-speed, high-capacity, low-cost advantages. With further research, it has been found that these two systems have Peak-to-average Power Ratio (PAPR) and load inequality problems, which have drastic effect on the system performances. To solve these problems, this thesis analyzes the key technologies of ROF PAPR suppression and load balancing of ROF systems. The main work and innovative results are summarized as follows:(1) Two joint algorithms are proposed to reduce PAPR problems of OFDM-ROF system. Compared with the traditional method, these algorithms achieve good PAPR suppression effect, reduce the computational complexity as well as ensure improved bit error rate performance. OFDM, as a special case of FBMC, has been widely used in various communication systems. Several algorithms for PAPR reduction of the system have been proposed. Amongst these, clipping method and Partial Transmit Sequence (PTS) are typical methods. Clipping method is relatively simple, but has high rate of errors. PTS method has good performance, but has the high computational complexity. In order to reduce the bit error rate of clipping methods, two improved algorithms are proposed including Quantification clipping method and recoverable clipping method. By setting multiple thresholds, different signal amplitude quantization is achieved. This will reduce signal distortion caused by clipping. The latter will compress the high power signal amplitude and a sequence is used to mark the compression. Then two algorithms combine PTS to reduce PAPR of OFDM-ROF system. Finally, simulation is done to test the proposed algorithms.(2) An improved PTS algorithm to reduce PAPR of FBMC-OQAM-ROF system is proposed. Compared with the traditional method, the algorithm achieves good results while reducing the computational complexity. FBMC-OQAM, as a typical representative of FBMC, is considered a multi-carrier OFDM that can substitute OFDM technique. But, this technique also has a high PAPR problem. In order to reduce PAPR, this thesis, drawing on traditional thought on PTS algorithm, considering FBMC-OQAM adjacent blocks overlapping problems, proposes a forward-PTS (F-PTS) algorithm. The algorithm divides FBMC-OQAM data blocks into two parts. When the data blocks of each part are chosen as the optimal transmission sequence, the former data blocks are considered according to the impact of different rules. Finally, a simulation model is used to test its performances.(3) An improved SLM algorithm is put forward to reduce PAPR of FBMC-OQAM-ROF system. Compared with the traditional method, with the premise that the candidate sequence is the same, the method can achieve better PAPR results. Traditional selective mapping (SLM) method is another multi-carrier systems which can effectively reduce PAPR, but suffers the problem of high computational complexity. Especially with the increasing number of the phase sequence, the computational complexity will be much greater. In this thesis, a linear relationship between the SLM algorithm candidate transmission sequence is proposed. The proposed algorithm is an extension of the candidate transmission sequence in the system to reduce the PAPR of FBMC-OQAM. Finally, a numerical simulation model is used to test its performance.(4) A load balancing algorithm based non-cooperative differential game is proposed. The algorithm has reasonable allocations of each wireless access point (AP) as well as avoid the "ping-pang effect". This thesis studies the problem of load balancing of ROF system. By analysis the advantages and disadvantages of the existing load balancing algorithm,a load balancing algorithm based non-cooperative differential game is proposed. Non-cooperative differential game theory can detect the real-time situation of loads for each AP and distribute the optimal load for each AP from the overall consideration. Finally, a numerical simulation model is used to test its performance.