Study On LTE Co-frequency Co-time Full Duplex Self-interference Cancellation Technology

As we all know,spectrum resources are extremely precious in the field of wireless communication. With the development of wireless communication technology, the requirements for wireless access have been growing exponentially. Therefore spectrum resources have been stretched. Within a certain frequency band, a node can access more users and can meet each user higher connection speed, which is our goal. The traditional duplex modes, including TDD(Time Division Duplex) and FDD(Frequency Division Duplex), sacrifice efficiency or spectrum resources. Therefore they could not complete the true duplex. A long time in academia, it is widely believed that a node can not simultaneously transmit and receive signals in the same frequency band, because the transmission signal can cause a strong self-interference in the receiving antenna in the same frequency. But recent studies have shown that CCFD(Co-frequency Co-time Full Duplex) is likely to be achieved. Therefore, we studied the co-frequency co-time full fuplex communication mode in this thesis. It is able to achieve that a node can transmit and receive signals at the same time within the same frequency band. In this thesis, we used the multi-layered interference cancellation architecture: antenna interference cancellation, analog cancellation and digital interference cancellation. Using this architecture we were able to cancel the local self-interference, and then we would achieve more efficient utilization of spectrum resources.Firstly, we studied the CCFD performance advantages over the prior technologies.We discussed the key issues and difficulties to achieve CCFD: to cancel the local self-interference in the same frequency. And we studied the changes of local self-interference from transmitter to receiver.We studied the three-antenna technology to achieve antenna cancellation. We used amplitude mismatch of two-way signals and RX antenna placement error to indicate the impact of the errors on the the performance of antenna cancellation.And then the simulations have shown the error impact of performance. To further reduce the local self-interference signal power, we also studied the analog cancellation technology.This method was similar to sinc difference. Used the local analog signal, we constructed N signals whose time intervals were equal. We saw N-channel signal as the Nyquist sampling sample values, and regarded the coefficient of each signal as the weight of digital samples. And we simulated this method and got its performance curve.In order to fully cancel the local self-interference signal, we discussed how to design nonlinear digital interference cancellation algorithm. First, we discuss ed a widely used power amplifier parallel hammerstein model, and modeled antenna cancellation and analog cancellation as FIR filters. Therefore we could model the nonlinear baseband self-interference channel, and obtained the self-interference signal. We used the local digital baseband signal to make nonlinear transformation matrix. In the digital signal receiving end in order to make the received signal minimum, we used the least squares to estimate the transformation matrix coefficients. Therefore we could use the constructed signal and self-interference to finish cancellation.