In cognitive radio network, the licensed spectrum for the primary user can be accessed in an opportunistic manner by secondary user, or unlicensed user. As a key technology of cognitive radio, spectrum sensing has an irreplaceable position. In this paper, we proposed a parallel CNN-LSTM network based deep learning algorithms for spectrum sensing. As much modulated signals and noise data as possible are generated to train the model to accommodate detection of multiple types signal. Various experiments are performed to prove the effectiveness of proposed method, and requiring no prior knowledge about the information of licensed user or channel state. The simulation results show that the model can detect multiple modulation types under a large scale of SNRs, especially in low SNR.

In this paper, stochastic delay differential equation (SDDE) and its application on communications are discussed. Based on SDDE, a novel communication scheme-delay time modulation (DTM) is proposed. In this modulation scheme, the information signal is conveyed by the delay time of a delayed linear Langevin equation, which exhibits a linear relationship with the variance of the SDDE system output. The information signal can be retrieved at the receiving end by estimating the variance of the received signal. To evaluate the performance of DTM scheme, normalized mean square error (NMSE) in additive white Gaussian channel is derived for analog information transmission, as well as BER (bit error rate) for binary information communications. Both analytical and simulation results demonstrate the feasibility and robustness of the proposed scheme.

Based on separation of the sum of chaotic signals, this paper proposes a novel spread spectrum modulation scheme-initial condition modulation (ICM), which is suitable for high data rate communications. The success of signal separation makes it possible to transmit multiple information streams through single channel. This technique significantly improves data transmission rate and implies good information security. Our theoretical analysis shows that this approach can also cleanse the additive white Gaussian noise imposed by communication channel. Computer simulations confirm that the proposed method has a good noise performance.