In the field of radiocommunication, modulation type identification is one of the most important characteristics in signal processing. This study aims to implement a modulation recognition system on two approaches to m...In the field of radiocommunication, modulation type identification is one of the most important characteristics in signal processing. This study aims to implement a modulation recognition system on two approaches to machine learning techniques, the K-Nearest Neighbors (KNN) and Artificial Neural Networks (ANN). From a statistical and spectral analysis of signals, nine key differentiation features are extracted and used as input vectors for each trained model. The feature extraction is performed by using the Hilbert transform, the forward and inverse Fourier transforms. The experiments with the AMC Master dataset classify ten (10) types of analog and digital modulations. AM_DSB_FC, AM_DSB_SC, AM_USB, AM_LSB, FM, MPSK, 2PSK, MASK, 2ASK, MQAM are put forward in this article. For the simulation of the chosen model, signals are polluted by the Additive White Gaussian Noise (AWGN). The simulation results show that the best identification rate is the MLP neuronal method with 90.5% of accuracy after 10 dB signal-to-noise ratio value, with a shift of more than 15% from the k-nearest neighbors’ algorithm.展开更多
Based on Immune Programming(IP), a novel Radial Basis Function (RBF) networkdesigning method is proposed. Through extracting the preliminary knowledge about the widthof the basis function as the vaccine to form the im...Based on Immune Programming(IP), a novel Radial Basis Function (RBF) networkdesigning method is proposed. Through extracting the preliminary knowledge about the widthof the basis function as the vaccine to form the immune operator, the algorithm reduces thesearching space of canonical algorithm and improves the convergence speed. The application ofthe RBF network trained with the algorithm in the modulation-style recognition of radar signalsdemonstrates that the network has a fast convergence speed with good performances.展开更多
In this paper, the design and verification process of an automobile-engine-fan control system on chip （SoC） are introduced. The SoC system, SHU-MV08, reuses four new intellectual property （IP） cores and the design...In this paper, the design and verification process of an automobile-engine-fan control system on chip （SoC） are introduced. The SoC system, SHU-MV08, reuses four new intellectual property （IP） cores and the design flow is accomplished with 0.35 btm chartered CMOS technology. Some special functions of IP cores, the detailed integration scheme of four IP cores, and the verification method of the entire SoC are presented. To settle the verification problems brought by analog IP cores, NanoSim based chip-level mixed-signal verification method is introduced. The verification time is greatly reduced and the first tape-out achieves success which proves the validity of our design.展开更多
A static approach method method is used to get the spectrum expression of the pulse interval modulation signal.The expression is simple,has a clear physical concept and it has been confirmed by a simultaneous experiment.
A novel radio-over-fiber(ROF) scheme to simultaneously generate and transmit the 3rd generation telecommunication(3G) and millimeter-wave(MMW) signals by using a single dual-electrode Mach-Zehnder modulator(MZM) is pr...A novel radio-over-fiber(ROF) scheme to simultaneously generate and transmit the 3rd generation telecommunication(3G) and millimeter-wave(MMW) signals by using a single dual-electrode Mach-Zehnder modulator(MZM) is proposed.There is no apparent nonlinearity induced by the ROF system.By employing this analog ROF signal transmission technique,highly transparent fiber-wireless convergence networks can be realized,which are ideal for multi-standard wireless system operation.展开更多
文摘In the field of radiocommunication, modulation type identification is one of the most important characteristics in signal processing. This study aims to implement a modulation recognition system on two approaches to machine learning techniques, the K-Nearest Neighbors (KNN) and Artificial Neural Networks (ANN). From a statistical and spectral analysis of signals, nine key differentiation features are extracted and used as input vectors for each trained model. The feature extraction is performed by using the Hilbert transform, the forward and inverse Fourier transforms. The experiments with the AMC Master dataset classify ten (10) types of analog and digital modulations. AM_DSB_FC, AM_DSB_SC, AM_USB, AM_LSB, FM, MPSK, 2PSK, MASK, 2ASK, MQAM are put forward in this article. For the simulation of the chosen model, signals are polluted by the Additive White Gaussian Noise (AWGN). The simulation results show that the best identification rate is the MLP neuronal method with 90.5% of accuracy after 10 dB signal-to-noise ratio value, with a shift of more than 15% from the k-nearest neighbors’ algorithm.
文摘Based on Immune Programming(IP), a novel Radial Basis Function (RBF) networkdesigning method is proposed. Through extracting the preliminary knowledge about the widthof the basis function as the vaccine to form the immune operator, the algorithm reduces thesearching space of canonical algorithm and improves the convergence speed. The application ofthe RBF network trained with the algorithm in the modulation-style recognition of radar signalsdemonstrates that the network has a fast convergence speed with good performances.
基金Project supported by the IC Special Foundation of Shanghai Municipal Commission of Science and Technology (Grant No.09706201300)the Shanghai Municipal Commission of Economic and Information (Grant No.090344)the Shanghai High-Tech Industrialization of New Energy Vehicles (Grant No.09625029),and the Graduate Innovation Foundation of Shanghai University
文摘In this paper, the design and verification process of an automobile-engine-fan control system on chip （SoC） are introduced. The SoC system, SHU-MV08, reuses four new intellectual property （IP） cores and the design flow is accomplished with 0.35 btm chartered CMOS technology. Some special functions of IP cores, the detailed integration scheme of four IP cores, and the verification method of the entire SoC are presented. To settle the verification problems brought by analog IP cores, NanoSim based chip-level mixed-signal verification method is introduced. The verification time is greatly reduced and the first tape-out achieves success which proves the validity of our design.
文摘A static approach method method is used to get the spectrum expression of the pulse interval modulation signal.The expression is simple,has a clear physical concept and it has been confirmed by a simultaneous experiment.
基金supported by the National Natural Science Foundation of China(No.61371061)the Zhejiang Provincial Natural Science Foundation of China(Nos.LY12F01010 and LQ13F010007)the Open Fund of the Information and Communication Engineering of Top Key Discipline of Zhejiang Province
文摘A novel radio-over-fiber(ROF) scheme to simultaneously generate and transmit the 3rd generation telecommunication(3G) and millimeter-wave(MMW) signals by using a single dual-electrode Mach-Zehnder modulator(MZM) is proposed.There is no apparent nonlinearity induced by the ROF system.By employing this analog ROF signal transmission technique,highly transparent fiber-wireless convergence networks can be realized,which are ideal for multi-standard wireless system operation.
