Author Affiliations
School of Computer and Communication, Lanzhou University of Technology, Lanzhou 730050, Gansu , Chinashow less
Fig. 1. OSPPM-FTN system model
Fig. 2. MNN decoder block diagram
Fig. 3. Comparison of theoretical upper bound and simulation performance of BER in OSPPM-FTN system
Fig. 4. BER performance of OSPPM-FTN system with different parameters
Fig. 5. Performance comparison of OSPPM and OSPPM-FTN systems. (a) BER performance of OSPPM and OSPPM-FTN systems with different ; (b) transmission rate and spectral efficiency of OSPPM and OSPPM-FTN systems with different
Fig. 6. Relationship between loss and training rounds of MNN decoder
Fig. 7. Comparison of computational complexity of different decoders
Fig. 8. BER performance of ML and MNN decoders under different turbulence
Input bits | LD index | 4-PPM signal | 4PPM-FTN signal |
---|
0000 | | | | 0001 | | | | 0010 | | | | 0011 | | | | | | | | 1110 | | | | 1111 | | | |
|
Table 1. Mapping table of OSPPM-FTN system
Parameter | / |
---|
Strong turbulence | | Middle turbulence | | Weak turbulence | |
|
Table 2. Turbulence model parameters
Modulation | Spectral efficiency /(bit · s-1 · Hz-1) | Transmission rate /(bit/channel) |
---|
OSPPM | | | OSPPM-FTN | | |
|
Table 3. Spectral efficiency and transmission rate of different schemes
Parameter | Content |
---|
Hidden layer activation | ReLU | Output layer activation | Sigmoid | Loss function | Cross entropy loss | Optimizer | SGD | Epoch | 30 | Learning rate | 0.001 | Number of training set | | Number of validation set | | Hidden nodes | 16, 24, 16 |
|
Table 4. Parameters of MNN decoder
Hidden layer number | Accuracy /% |
---|
2 | 99.930015 | 3 | 99.978038 | 4 | 99.953223 | 5 | 99.821631 |
|
Table 5. Relationship between hidden layer number and decoding accuracy
Learning rate | Accuracy /% |
---|
0.01000 | 99.640017 | 0.00100 | 99.996005 | 0.00010 | 99.987218 | 0.00001 | 99.705215 |
|
Table 6. Relationship between learning rate and decoding accuracy
Decoder | Complexity /FLOPs |
---|
ML | | MNN | |
|
Table 7. Computational complexity of different decoders