5.4 Conclusion
197
Fig. 5.33 Comparison of prediction results of different networks
Table 5.8 Average
prediction relative error of
different networks
Network
Average prediction relative error/%
Wavelet
0.71
PSO
2.77
GA-BP
3.44
Table 5.9 Average time cost
of different networks
Network
Average time cost/s
wavelet
2.17
PSO
264.28
GA-BP
249.61
5.4 Conclusion
This chapter introduces optimization algorithm and physical anti-collision of RFID
systems. First, we introduce a multi-tag optimization method based on PSO neural
network. When the tags are distributed in space, their coordinates are accurately
measured through a dual CCD vision system. Then, the model is established based
on the relationship between the spatial distribution of the tags and the reading distance
by means of PSO neural network, and the simulation of maximum reading distance
is verified. The result shows that, with the help of this method, the reading distance
of the RFID system has been improved. Second, the relationship between the tags’
197
Fig. 5.33 Comparison of prediction results of different networks
Table 5.8 Average
prediction relative error of
different networks
Network
Average prediction relative error/%
Wavelet
0.71
PSO
2.77
GA-BP
3.44
Table 5.9 Average time cost
of different networks
Network
Average time cost/s
wavelet
2.17
PSO
264.28
GA-BP
249.61
5.4 Conclusion
This chapter introduces optimization algorithm and physical anti-collision of RFID
systems. First, we introduce a multi-tag optimization method based on PSO neural
network. When the tags are distributed in space, their coordinates are accurately
measured through a dual CCD vision system. Then, the model is established based
on the relationship between the spatial distribution of the tags and the reading distance
by means of PSO neural network, and the simulation of maximum reading distance
is verified. The result shows that, with the help of this method, the reading distance
of the RFID system has been improved. Second, the relationship between the tags’