"Research on Photo-electric ISAC to Predict Nitrogen Content in lettuce"

"The existing communication network has covered the main agricultural production areas, but the low utilization of hardware resources because the independence of communication and perception modules, the network layer has a single function. In this paper, an integrated sensing and communication (ISAC) system for visible light communication (VLC) for lettuce nitrogen(N) content perception is designed, in the transmitter uses an arbitrary waveform generator (AWG) to generate binary frequency shift keying (BFSK), and transmits an optical modulation signal carrying the encoded information through a DC controller (DC) combined with a spectral sensing LED. The photodetector (PDA) with high sensitivity response is employed photoelectric conversion, and the oscilloscope (OSC) demodulates the signal and constructs a 1D CNN N regression model based on the photo-electrical reflection characteristics of the lettuce content at the receiver. The studies showed that the signal-to-noise ratio (SNR) was greater than 39.5dB and the bit error rate (BER) was less than 10-6 at 1kilobit per second kbps) with a single LED configuration. Therefore, the visible light signal transmission quality was stability and anti-interference ability. The N content in the lettuce canopy had correlation with the multidimensional characteristics of the wireless signal, and the principal component analysis (PCA) one-dimensional convolutional neural network(1D-CNN) achieved a coefficient of determination (R2)=0.9847, root mean square error (RMSE)=3.08%, and mean absolute error (MAE)=1.98%. therefore, the system has the functions of lighting, communication and perception. It will further empower the communication system to expand from transmitting information to perception, and advance comprehensive agriculture development."