Throughput and Detection Probability of Interweave Cognitive Radio Networks Using Intelligent Reflecting Surfaces

In this paper, we derive a tight lower bound of the detection probability of the energy detector when intelligent reflecting surface (IRS) are used. The secondary source uses the energy detector to detect primary source activity. There is IRS between primary source and secondary source. The secondary sources compute the energy of the received signal from primary source and reflected on IRS. The proposed spectrum sensing algorithm using IRS offers 15, 21, 27, 33 dB gain with respect to conventional sensing without IRS for a number of reflectors K = 8, 16, 32, 64. We also used IRS for data communication between primary source and destination as well as the communication between secondary nodes. The proposed primary and secondary networks of cognitive radio network (CRN) using IRS offer 23, 29, 36, 43, 49 and 56 dB gain with respect to conventional CRN without IRS for a number of reflectors K = 8, 16, 32, 64, 128, 256. We show that the use of N = 20, 10, 5 symbols in energy detection offers up to 8.5, 7.7 and 4.7 dB gain with respect to a single symbol. We plot the miss detection probability P-md versus the false alarm probability P-f. For K = 16 reflectors, average SNR per bit E-b/N-0 = -10dB and P-f = 0.01, P-md = 210-3, 710-3, 2.510-2 when N = 20, 10, 5 symbols are used in energy detection, whereas P-md = 0.45 when a single symbol is used.