Cooperative Multiband Spectrum Sensing Using Radio Environment Maps and Neural Networks

. 2023 May 30;23(11):5209. doi: 10.3390/s23115209

Algorithm 2 Central Entity Computed by NNs
Step 2.1	$The central entity collects the data from each of the SUs in the database at a$ $specific time T_{x}$ $. In the module input estimator and processor, a processing is$ $carried out in order to estimate the inputs driving a NNs ensemble . As$ $mentioned above, the three vectors coming from each SU for time T_{x}$ $do not$ $have the same length as the vectors for time T_{x + t_{1}}$ $or T_{x - t_{2}}$ $. Even, the size of the$ $vectors differs from one SU to another SU even though they are at the same$ $time T_{x}$ $. Based on this fact, this module oversees building the input vectors$ $P S D_r e c_{T_{t}} \{R_{i}^{'} (k)\}$ $and E d g e_r e c_{T_{t}} \{R_{i}^{'} (k)\}$ driving the NNs. For this, these vectors must have always the same length (in our case this length is set to 13 samples). This block is detailed below.
Step 2.2	$The vector P S D_r e c_{T_{t}} \{R_{i}^{'} (k)\}$ $is evaluated by N N_{1}$ $, at the same time the vector$ $E d g e_r e c_{T_{t}} \{R_{i}^{'} (k)\}$ $is evaluated by N N_{2}$ $and by N N_{3}$ $. This evaluation is$ $performed sequentially, i . e ., each vector of each SU will be evaluated by its$ $corresponding NN one after the other until the result of the i - th connected SU$ $be obtained . As a result of this step, N N_{1}$ $provides an approximate number of$ $detected PUs and their powers . N N_{2}$ $gives an approximate number of detected$ $PUs and their central frequencies . Finally, N N_{3}$ returns an approximate number of detected PUs and their bandwidths. At the end of this section, it is discussed what would happen if the number of PUs detected by each NN is not the same.
Step 2.3	The information obtained from Step 2.2 is evaluated to determine the number of PUs in the spectrum and their corresponding power, bandwidth, and carrier frequency. This evaluation is detailed below.
Step 2.4	$The information obtained from Step 2.3 will be shared with the REM estimator$ $module . This block receives (i) the geographic coordinates of the SUs in the$ $network, (ii) the data of the possible PUs in the spectrum (power, carrier, and$ $bandwidth) and (iii) an identifier that corresponds to time T_{x}$ in which the spectrum was monitored.