Sardar, Mishra, and Khan Figure 6. Floorplan representing the position of the receiver, person, and transmitter for indoor human-target detection using the CommSense experiment. clusters with and without a human in an indoor environment from the scatter plots. The following experimental procedure was adopted for this purpose: 1. Six indoor environments were chosen. 2. LTE DL signals were captured with and without the human target for the six backgrounds. A floorplan depicting the position of the receiver, person, and transmitter is shown in Figure 6. 3. The distance of the human target from the USRP device was kept constant. The human target was placed 2.5 m from the SDR platform, which is working as an LTE receiver. Figure 7 shows the different indoor environments with the target object. 4. For all preceding cases, data were divided into two categories: with and without target. 5. The channel estimate for all cases were evaluated. 6. PCA was performed on the channel estimates to reduce its dimension. 7. Support vector machine (SVM)-based classification of the reduced dataset was performed. Figure 8 shows the PCA results of the received data captured using USRP N200 working as an LTE receiver. The dimension of the data was reduced using PCA. To select how many prin- Figure 7. Different indoor environments used for the invariant human-target detection experiment. LTE DL data recordings are performed in these five backgrounds with and without the human target. JULY 2018 IEEE A&E SYSTEMS MAGAZINE 53