Media-based modulation (MBM) is a recently proposed modulation scheme which uses radio frequency (RF) mirrors at the transmit antenna(s) in order to create different channel fade realizations based on their ON/OFF status. These complex fade realizations constitute the modulation alphabet. MBM has the advantage of increased spectral efficiency and performance. In this paper, we investigate the performance of some physical layer techniques when applied to MBM. Particularly, we study the performance of i) MBM with generalized spatial modulation (GSM), ii) MBM with mirror activation pattern (MAP) selection based on an Euclidean distance (ED) based metric, and iii) MBM with feedback based phase compensation and constellation rotation. Our results show that, for the same spectral efficiency, GSM-MBM can achieve better performance compared to MIMO-MBM. Also, it is found that MBM with EDbased MAP selection results in improved bit error performance, and that phase compensation and MBM constellation rotation increases the ED between the MBM constellation points and improves the performance significantly. We also analyze the diversity orders achieved by the ED-based MAP selection scheme and the phase compensation and constellation rotation (PC-CR) scheme. The diversity orders predicted by the analysis are validated through simulations.

No abstract

No abstract

In this paper, we propose an automatic approach to skin lesion region segmentation based on a deep learning architecture with multi-scale residual connections. The architecture of the proposed model is based on UNet [22] with residual connections to maximise the learning capability and performance of the network. The information lost in the encoder stages due to the max-pooling layer at each level is preserved through the multi-scale residual connections. To corroborate the efficacy of the proposed model, extensive experiments are conducted on the ISIC 2017 challenge dataset without using any external dermatologic image set. An extensive comparative analysis is presented with contemporary methodologies to highlight the promising performance of the proposed methodology.

No abstract