Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
3 results
Search Results
Item Redesigned Spatial Modulation for Spatially Correlated Fading Channels(Springer New York LLC barbara.b.bertram@gsk.com, 2017) G.D., G.S.; Koila, K.; Neha, N.; Raghavendra, R.; Sripati, U.In this paper, a new variant of Spatial Modulation (SM) Multiple-Input Multiple-Output (MIMO) transmission technique, designated as Redesigned Spatial Modulation (ReSM) has been proposed. In ReSM scheme, a dynamic mapping for antenna selection is adopted. This scheme employs both single antenna as well as double antenna combinations depending upon channel conditions to combat the effect of spatial correlation. When evaluated over spatially correlated channel conditions, for a fixed spectral efficiency and number of transmit antennas, ReSM exhibits performance improvement of at least 3 dB over all the conventional SM schemes including Trellis Coded Spatial Modulation (TCSM) scheme. Furthermore, a closed form expression for the upper bound on Pairwise Error Probability (PEP) for ReSM has been derived. This has been used to calculate the upper bound for the Average Bit Error Probability (ABEP) for spatially correlated channels. The results of Monte Carlo simulations are in good agreement with the predictions made by analytical results. The relative gains of all the comparison plots in the paper are specified at an ABER of 10?4. © 2017, Springer Science+Business Media, LLC.Item Signal constellations employing multiplicative groups of Gaussian and Eisenstein integers for Enhanced Spatial Modulation(Elsevier B.V., 2017) G.D., G.S.; Raghavendra, R.; Koila, K.; Shripathi Acharya, U.In this paper, we propose two new signal constellation designs employing Gaussian and Eisenstein Integers for Enhanced Spatial Modulation (ESM). ESM is a novel technique which was propounded by Cheng et al. The advantage of ESM over other Spatial Modulation (SM) schemes lies in its ability to enhance spectral efficiency while keeping the energy efficiency intact. This is done by activating either one or two antennas judiciously depending upon the required trade-off. In ESM, information radiated from the antennas depends upon index of the active transmit antenna combination(s) and also on the set of constellation points chosen, which may include points from multiple constellations. In this paper, we propose signal constellations based on multiplicative groups of Gaussian and Eisenstein integers. The set comprising of Gaussian and Eisenstein integers serves as primary and secondary constellation points for Gaussian Enhanced Spatial Modulation (GESM) scheme. The secondary constellation points are deduced from a single geometric interpolation from the primary constellation points. The Monte Carlo simulation results indicate that the proposed nonuniform constellations achieve impressive SNR gains compared to conventional constellation points used in the design of ESM. This new design has been described for MIMO employing 4 × 4 and 8 × 8 antenna configurations with only two active antennas. Furthermore, a closed form expression for the pairwise error probability (PEP) for the GESM scheme has been deduced. The PEP is utilized to determine the upper bound on the average bit error probability (ABEP). Our simulations indicate that the proposed GESM from Gaussian and Eisenstein integers scheme outperforms all the other variants of SM including conventional ESM by at least 2.5 dB at an average bit error ratio (ABER) of 10?5. Close correspondence between the theoretical analysis and the Monte Carlo simulation results are observed. © 2017 Elsevier B.V.Item A new deep learning architecture for dehazing of aerial remote sensing images(Springer, 2022) Kalra, A.; Sequeira, A.; Manjunath, A.; Lal, S.; Raghavendra, R.A major problem in most aerial remote image processing applications is the presence of haze in images. It is a phenomenon by which particles in the atmosphere disperse light, thus altering the quality of the overall image. This can be detrimental to the performance of vision-based algorithms such as those concerned with object detection. There have been numerous attempts using traditional image processing techniques as well as using deep learning approaches to eliminate this haze. In most cases, models tend to make assumptions on the nature of haze that are rarely true in reality. In this paper, we propose an end-to-end deep learning architecture that can dehaze aerial remote sensing images efficiently with minimal deviation from the ground truth. Many of the assumptions made in other models are eliminated and the relationship between hazed and dehazed images is directly computed. The proposed model is based on the observation that identifying structural and statistical portions separately from an image and using those features to reconstruct the image can give a realistic dehazed image. It also makes use of information exposed by different color spaces to achieve this using lesser computation. The experimental quantitative and qualitative results of the proposed architecture are compared with recent benchmark dehaze models on NYU hazy dataset and real-world hazy images. Experimental results yield that the proposed architecture outperforms benchmark models on test aerial remote sensing images. © 2022, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.