24 May 2017 in IEEE Transactions on Biomedical Engineering
Objective:Diabetic Retinopathy (DR) is characterized by the progressive deterioration of retina with the appearance of different types of lesions that include microaneurysms, hemorrhages, exudates etc. Detection of these lesions plays significant role for early diagnosis of DR. Methods: To this aim, this paper proposes a novel and automated lesion detection scheme which consists of the four main steps: vessel extraction and optic disc removal, pre-processing, candidate lesion detection and post-processing. The optic disc and the blood vessels are suppressed first to facilitate further processing. Curvelet based edge enhancement is done to separate out the dark lesions from the poorly illuminated retinal background while the contrast between the bright lesions and the background is enhanced through an optimally designed wideband bandpass filter. The mutual information of the maximum matched filter response and the maximum Laplacian of Gaussian response are then jointly maximized. Differential Evolution algorithm is used to determine the optimal values for the parameters of the fuzzy functions that determine the thresholds of segmenting the candidate regions. Morphology based post-processing is finally applied to exclude the falsely detected candidate pixels. Results and Conclusions:Extensive simulations on different publicly available databases highlight an improved performance over the existing methods with an average accuracy of 97.71% and robustness in detecting the various types of DR lesions irrespective of their intrinsic properties.
26 February 2024 in IEEE Transactions on Consumer Electronics
Emergence of 5G and beyond promise development of several applications specific Internet-of-Things (IoT) services involving consumer electronics devices doing trustworthy intelligent operations. One such application is smart healthcare support in hospital or home premises where battery driven wearable wireless nodes collect patient data, transmit securely and seamlessly in cooperative communications for monitoring. To meet the goal, this work suggests device-to-device (D2D) communications, operated in cognitive radio network (CRN), protecting from eavesdropping by exploiting artificial intelligence driven channel state information (CSI) estimation. IoT devices (IoDs) harvest energy from radio frequency (RF) signals and transmit own data with relaying message of primary users (PUs). The goal is to maximize energy efficiency (EE) of IoDs satisfying the constraints of own data transmission rate, cooperative outage of PUs, and secrecy outage rate with self-powering. A long short term memory (LSTM) based CSI estimation on indoor complex D2D links is suggested and shows comparable performance on EE maximization and outage secrecy, when compared with known CSI. Simulation results show about 20% EE performance improvement at 7 dB signal-to-noise-ratio (SNR) over 8 dB SNR at the power splitting factor 0.5 and time switching factor 0.07 using LSTM based CSI.
21 December 2017 in IEEE Transactions on Cognitive Communications and Networking
This paper explores joint power allocation and route selection in a multi-hop cognitive radio network consisting of secondary transmitter and receiver connected through decode-and-forward relays. A novel frame structure of radio frequency energy harvesting (EH)-cooperation-transmission is considered that operates in time switching mode. The relays also help (cooperation phase) in data transmission between the primary transmitter and receiver using a fraction of the harvested energy. An optimization problem is formulated to minimize the end-to-end secondary outage probability under the constraints of energy causality and primary user cooperation rate. Closed form expressions of the optimal time duration for EH and power allocation factor on each relay are also derived. In order to minimize the total power consumption (cost) and to enhance the network lifetime, the optimal route selection is also explored using Bellman-Ford algorithm and the efficacy of the same over Dijkstra’s algorithm is also demonstrated by a large set of simulation results. Simulation results also show that a gain in outage probability ∼43.57% and ∼56.66% is achieved for the proposed approach over the existing works.
29 April 2022 in IEEE Transactions on Cognitive Communications and Networking
This paper reports a relative performance analysis of decode-and-forward (DF) and amplify-and-forward (AF) relaying in a multi-antenna cooperative cognitive radio network (CCRN) that supports device-to-device (D2D) communications using spectrum sharing technique in cellular network. The system model considers cellular system as primary users (PUs) while internet of things devices (IoDs), involved in D2D communications, as secondary system. The devices access the licensed spectrum by means of the cooperation in two-way primary communications. Furthermore, IoDs are energized through energy harvesting (EH) of PU radio frequency (RF) signals, using simultaneous wireless information and power transfer (SWIPT) protocol. Closed form expressions of the outage probability for both cellular and D2D communications are derived and the impact of various design parameters for both AF and DF relaying techniques are studied. Based on the simulation results, it is found that the proposed spectrum sharing protocol outperforms by 209% and 49% for DF relaying and AF relaying schemes, respectively. It is also observed that DF relaying performs better in term of peak energy efficiency (EE) at low transmit power while AF relaying at high transmit power.
1 March 2016 in Computers in Biology and Medicine
6 March 2019 in IEEE Transactions on Cognitive Communications and Networking
This paper addresses Energy Efficiency (EE)-Spectrum Efficiency (SE) trade-off in an energy harvesting (EH) cooperative cognitive radio network (CCRN) operated in a frame structure comprising of spectrum sensing (SS) and cooperation-transmission mode. The secondary users (SUs) are clustered into two groups and function in a way that when the SUs in Group I participate in cooperative SS, the SU nodes in Group II harvest energy from the radio frequency (RF) signal of primary user (PU). If PU is detected to be in transmit mode, the SU nodes in Group II cooperate in PU’s transmission while the SUs in Group I perform EH. All the SU nodes transmit data when PU is detected idle. The objective is to maximize EE and SE via joint optimization of SS time, final decision threshold, self and cooperative transmission power gains. Mathematical analysis is done to solve the problem of EE-SE trade-off under the constraints of SS reliability, SU power budget, PU cooperative SE and fairness in SE for each SU. Numerical results highlight about 19% and 14% gain in normalized EE and SE values, respectively compared to the existing work and the issue of EE-SE trade-off is clearly investigated.