RESUMO
In wake of covid19, many countries are shifting their paper-based health record management from manual processes to digital ones. The major benefit of digital health record is that data can be easily shared. As health data is sensitive, more security is to be provided to gain the trust of stakeholders. In this paper, a novel secure authentication protocol is planned for digitalizing personal health record that will be used by the user. While transacting data, a key is used to secure it. Many protocols used elliptic curve cryptography. In this proposed protocol, at an initial stage, an asymmetric and quantum-resistant crypto-algorithm, Kyber is used. In further stages, symmetric crypto-algorithm, Advanced Encryption Standard in Galois/Counter mode (AES-GCM) is used to secure transferred data. For every session, a new key is generated for secure transactions. The more interesting fact in this protocol is that transactions are secured without exchanging actual key and also minimized the key exchange. This protocol not only verified the authenticity of user but also checked rightful citizenship of user. This protocol is analyzed for various security traits using ProVerif tool and provided better results relating to security provisioning, cost of storage, and computation as opposed to related protocols.
RESUMO
A cognitive radio network (CRN) is integrated with the Internet of Connected Vehicles (IoCV) in order to address spectrum scarcity and communication reliability issues. However, it is limited, possessing less throughput, a low packet delivery ratio, high latency, and high mobility in the spectrum. In this research study, the existing issues are addressed by proposing a 6G cognitive radio network-Internet of connected vehicles (6GCRN-IoCV) approach. Initially, all the entities such as secondary users (SUs), primary users (PUs), and pedestrians are authenticated in blockchain to ensure security. The edge-assisted roadside units (ERSU) initiate clustering only for authenticated SUs using the improved DBSCAN algorithm in consideration of several metrics. The ERSU then generates an intersection-aware map using the spatial and temporal-based logistic regression algorithm (STLR) to reduce collisions in the intersection. The spectrum utilization is improved by performing spectrum sensing in which all the SUs involved in spectrum sensing use lightweight convolutional neural networks (Lite-CNN) in consideration of several metrics and provide the sensing report to the fusion center (FC) in an encrypted manner to reduce the spectrum scarcity and security issues. The communications between the SUs are necessary to avoid risks in the IoCV environment. Hence, optimal routing is performed using the Dingo Optimization Algorithm (DOA), which increases throughput and packet delivery ratio. Finally, communication reliability is enhanced by performing hybrid beamforming, and this exploits the multi-agent-based categorical Deep-Q Network (categorical DQN), which increases spectral efficiency based on its adaptive intelligent behavior. The proposed study is simulated using the SUMO and OMNeT++ simulation tools and the performances are validated with existing works using several performance metrics. The result of the simulation shows that the proposed work performs better than the existing approaches.
RESUMO
Wireless body area networks (WBANs) are a research area that supports patients with healthcare monitoring. In WBAN, the Internet of Things (IoT) is connected with WBAN for a smart/remote healthcare monitoring system in which various medical diseases are diagnosed. Quality of service (QoS), security and energy efficiency achievements are the major issues in the WBAN-IoT environment. Existing schemes for these three issues fail to achieve them since nodes are resource constrained and hence delay and the energy consumption is minimized. In this paper, a blockchain-assisted delay and energy aware healthcare monitoring (B-DEAH) system is presented in the WBAN-IoT environment. Both body sensors and environment sensors are deployed with dual sinks for emergency and periodical packet transmission. Various processes are involved in this paper, and each process is described as follows: Key registration for patients using an extended version of the PRESENT algorithm is proposed. Cluster formation and cluster head selection are implemented using spotted hyena optimizer. Then, cluster-based routing is established using the MOORA algorithm. For data transmission, the patient block agent (PBA) is deployed and authenticated using the four Q curve asymmetric algorithm. In PBA, three entities are used: classifier and queue manager, channel selector and security manager. Each entity is run by a special function, as packets are classified using two stream deep reinforcement learning (TS-DRL) into three classes: emergency, non-emergency and faulty data. Individual packets are put into a separate queue, which is called emergency, periodical and faulty. Each queue is handled using Reyni entropy. Periodical packets are forwarded by a separate channel without any interference using a multi objective based channel selection algorithm. Then, all packets are encrypted and forwarded to the sink nodes. Simulation is conducted using the OMNeT++ network simulator, in which diverse parameters are evaluated and compared with several existing works in terms of network throughput for periodic (41.75 Kbps) and emergency packets (42.5 Kbps); end-to-end delay for periodic (0.036 s) and emergency packets (0.028 s); packet loss rate (1.1%); residual energy in terms of simulation rounds based on periodic (0.039 J) and emergency packets (0.044 J) and in terms of simulation time based on periodic (8.35 J) and emergency packets (8.53 J); success rate for periodic (87.83%) and emergency packets (87.5%); authentication time (3.25 s); and reliability (87.83%).
