The application of machine learning (ML) techniques to electronic health records (EHR) is gaining more and more attention as a method to extract valuable information that has the potential to enhance the decision-making process within the healthcare domain. A useful approach comes from the fed-erated learning (FL) scenario, which facilitates the decentralised training of machine learning models using datasets that are stored locally, hence eliminating the necessity of data aggregation on a central server. Federated learning also ensures data privacy because the federated devices do not share the actual data and store it locally. It becomes a useful tool when integrated with blockchain technology, which provides some properties such as immutability and traceability that are useful to enhance the security of such applications. With the growing use of IoT health care (loHT) devices, it is becoming challenging to manage them centrally and ensuring the health care data privacy. In this work, we propose an architecture to address the scalability issue related to the healthcare data management for federated learning networks with a sharding-based blockchain technique. We discuss some basic properties and report some results also coming from the implementation in Hyperledger Fabric.

Android has become ubiquitous by taking more than 86 % of the market share. This rapid expansion has brought a slew of new threats, including privacy breaches, data thefts, and cyberespionage. As today's malware are highly sophisticated and capable to evade even state-of-the-art malware detection tools using obfuscation and repacking techniques. However, existing surveys possess challenges for researchers as they classify tools based on analysis methodology such as static and dynamic analysis. They do not cover accuracy rate, false-positive rate, effectiveness, performance evaluation, and additional features such as secure backup and spam blocking. Therefore, to check the strength of existing anti-malware tools, we present an in-depth study of Android anti-malware tools built over the last decade. This research classifies anti-malware tools, according to their analysis methodology along with their protection capabilities, performance, accuracy rate, usability, and ability to classify malware families. Based on our thorough studies, shortcomings are discussed and research gaps have been identified.

In the past few years, Android security is enhanced \nand state-of-the-art anti-malware tools have been introduced \nto counter Android malware. These tools use both static and \ndynamic analysis techniques to detect malicious applications. \nDespite these, the attack surface against Android phones has \nrisen exponentially and malware detection tools are failed to \ncounter sophisticated threats. Therefore, it is a need to audit \nand evaluate Anti Malware Solutions (AMTs). In our research, \nwe have analyzed various Android malware evasion techniques, \nalong with their pros and cons. Moreover, we conducted a detailed \ncomparison of existing anti-malware tools and measured their \nefficacy against the discussed evasion techniques. Finally, a more \nsophisticated anti-malware evasion technique is proposed that \nuses exhaustive obfuscation and remote code execution to audit \nstatic and dynamic detection capabilities of AMTs. The proposed \ntechnique is practically validated and results prove that it evades \nall known anti-malware solutions. This technique can be utilized \nby anti-malware solution providers for making their products \nmore resilient and powerful.

In this paper, we present a bandwidth efficient retransmission method employing selective retransmission approach at a modulation layer under orthogonal frequency division multiplexing (OFDM) signaling. The proposed cross-layer design embeds a selective retransmission sub-layer in physical layer (PHY) that targets the retransmission of information symbols transmitted over poor quality OFDM sub-carriers. Most of the times, a few errors in decoded bit stream result in packet failure at medium access control (MAC) layer. The unnecessary retransmission of good quality information symbols of a failed packet has detrimental effect on the overall throughput of transceiver. We propose a cross-layer Chase combining with selective retransmission (CCSR) method by blending Chase combining approach at MAC layer and selective retransmission in PHY. The selective retransmission in PHY targets the poor quality information symbols prior to decoding, which results in lower hybrid automatic repeat reQuest retransmissions at MAC layer. We also present bit-error rate upper bound and throughput lower bound for the CCSR method. In order to maximize the throughput, we formulate optimization problem with respect to the amount of information to be retransmitted in selective retransmission. We also present an impact of selective retransmission on latency. The proposed CCSR method achieves a significant throughput gain as compared with the conventional Chase combining method.

In this study, the authors present a novel method to improve the performance of automatic repeat request (ARQ) protocol for multiple‐input–multiple‐output systems that use orthogonal frequency division multiplexing signalling by incorporating partial ARQ (PARQ) mechanism. As a typical ARQ frame has few number of errors, it is useful to retransmit partial sequence of symbols and recover from errors by consuming relatively low bandwidth. The observations from the first transmission and subsequent retransmission are used for joint detection. As a direct consequence of partial retransmission, the bandwidth and power efficiency of the overall system is improved considerably without substantial decrease in the performance. They achieve further bandwidth saving by omitting pilot sequence from the partial retransmission and acquire channel estimate in semi‐blind fashion. Partial retransmission of the original information symbol sequence allows the authors to formulate second‐order statistics based joint blind channel estimation problem by combining observations from two transmissions. Channel estimation ambiguity inherent to blind methods is resolved by combining cost functions of blind detection and least‐square estimator. To evaluate efficacy of the proposed PARQ method, they compare throughput of chase‐type hybrid‐ARQ and the proposed PARQ methods.

We propose a new joint receiver for the blind detection of synchronous co-channel multiuser signals modulated with orthogonal space-time block codes (OSTBC). We exploit the algebraic structure of the orthogonal space-time block codes and the constellation feature of Quadrature Amplitude Modulation (QAM) signals to develop a blind zero-forcing (ZF) detector based on quadratic programming (QP) with desirable convergence properties. Focusing on high rate space-time block codes in synchronous co-channel multiuser systems, we propose a successive processing joint receiver based on the proposed blind detector with low computational complexity and fast convergence. For signal detection in single user MIMO systems, the proposed QP algorithm has performance similar to that of maximum-likelihood receiver with wireless channel information. Unlike many existing techniques, our new receiver is applicable to full-rate orthogonal space-time block codes such as the practically popular Alamouti code. Furthermore, our QP receiver for multiuser detection relaxes stringent conditions on the number of receive antennas of the multi-input-multi-output (MIMO) transceiver systems.

We present a new approach to blind equalization for generalized orthogonal space-time block codes. Our method takes the form of linear programming (LP) and is globally convergent. We exploit the implicit structure of orthogonal space-time block codes to cast the problem as linear programming that can be solved efficiently. Unlike several known methods, the proposed technique is applicable to many full-rate orthogonal space time codes such as the popular Almouti code. Our algorithm allows receiver detection of full diversity codes without channel knowledge with detection performance comparable to the optimum maximum-likelihood (ML) detection.

In this work, we develop a joint ARQ receiver design for channel estimation and equalization under bandwidth efficient automatic repeat request (ARQ) in multiple-input multiple- output (MIMO) system. Our retransmission proposal improves bandwidth efficiency by sending a partial response of the original data sequence and by potentially omitting training sequence. To estimate unknown channels and to detect the transmitted data, our design of joint ARQ receivers can implement semiblind channel estimation and joint detection based on multiple transmissions. We demonstrate bandwidth savings without significant performance loss. We also propose a low complexity joint ARQ equalizer that minimizes the mean square error and provides a good design trade-off between performance and complexity.