In the era of the Internet of Things (IoT), blockchain is a promising technology for improving the efficiency of healthcare systems, as it enables secure storage, management, and sharing of real-time health data collected by the IoT devices. As the implementations of blockchain-based healthcare systems usually involve multiple conflicting metrics, it is essential to balance them according to the requirements of specific scenarios. In this paper, we formulate a joint optimization model with three metrics, namely latency, security, and computational cost, that are particularly important for IoT-enabled healthcare. However, it is computationally intractable to identify the exact optimal solution of this problem for practical sized systems. Thus, we propose an algorithm called the Adaptive Discrete Particle Swarm Algorithm (ADPSA) to obtain near-optimal solutions in a low-complexity manner. With its roots in the classical Particle Swarm Optimization (PSO) algorithm, our proposed ADPSA can effectively manage the numerous binary and integer variables in the formulation. We demonstrate by extensive numerical experiments that the ADPSA consistently outperforms existing benchmark approaches, including the original PSO, exhaustive search and Simulated Annealing, in a wide range of scenarios.

翻译：在物联网（IoT）时代，区块链作为一种有前景的技术，能够提升医疗系统的效率，因为它实现了对物联网设备采集的实时健康数据进行安全存储、管理和共享。由于基于区块链的医疗系统实施通常涉及多个相互冲突的指标，因此需要根据具体场景的要求在这些指标之间取得平衡。本文构建了一个包含延迟、安全性和计算成本这三个对物联网医疗尤为重要的指标的联合优化模型。然而，对于实际规模的系统，精确求解该问题的最优解在计算上是不可行的。因此，我们提出了一种名为自适应离散粒子群算法（ADPSA）的算法，以低复杂度方式获得近优解。该算法源于经典粒子群优化（PSO）算法，能够有效处理模型中大量的二元和整数变量。通过大量数值实验表明，在多种场景下，ADPSA算法在性能上始终优于现有基准方法，包括原始PSO算法、穷举搜索法和模拟退火法。