Faculty of Computing, Universiti Teknologi Malaysia, 81310 Skudai, Johor, Malaysia.
Sensors (Basel). 2013 Dec 24;14(1):299-345. doi: 10.3390/s140100299.
For the past 20 years, many authors have focused their investigations on wireless sensor networks. Various issues related to wireless sensor networks such as energy minimization (optimization), compression schemes, self-organizing network algorithms, routing protocols, quality of service management, security, energy harvesting, etc., have been extensively explored. The three most important issues among these are energy efficiency, quality of service and security management. To get the best possible results in one or more of these issues in wireless sensor networks optimization is necessary. Furthermore, in number of applications (e.g., body area sensor networks, vehicular ad hoc networks) these issues might conflict and require a trade-off amongst them. Due to the high energy consumption and data processing requirements, the use of classical algorithms has historically been disregarded. In this context contemporary researchers started using bio-mimetic strategy-based optimization techniques in the field of wireless sensor networks. These techniques are diverse and involve many different optimization algorithms. As far as we know, most existing works tend to focus only on optimization of one specific issue of the three mentioned above. It is high time that these individual efforts are put into perspective and a more holistic view is taken. In this paper we take a step in that direction by presenting a survey of the literature in the area of wireless sensor network optimization concentrating especially on the three most widely used bio-mimetic algorithms, namely, particle swarm optimization, ant colony optimization and genetic algorithm. In addition, to stimulate new research and development interests in this field, open research issues, challenges and future research directions are highlighted.
在过去的 20 年中，许多作者都专注于研究无线传感器网络。已经广泛探索了与无线传感器网络相关的各种问题，例如能量最小化（优化）、压缩方案、自组织网络算法、路由协议、服务质量管理、安全性、能量收集等。这些问题中最重要的三个问题是能效、服务质量和安全管理。为了在无线传感器网络优化中尽可能在一个或多个这些问题上取得最佳效果，优化是必要的。此外，在许多应用中（例如，身体区域传感器网络、车载自组织网络），这些问题可能会发生冲突，需要在它们之间进行权衡。由于高能耗和数据处理要求，经典算法的使用在历史上一直被忽视。在这种情况下，当代研究人员开始在无线传感器网络领域使用基于仿生策略的优化技术。这些技术多种多样，涉及许多不同的优化算法。据我们所知，大多数现有工作往往只专注于优化上述三个问题中的一个特定问题。现在是将这些单独的努力放在一起，并采取更全面的视角的时候了。在本文中，我们通过对无线传感器网络优化领域的文献进行调查，朝着这个方向迈出了一步，特别是集中研究了三种应用最广泛的仿生算法，即粒子群优化、蚁群优化和遗传算法。此外，为了激发该领域的新研究和开发兴趣，突出了开放的研究问题、挑战和未来的研究方向。
