Cognitive RF-based localization for mission-critical applications in smart cities: An overview

https://doi.org/10.1016/j.compeleceng.2020.106780Get rights and content

Abstract

The accessibility of accurate location information for operators in mission-critical scenarios would considerably increase their mission success. In order to obtain precise location information, numerous algorithms and technologies have been suggested. These methods and systems show varying performances under different conditions, and with the help of machine learning techniques, their reliability can be enhanced dramatically. In this paper, we overview the state-of-the-art in emerging algorithms and technologies employing cognitive solutions in mission critical localization applications. We compare these algorithms in terms of different localization parameters such as scalability, power consumption, availability, service quality and accuracy. Consequently, this survey will assist researchers who are working in the area of RF-based localization to achieve better performance in mission critical scenarios that can be experienced in smart city applications.

Introduction

IoT is described as a network of interrelated physical devices, sensors and everyday objects capable of interacting with each other and the surrounding environment [1]. With other enabling technologies such as big data analytics, artificial intelligence, and machine learning, these devices are promising great opportunities that can improve the quality of our lives. The main goal is to enhance efficiency, reduce costs, and to engage more effectively and actively. The idea is the pervasive presence of things around people that can observe, understand, and even act. Consumers have already started experiencing in their daily lives some applications of this idea such as smart homes and smart environment. Other sectors such as manufacturing industry are rapidly adopting IoT technologies for facility management, production flow monitoring and maintenance, and creating digitally connected factories. Similarly, health industry is using smart technologies for patient monitoring (e-health), management, maintenance and security to achieve more efficient hospitals [1]. Another important application area of IoT technologies is cities. According to United Nations, urban population will increase from 55% to 68% and approximately 2.5 billion more people will live in cities by 2050, which will create huge management problems such as energy sustainability, housing, traffic congestion and security [2]. Several studies have been conducted to address these issues. However, one extremely challenging aspect of the cities that needs attention from the authorities and the researchers is crisis where human life is at risk and immediate intervention is required. A typical example is a disaster situation such as hurricane or flood where the known environment significantly changes. In these situations, obtaining accurate and uninterrupted useful information (for instance location data) for responsible authorities (rescue, police, firefighter, local-security, etc.) is a MC requirement that has a paramount effect on the success of handling the problem at hand.

In this survey paper, we focused on the location data and localization problem in smart cities when localization is essential to the success of the assisting systems. Localization, in its simplest form, is defined as the process of predicting location of people, devices and other objects of interest [3]. MC localization has different set of rules and in order to comply with them, improvements are required for applicable algorithms, measurement technologies, prediction methods and performance metrics. Some of the guidelines published by National Institute of Standards and Technology are: the systems should (i) provide positioning accuracy of approximately one meter, (ii) be able to work in any building, (iii) be resistant to structural changes, (iv) be light and easy to carry, (v) have reasonable costs, and (vi) operation does not necessarily require training in different fields [4]. In this work, we mainly focus on the first two requirements, and discuss enabling technologies, communication protocols and prediction techniques for accurate positioning in indoor environments.

Several surveys have been performed on the reliable and precise localization for different environments in literature. These studies can be classified into three categories, which are, localization in MC scenarios, localization in smart cities, and fundamentals of localization techniques/technologies. In addition, in this work, studies were evaluated by considering cognitive radios. First group of surveys were mainly for localization for MC applications. Discussed techniques have stringent requirements and have to consider different operating conditions. Techniques discussed in the second set of surveys typically aim to increase the positioning accuracy in smart cities by taking into account the existing infrastructures. In addition, last group of surveys reviewed available technologies and basic techniques, compared localization and positioning systems and summarized their working principles. In this study, we survey localization issues with cognitive radios for MC situations in smart cities. To the best of our knowledge, this study is the first survey paper that reviews localization for MC scenarios with cognitive solutions in smart cities.

In the first group of surveys, concerning position estimation in MC scenarios, Fuchs et al. discussed the requirements of indoor tracking and reviewed available techniques. They classified the state of art techniques with respect to the MC requirements, and compared in terms of reliability, accuracy and capability in MC scenarios [4]. In another survey, Ferreira identified the specific requirements of indoor positioning systems for emergency responders. They covered the localization methods and highlighted their advantages and disadvantages. They also compared the available systems, discussed the issues about their designs, requirements, additional features and performances. Moreover, they proposed a taxonomy for the main design choices to classify the existing systems for emergency responders. This taxonomy addressed the technological principles, localization principles, deployment techniques, algorithms and working environments [5]. This group of articles mainly focus on the technological details of localization and do not address the MC scenarios for smart environments.

In the second group of surveys, Pahlavan et al. presented the relation between the smart environments and localization technologies. They classified localization applications for smart devices and existing environments into four categories, namely, cyber world, social world, physical world and thinking world. Researchers described the technologies in terms of their complexity and implementation challenges. They also claimed that localization is a complex phenomenon, which demands more investigation to provide the precision needed for the applications on diverse platforms and environments [6]. Different from other studies, Khelifi et al. presented application of several approaches of the localization systems in IoT and smart cities, and compared different location-based system. Finally, they discussed open research topics for localization in smart environments. In their study, localization approaches were classified into three groups, which are, centralized, distributed, and iterative [7]. In this group of surveys, studies could also address the possible ways of combining information gathered by smart environments and other sensing technologies used by the rescue teams to address the requirements of critical situations.

In the last group of surveys, [8], [9], [10], authors discussed the advantages and disadvantages of existing localization techniques, provided reviews of the recent improvements in wireless indoor localization and discussed the future trends. Table 1 shows the abbreviations used in the paper.

Besides these groups of articles, Palmeri et al., in their pioneering study, presented a hybrid cloud architecture for MC scenarios [11]. They proposed a method using multiple sensors for pedestrian navigation in both indoor and outdoor environments. Their method combined PDR with RSS based FP. In order to increase reliability, sensor information was fused using Kalman filtering. In another study, Wan et al. proposed a high level architecture for the smart cities for emergency situations. The system is composed of a central agent and three layers (sensor network, UAV and multi-robot). In order to provide a system wide monitoring, they used UAVs at disaster locations. The design required communication between different layers to detect and handle crisis scenarios [12].

As a result, the smart city concept offers a unique opportunity for positioning in MC applications. Fusion of technologies in and around the infrastructures, portable devices, smart vehicles and others offer additional capabilities for positioning in MC applications. Studies should be carried out by evaluating these technologies together. The rest of the paper is organized as follows. The second section discusses indoor localization applications, including application scenarios, technologies, performance metrics and measurement methods. In Sections 3 and 4, different localization technologies and performance metrics are reviewed respectively. The fifth section discusses application scenarios and the last section explains open research issues.

Section snippets

Techniques and methods

Localization is the process of predicting location of an object of interest, and is often confused with positioning, which refers to the raw data that expresses the position of the object with coordinates [3]. Localization includes the environmental information in addition to the coordinates of the object. Localization process can be divided into three discrete steps, (i) estimating distance by using selected technique, (ii) position calculation, and (iii) localization from position

Technologies

Localization techniques are typically divided into two groups, namely, indoor and outdoor localization. For outdoor environment, GPS is the most widely used positioning system and it uses the ToA of radio signals to estimate the distance between the GPS satellites and the receiver. Each satellite has a synchronized atomic clock. The GPS system is sensitive to environmental conditions and nLoS situations. Typically, localization accuracy decreases in cities near high buildings, in indoor

Performance metrics

In order to support MC systems, the performance parameters such as power consumption, availability, latency and reliability have to be considered during system design.

Application scenarios

Recent advances in ubiquitous communication and IoT enabled us to develop intelligent systems for smart cities in the fields of transportation, energy, tourism, lighting, monitoring and surveillance. Localization within the scope of the development of public services is particularly important in emergency responses. These services are used by all residents, and the use of innovative technologies is beneficial for the prestige of the cities. A new application scenario can be integrating

Open research issues

Addressing MC scenarios in smart cities using cognitive RF based localization poses various difficulties. Open research issues often try to address distribution, reliability, complexity and power consumption topics. In addition, the future research areas are related to artificial intelligence, machine learning, deep learning and big data. Some studies are performed on positioning for emergencies in challenging areas such as underwater. In such cases, efficient location calculations cannot be

Conclusion

With the development of recent technologies, new standards are being established for urban areas, roads, facilities and buildings. It is now possible to continuously monitor for the safety of people and infrastructures, and in case of an emergency, secure and quick response to the situation is required. Generally, the studies aim to solve the problems that may be experienced during the normal operation of the systems. However, these scenarios need to be developed to cover the moments of crisis.

Author participations

Fadi Al-Turjman and Umit Deniz Ulusar participated in conceptualization, administration, management, coordination and planning of the study. All authors have participated in execution, draft preparation and creation of the work, specifically writing the initial draft, making changes and writing response to reviewers.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement

This study was funded by Akdeniz University Scientific Research Unit.

Umit Deniz Ulusar: Umit Deniz Ulusar received his B.Sc. in Computer Engineering from Marmara University, was awarded his M.Sc. at Bogazici University and Ph.D. from University of Arkansas, USA. He is currently associate professor, head of IT department and ComNets laboratory in Akdeniz University. His main research interests are IoT, big data and medical signal processing.

References (30)

  • Z. Farid et al.

    Recent advances in wireless indoor localization techniques and system

    J Comput Netw Commun

    (2013)
  • A. Alarifi

    Ultra wideband indoor positioning technologies: analysis and recent advances

    Sensors

    (2016)
  • M. Segura et al.

    Ultra wideband indoor navigation system

    Sonar Navig IET Radar

    (2012)
  • S. Wan et al.

    To smart city: public safety network design for emergency

    IEEE Access

    (2018)
  • A. Boukerche et al.

    Localization systems for wireless sensor networks

    IEEE Wirel Commun

    (2007)
  • Cited by (11)

    View all citing articles on Scopus

    Umit Deniz Ulusar: Umit Deniz Ulusar received his B.Sc. in Computer Engineering from Marmara University, was awarded his M.Sc. at Bogazici University and Ph.D. from University of Arkansas, USA. He is currently associate professor, head of IT department and ComNets laboratory in Akdeniz University. His main research interests are IoT, big data and medical signal processing.

    Gurkan Celik: Gurkan Celik received his B.Sc. in Electronics & Communication Engineering from Kocaeli University and was awarded his master's degree in Electrical & Electronics Engineering from Akdeniz University. He is currently a Ph.D. candidate at Akdeniz University and research assistant at ComNets laboratory. His main research interests are IoT, WSNs and wireless localization techniques.

    Fadi Al-Turjman: Fadi Al-Turjman received his Ph.D. from Queen's University, Canada. He is a full professor and a research center director at Near East University, Nicosia. He is a leading authority in the areas of smart/intelligent systems. His publication history spans over 250 publications in journals, conferences, patents, books, and book chapters, in addition to numerous plenary talks at flagship venues.

    This paper is for CAEE special section VSI-mis. Reviews processed and recommended for publication to the Editor-in-Chief by Guest Editor Dr. Chan-Yun Yang.

    View full text