**1. Introduction**

New generation networks, also known as 5G networks, are slowly becoming the part of our lives. In the future, they will bring new opportunities and different views of the network of today. With the phenomenon known as the Internet of Things (IoT), the Internet will become increasingly complex, smart, and pervasive. Upcoming trends include smart homes, smart cities and Industry 4.0 with different applications, such as industrial automation, public health and information systems, city management, energy efficiency and public safety. Everything will be possible thanks to the new type of mobile networks, called heterogeneous network that will work as one functional complex network. New 5G networks are considered to be a promising technology that incorporates different types of networks to provide needed functionality and applications. Massive transport of the IoT data will require to use the alternative type of mobile network such as Wireless Sensor Networks (WSN), Mobile Ad-Hoc Networks (MANET), Drone Networks (DRONET) and use their benefits.

One of the urgent goals for the next generation networks are to provide uninterrupted public safety and health service in different scenarios, where fixed infrastructure will be disrupted. This scenario includes natural disasters (earthquakes, fires, floods, hurricanes), human errors (nuclear, chemical, biological, radiological exposures or railway and car accidents), and malicious criminal actions (terrorists or cyber-attacks) [1]. Other goals for designers of Public Protection and Disaster Relief (PPDR) agencies are not only to provide a reliable communication network for public safety agencies

and departments such as police, emergency, etc., but to provide the data services to all people during emergency and disaster situations, where existing fixed communication infrastructure could be destroyed [2,3].

The one possible solution to these problems could be a resilient multilayered network model, which is able to provide necessary services even in unpredictable situations mentioned above. In this paper, the multilayered network model composed of Wireless Sensor Networks (WSN), Mobile Ad-Hoc Networks (MANET), and Drone Networks (DRONET) will be introduced. WSN networks are considered to be a type of IoT network composed of numerous low energy sensors, which are responsible for collecting measured data. WSN sensors could be situated in different types of environments, such as hospitals for collecting critical health data, on the beach to collect sea level data before the tsunami waves or to the amount of CO2 and quality of the air during fires. Collected data are then distributed in a multihop ad-hoc manner to cloud data centers for processing.

Our proposal deals with destroyed fixed infrastructures, where fixed Access Points are not functional. Urgent data from the WSN network cannot be delivered and this could lead to system errors or misleading information on the server-side. The solution of this problem could be the MANET network, which is an autonomous self-organized network that could offer support for rescue operations and was used before by the military to surveying inaccessible areas. Thanks to the support of mobility, higher data rates, and lower energy constraints, urgent data could be delivered from the WSN network through the MANET network to another operational Access Point in a multihop manner with respect to device-to-device (D2D) communication principals. The reach can be extended by using unmanned aerial vehicles (UAVs) such as drones in DRONET network. UAVs can be used for data collection, or for delivery of urgent data from isolated MANET subnetworks, which were created by the movement of MANET nodes.

The aim of this study is to provide innovative and resilient way to deal with the failure or disruption of fixed infrastructure for upcoming 5G networks and its urgent applications. With a unique combination of WSN, MANET, and DRONET networks, it is possible to preserve the functionality of urgent applications, such as public health and safety, in different adverse situations. Based on this motivation, the multilayered model was build in order to use the advantages of mentioned networks and also empower the potential of new generation 5G networks and IoT solutions. The simulations of the proposed multilayered model show that it is able to provide a fully functional backup solution that preserves functionality and service demands required by 5G standards.

## **2. Related Work**

In recent years, many studies investigated the possibilities of different networks convergence scenarios. Most of the works include MANET networks, WSN networks, Vehicular Ad-Hoc Networks (VANET) or networks composed of unmanned aerial vehicles (UAVs). Bellavista et al. [4] proposed a MANET and WSN convergence network model to support a cost-effective realization of wide-scale urban monitoring applications. The authors assumed a tree-based data collection for WSN with generic tree-based protocol to easily enable its deployment and immediate usage with all emerging collection solutions and standard specifications. The MANET network organized in small local clusters acts as a WSN backbone network that allows urgent data to pass.

Erdelj et al. [1] described the advances in wireless sensor network (WSN) technology and unmanned aerial vehicles (UAVs) to enhance the ability of network-assisted disaster prediction, assessment, and response. UAVs are responsible for the data collection from fixed WSN sensors deployed in different areas. The authors introduced recommendations for WSN and UAV use during different disaster stages, but there are missing technology and protocol background. In [5], the solution was extended about the measurement of major UAV communication technologies and authors discussed the possible communications technology with Quality of Service (QoS) point of view. A conceptual mobile UAV station for disaster management was proposed as well.

Mukherjee and Biswas [6] propose IoT network hierarchy comprising the Internet, WLAN and/or Internet gateway, MANET and WSN networks. Wireless sensor nodes were deployed in an IoT system that collecting data from the environment and sent them to the gateway node. Data can be directly sent to the Internet at the highest hierarchy level through WLAN Access Point or Internet gateway. MANET nodes are acting as intermediate nodes responsible for collecting data from WSN nodes if the direct Internet gateway is not available. Different communication technologies along with protocol stacks are also discussed [6]. In this work [6] to overcome interference between MANET (Wi-Fi IEEE802.11b) and WSN (ZigBee IEEE802.15.4) communication, the authors defined usable non-overlapping channels for WSN and MANET. Other related works considering convergence scenarios are [7–9].

Unlike the research presented above that interconnects different variations of networks, in our proposal, we introduce a convergence scenario that interconnects WSN, MANET, and DRONET networks into a layered model with the ability to collect data and send it to the cloud services for processing. The layers are independent, but in case of network disruption, urgent data can pass layers based on a system of gateways that enables the interconnection of multiple layers. For each network layer, we provide communication technology and routing protocols recommendation. Besides that, our contribution includes the description of necessary changing and exceptions for routing protocol deployed on each layer (i.e., network) that allows the transporting of urgent data through multiple layers in the multilayered network model. In particular, in DRONET layer, we provide a simplified mechanism for UAV management in order to cover MANET nodes by slicing MANET networks into clusters. The main aim of this proposal is to provide a conceptual way for critical data of urgent applications to be continuously delivered in disrupted network scenarios caused by unpredictable situations.
