An Energy-Efficient Model for Opportunistic Data Collection in IoV-Enabled SC Waste Management

1. Introduction

The increasing connection of vehicles to the Internet of Things (IoT) has ostensibly given rise to the recent technological advancements which have paved way for the emergence of a new era of vehicular networks known as the Internet of Vehicles (IoV). In the context of Smart City (SC) scenarios, these vehicular networks can be exploited as Mobile Data Collectors (MDCs) or data MULEs for waste management strategies. The era of IoT technology has given birth to the development of several innovative applications for smart cities. Essentially, IoT is expected to be the bedrock for novel applications and technological platforms for the emergence of the IoV from the existing vehicular ad hoc networks (VANETs), characterized by a seamless integration of sensor-driven mobile vehicles, and mobile devices within the network environment. Hence, IoT is expected to permit the evolution of IoV with new technological features and transformations in SCs. One such category of applications is the use of smart vehicles (IoV) as data MULEs for data collection in a SC waste management, which is the essence of this study. Other applications of the use of vehicular networks for data collection in smart cities may be found in transportation (i.e. intelligent transport), traffic light, environmental data sensing, monitoring, surveillance etc. The electronic toll which involves a communication between vehicles and roadside (V2R) in which vehicles send signals to the toll to make payments to the highway authorities is also another form of IoV applications. As mentioned earlier, to the best of our knowledge, this scheme is yet to be exploited in the context of SC waste management scenarios. Essentially, vehicular networks can be exploited for data collection meant for SC applications. Hence, Berrilio (2014) proposed the use of a network infrastructure for Data Collection for Low Energy Devices (DC4LED) integrated with sensors for data transmission to the internet and capable of providing connectivity to vehicular-sensors and access point so that data generated from the sensors could be transmitted to the internet via the access point. Other applications of IoV may include traffic & routing information (also known as a GPS application) that informs individuals of the route to a destination, security and collision avoidance where the driver is warned about a possible danger ahead such as an oil spill, a car accident, and so on (Berrilio, 2014).

Silva et al. (2017) proposed the use of ITS-based Decision Maker (DM) architecture for opportunistic networking based on heterogeneous and dynamic access network. The proposed scheme is expected to monitor a variety of information and will enable information exchange between vehicles and their environments for accuracy and better decision strategies. A feature of this scheme is its ability to provide the best route for a given process. Researchers (Er et al., 2017; Er et al., 2019) conducted the performance evaluation of four basic Vehicular Delay-Tolerant Networks (VDTN) routing protocols using the Opportunistic Networking Environment (ONE) simulator in terms of V2X communications for data delivery in SCs. DC4LED is a simple hierarchical forwarding routing scheme in the context of VDTN network. The focus of their work is the use of VDTN as an opportunistic network that can provide a data collection service by exploiting its V2X communication abilities. The results of their work showed that vehicles with predetermined routes presented high data delivery performance. Further to this was the use of DC4LED forwarding algorithm to determine the mobility features of vehicles by considering three kinds of vehicles (e.g. buses, taxis and cars), which can be considered advantageous for VDTN routing as well as sensor-driven mobile collectors.

Based on perceived dissimilarities in the mobility patterns of the selected vehicles, it is observed that buses follow almost fixed routes especially during their service hours, thus making them the best fit for VDTN routing applications, if the internet Point-of-Presence (PoP) is planted along their paths. On the other hand, cars and taxis traverse the streets which are not bus-routes, and thus collect data from the sensors planted along the streets in the city. However, compared to cars, taxis appear to traverse the cities much more. Cars make few trips with longer stationary times. Hence, based on this comparison, buses are found to be the best bet for sensor-driven mobile data collectors followed by taxis while cars are the least reliable in the hierarchical routing scheme.