An Efficient Framework Model for Optimizing Routing Performance in VANETs

Nori Al-Kharasani, Zuriati Zulkarnain, Shamala Subramaniam, Zurina Hanapi
2018 Sensors  
Routing in Vehicular Ad hoc Networks (VANET) is a bit complicated because of the nature of the high dynamic mobility. The efficiency of routing protocol is influenced by a number of factors such as network density, bandwidth constraints, traffic load, and mobility patterns resulting in frequency changes in network topology. Therefore, Quality of Service (QoS) is strongly needed to enhance the capability of the routing protocol and improve the overall network performance. In this paper, we
more » ... uce a statistical framework model to address the problem of optimizing routing configuration parameters in Vehicle-to-Vehicle (V2V) communication. Our framework solution is based on the utilization of the network resources to further reflect the current state of the network and to balance the trade-off between frequent changes in network topology and the QoS requirements. It consists of three stages: simulation network stage used to execute different urban scenarios, the function stage used as a competitive approach to aggregate the weighted cost of the factors in a single value, and optimization stage used to evaluate the communication cost and to obtain the optimal configuration based on the competitive cost. The simulation results show significant performance improvement in terms of the Packet Delivery Ratio (PDR), Normalized Routing Load (NRL), Packet loss (PL), and End-to-End Delay (E2ED). Adjusting routing protocols for V2V communication network in order to satisfy various QoS requirements in applications was studied in [6] [7] [8] . However, the natural characteristics of the vehicular network are constrained by several factors such as limited link capacity, bandwidth, and transmission range coverage. These restrictions lead to a reduced routing efficiency and an increased routing overhead [1] . The objectives of most of the current routing protocols focus on either tuning the routing parameters or adapting the efficiency of route selection to reduce network overhead. Several routing protocols were presented to utilize network resources and enhance the routing efficiency in VANETs [9] . However, most of these protocols still have shortcomings when meeting the QoS requirements and in guaranteeing the stability of network topology during the routing process. Adopting suitable routing parameters or routing functionality to enforce different QoS achieving a high level of reliability is dependent mainly on network characteristics and their constraints. A high level of communication reliability is needed to enhance traffic prioritization and optimize network resources. Due to the high-level perspective, MANET is not suitable in VANETs, it needs to be developed to meet the specific characteristics of VANETs. Different methods to tune the routing parameters had previously been introduced to ensure a high level of performance in the dynamic network [10], yes most of them prove inadequate. Most of the them considered QoS as a major requirement to enhance the routing efficiency [11] , overcomes a number of problems such as unpredictable node density [12] and the limitation of coverage [13, 14] . Moreover, to address characteristics of a highly dynamic scenarios, this study can be extended by several research areas such optimization-based multiple objective which enhance delivery via D2D Communications in term of energy efficiency [5], improving the transmission capacity in dense network [15] , and in a heterogeneous mobility network where the delay or disruption network tolerance presented in [16] . Most of the recent studies considered QoS to optimize the routing efficiency [17] by introducing the optimization concept [18] and Particle optimization (PSO) technique [19] in order to reduce the impact of roads constraints on routing performance [20]. Tuning routing configuration parameters is one of the most common strategies used to keep track of neighboring node status [21] . A number of studies in the literature adjusted routing configurations making it suitable and easily applied into different kinds of network [22, 23] . In addition, studies tried to improve routing efficiency mainly in terms of network connectivity or available bandwidth by using multi-objective optimization strategy [22, 24] . However, most of recent studies in literature succeeded somewhat in enhancing the routing performance for one or two features, but it also loses other features or part of its routing efficiency. This is because of the limitation capacity of MANETs routing protocols. Thus, the relationship should be detected between the routing performance and the influential factors of V2V system to correctly develop MANET protocols for VANET. Addressing the trade-off and balancing problem in VANET is critically required, it helps comprehending the trade-off status between QoS requirements and the routing efficiency. In this paper, we present a framework model designed specifically to tune the routing configuration parameters in high dynamic mobility environment. This model exploits the network resources to improve the stability and reliability of routing packets between vehicles, which it index the QoS metrics as a function to balances the QoS requirements and the nature of the dynamic mobility impact. This balance reduces the trade-off between the communication cost and time needed to deliver packets successfully. In addition, we present The routing protocol for vehicular networks in urban environment called Balanced Optimized Link State Routing-Particle swarm optimization (BOLSR-PSO). This protocol is the end result obtained when the framework tune Optimized Link State Routing-Particle swarm optimization (OLSR-PSO), which improved its efficiency using the network resources by maximizing PDR, Throughput, and minimizing both NRL and E2ED. The remainder of this paper is organized as follows; Section 2 provides an overview of the work aimed at optimizing the routing performance in dynamic topology changes. The proposed protocol and framework components are described together with the problem statement in Section 3. Section 4
doi:10.3390/s18020597 pmid:29462884 pmcid:PMC5855465 fatcat:iwpis3m6qzdc3grwxdjozaalv4