Existing wireless technologies provide communication and information services to all fields of life. The one of the emerging and popular field is vehicular ad hoc networks, with its
IntroductionVehicular ad hoc networks (VANETs) are expected to support the large spectrum of commercial and safety applications range from safety to comfort for travelers. The main objective of vehicular communication is providing real-time road information of dangerous situations in advance to drivers such as accident detection, weather information. However, due to high vehicle mobility, the vehicle nodes are frequently changing their positions and its impact in the shape of link disconnection, network overhead, high transmission delay and low data packet delivery ratio issues [1]. In order to incorporated these issues, geographical routing protocols are more suitable solutions to ensure forward progress toward the destination by flooding messages with node position information such as greedy perimeter stateless routing (GPSR) [2], greedy perimeter coordinator routing (GPCR) [3], vehicle assisted data delivery (VADD) [4]. In these protoocls, the beacon messages are periodically broadcasted to inform one-hop neighbors location and presence by global positioning systems (GPSs). The result of regular beaconing messages, the wireless channel is more congested and packet collisions with communication overhead occur in the network. Although, the recovery strategies have been proposed to solve these issues, but these approaches are based on planner graph traversals, which are not suitable for high-velocity and urban environment. Without an effective multi-hop routing, these features are limited and have several complexities. Therefore, an effective routing protocol requires for in-time data delivery in vehicular netwokrs.To solve the frequent beaconing challenges in the network, various beaconless forwarding approaches have been proposed such as CBF [5], BRAVE [6], CoopGeo [7]. These beaconless approaches contain own and destination position in the data packet header and broadcast it to next one-hop neighbors. Afterwards, these approaches determine different routing metrics to find optimal forwarder in the network and deal with unique vehicular environment. The important point in beaconless protoocls is to select the appropriate routing metric to deal with vehciualr envirnemnt in their handshacke mechanism. In this paper, we proposed a beaconless packet forwarding (BPF) protocol, which is based on score function and self-election among vehicle nodes. The beaconless self-election forwarding is made in a way to