Evaluation and analysis of an enhanced hybrid wireless mesh protocol for vehicular ad hoc network

Gateway vehicle selection and management is the prominent issue in the mechanism of multi-hop relay based on the Integrated VANET-cellular networks Heterogeneous Wireless Network. This is due to the non-similarity characteristics in vehicles, which lead to inability of the gateway to access. To solve this problem, one of the proposed solutions is to assemble them in groups (clusters) based on vehicles with similar characteristics and elect a cluster head (CH) for each group (cluster), which becomes the gateway node. Vehicle clustering is promising to resolve this issue, where cluster-head selection depends only on the geographical location of vehicles. In consequence, the cluster head of the cluster does not have the basic standards to manage the VANET groups, so that it enables them to connect with the cellular network.

The proposed mechanism stated in [3], Simplified Gateway Selection Scheme (SGSS), enabled vehicles to connect to the Internet and other mobile services continuously; it described two types of vehicles: vehicles within the coverage area had a dual interface and communicated directly to the UMTS base station transceiver. UTRAN and vehicles that move along the road looking for an associate to be used to connect to the UMTS base station transceiver had a single interface. Vehicles received different signals from the base station transceiver. This mechanism showed lower reliability in terms of availability. In addition, the researcher did not take into account the propagation delay between the relay request message and the relay response message.

A Cluster-based Multi-metric, adaptive mobile Gateway Management mechanism (CMGM) was presented, where two kinds of vehicles were described [4], namely, vehicles equipped with IEEE 802.11p and vehicles with UTRAN interfaces. It was discovered that the mechanism allowed the small number of vehicles (GW) with IEEE 802.11p to connect to the 3G networks and the UTRAN interfaced vehicles used this GW to connect to the 3GPP networks. The clustering scheme utilized to cluster GWc was based on the direction of movement, UMTS-RSS and IEEE 802.11p transmission ranges. If the GWc was closest to the middle of the cluster, then CH was elected in a cluster. However, the suggested mechanism in [2] was complex in terms of preserving the clusters for the fast-moving environment such as in VANET. In addition, the process of selecting the CH depends mainly on the geographical location of the vehicle. The result showed that the CH did not have the basic criteria for managing the cluster and enabling them to connect to the cellular network. In addition, the process of forming a cluster and the selection of the CH consumed more time compared with data exchange time.

In [5], an optimized hybrid wireless mesh protocol had been suggested to construct a routing protocol over VANET. A real-time packet loss estimation was utilized as an additional parameter to the parameters used in [4] (UMTS received signal strength, link stability, and available relay capacity) to elect an efficient mobile gateway node for the transfer of packets without losing any of them.

The authors in [6] suggested a Cluster-based Multi-metric, new Simplified Gateway Selection Scheme (SGSS), where the received signal strength (RSS) threshold was utilized as a cluster formation index considering the UMTS cell boundary. The Gateway had been selected based on the UMTS received signal strength, link stability, and available relay capacity. An enhanced hybrid wireless mesh protocol (E-HWMP) protocol was suggested to construct multi-hop routing through vehicular ad hoc networks (VANET).

The integration of IEEE802.11p VANETs with 3GPP LTE for seamless data connectivity had been proposed in heterogeneous wireless network architecture [7]. Issues on CH based on multicasting and QoS was described in this study. The approach had suggested the mechanism for the selection of an adaptive multi-metric CH to manage VANET sub-clusters and to provide multimedia communication services over spatially apart vehicular clusters. As expected from the effective cluster-head selection mechanism, there was a suggested approach for a virtual two-hop overlay mesh-based shared tree for lower level VANET multicasting. The results showed that the integrated system had an acceptable output in terms of throughput and end-end delay for the LTE network.

The mechanism proposed in [8] was to connect vehicular ad hoc networks to the infrastructure networks. The work included an experiential estimate of relay vehicle utilization when UMTS Node B was far away from the source vehicle. This scheme allowed the source vehicle to find a relay vehicle before exiting the coverage zone of the RSU. When implementing this mechanism, the achieved result was the extension of the network coverage zone.

In [9], the authors proposed spatial reusability-aware single-path routing (SASR) and any path routing (SAAR) protocols and compare them with existing single-path routing and any path routing protocols, respectively. The evaluation results show that our protocols significantly improve the end-to-end throughput compared with existing protocols. Specifically, for single-path routing, the median throughput gain is up to 60 %, and for each source-destination pair, the throughput gain is as high as 5.3×; for any path routing, the maximum per-flow throughput gain is 71.6 %, while the median gain is up to 13.2 %.

In [10] the authors propose a cross-layer distributed algorithm called interference-based topology control algorithm for delay-constrained (ITCD) MANETs with considering both the interference constraint and the delay constraint, which is different from the previous work. They focus on the delay-constrained topology control problem and take into account delay and interference jointly. The transmission delay, contention delay, and the queuing delay are taken into account in the proposed algorithm. Moreover, the impact of node mobility on the interference-based topology control algorithm is investigated, and the unstable links are removed from the topology. The simulation results show that ITCD can reduce the delay and improve the performance effectively in delay-constrained mobile ad hoc networks.

In [11], the authors propose a reliable multicast protocol, called CodePipe, with energy efficiency, high throughput, and fairness in lossy wireless networks. Building upon opportunistic routing and random linear network coding, CodePipe can not only eliminate coordination between nodes but also improve the multicast throughput significantly by exploiting both intra-batch and inter-batch coding opportunities. In particular, four key techniques, namely, LP-based opportunistic routing structure, opportunistic feeding, fast batch moving, and inter-batch coding, are proposed to offer significant improvement in throughput, energy efficiency, and fairness. Moreover, the authors design an efficient online extension of CodePipe such that it can work in a dynamic network where nodes join and leave the network as time progresses. The proposed protocol evaluated on ns2 simulator by comparing with other two state-of-art multicast protocols, MORE and Pacifier. Simulation results show that CodePipe significantly outperforms both of them.

In [12], the authors present a directional routing and scheduling scheme (DRSS) for green vehicle DTNs by using Nash Q-learning approach that can optimize the energy efficiency with the considerations of congestion, buffer, and delay. Their scheme solves the routing and scheduling problem as a learning process by geographic routing and flow control toward the optimal direction. To speed up the learning process, their scheme uses a hybrid method with forwarding and replication according to traffic pattern. The DRSS algorithm explores the possible strategies and then exploits the knowledge obtained to adapt its strategy and achieve the desired overall objective when considering the stochastic non-cooperative game in online multi-commodity routing situations. The simulation results of a vehicular DTN with predetermined mobility model show DRSS achieves good energy efficiency with learning ability, which can guarantee the delivery ratio within the delay bound.

In [13], the authors study the heterogeneous media provision in peer-to-peer (P2P)-based vehicular networks and develop fully dynamic service schemes with the goals of maximizing the total user-satisfaction and achieving a certain amount of fairness. We first construct a general user-satisfaction model according to the network transmission mechanism, as well as different media delay-satisfaction characteristics. Then, the media service was formulated as an optimization problem and proposes a joint content dissemination and cache update scheme. Also, the exact steps to achieve the optimal solution at equilibrium, given the user-satisfaction function, were provided. Furthermore, the proposed service scheme was extended by addressing the fairness problem. Unlike prior works that target at bandwidth or demand fair, a media-aware satisfaction-fairness strategy, which is aware of the characteristic of user-satisfaction and media content and ensures max-min satisfaction-fairness sharing among multiple vehicles, was proposed. It is worth noting that both schemes are designed in a distributed manner, which is amenable to online implementation for vehicle networks. In addition, extensive simulation results that demonstrate the effectiveness of our proposed schemes were provided.

In [14], the authors present overview characteristics of cloud computing and introduces the basic concepts of vehicular networks. An architecture for multimedia cloud computing is proposed to suit subscription service mechanisms. The tendency to equip vehicles with advanced and embedded devices such as diverse sensors increases the capabilities of vehicles to provide computation and collection of multimedia content in the form of the vehicular network. Then, the taxonomy of cloud-based vehicular networks is addressed from the standpoint of the service relationship between the cloud computing and vehicular networks. In this paper, we identify the main considerations and challenges for cloud-based vehicular networks regarding multimedia services identified and propose potential research directions to make multimedia services achievable. More specifically, the authors quantitatively evaluate the performance metrics of these researches. For example, in the proposed broadcast storm mitigation scheme for vehicular networks, the packet delivery ratio and the normalized throughput can both achieve about 90 %, making the proposed scheme a useful candidate for multimedia data exchange. Moreover, in the video uplink scenarios, the proposed scheme is favorably compared with two well-known schedulers, M-LWDF and EXP, with the performance much closer to the optimum.

In [15], the authors proposed a novel threshold credit-based incentive mechanism (TCBI) based on the modified model of population dynamics to efficiently resist the node compromise attacks, stimulate the cooperation among intermediate nodes, maximize vehicular nodes’ interest, and realize the fairness of possessing the same opportunity of transmitting packets for credits. Then, a TCBI-based privacy-preserving packet forwarding protocol was proposed to solve the open problem of resisting layer-adding attack by outsourcing the privacy-preserving aggregated transmission evidence generation for multiple resource-constrained vehicles to the cloud side from performing any one-way trapdoor function only once. The vehicle privacy is well protected from both the cloud and transportation manager. Finally, formal security proof and the extensive simulation show the effectiveness of their proposed TCBI in resisting the sophisticated attacks and the efficiency in terms of high reliability, high delivery ratio, and low average delay in cloud-assisted vehicular DTNs.

In [16], the authors classify and describe the most relevant vehicular propagation and channel models, with a particular focus on the usability of the models for the evaluation of protocols and applications. First, they classify the models based on the propagation mechanisms they employ and their implementation approach. They also classify the models based on the channel properties they implement and pay special attention to the usability of the models, including the complexity of implementation, scalability, and the input requirements (e.g., geographical data input). They also discuss the less-explored aspects in vehicular channel modeling, including modeling specific environments (e.g., tunnels, overpasses, and parking lots) and types of communicating vehicles (e.g., scooters and public transportation vehicles). They conclude their paper by identifying the under researched aspects of vehicular propagation and channel modeling that require further modeling and measurement studies.

Considering the above issues, an envisioned Integrated VANET-cellular networks Heterogeneous Wireless Network architecture is proposed. The architecture is depicted in Fig. 1. The scenario in this paper depicts a highway with a dual carriageway consisting of two lanes in each direction. With regard to placement, a GPS device was installed in all vehicles. Vehicles connect to different signal strengths on the road. New gateway is elected before the serving gateway loses its optimality to avoid the delay that occurs during the process of handoff. The connection between VANET nodes is multi-hop and is based on peer to peer.

The proposed clustering mechanism is employed upon the available GWcs (vehicles inside the converge zone and equipped with dual interface) cluster. The proposed algorithm1 is multi-metric algorithm, based on the Simple Additive Weighting (SAW) technique is used for gateway selection and for outranking the optimum gateway node. Three metrics are used for mobile gateway selection: LTE RSS (x ₁), available route capacity (x ₂), and stability of the link (x ₃). The stability of the link is defined by two metrics between the source and the CH: (1) Link Expiration Time (LET) and (2) Route Expiration Time (RET). LET is the link lifetime between the two adjacent vehicles and the minimum lifetime between the vehicles to select gateway which is referred to as Route Expiration Time (RET); the information of each metrics of each CH lies with the GWcs of the cluster. According to the given fact, the hybrid gateway discovery mechanism is employed.

E-HWMP is standard for wireless access to vehicular environments (WAVE) that employs a vehicular connection in vehicles (V2V). A Vehicle-to-Infrastructure (V2I) connection is not supported by the standard for wireless access. The Evolved Universal Terrestrial Radio Access Network (E-UTRAN) interface is utilized by the VANET to connect to the internet through eNodeB. Gateway numbers are required to be lower. The proposed scheme is cluster based. The clusters are formulated only at a certain point of the 3GPP LTE cell’s edges when the received signal strength (RSS) crumbles to less than assured threshold.

This proposed scheme adopts a hybrid [2] that is a combination of both proactive and the reactive approaches. Firstly, the gateway broadcasts the Gateway Advertisement message (GANN) proactively to an area confined to a limited number of hops which is known as the TTL value of the message. The TTL is set to advertisement zone so that the advertisement message can be forwarded only up to this maximal number of hops through the ad hoc network. This area is known as the proactive zone. The nodes that are outside the proactive zone establish connection to the gateway reactively, On-demand Broadcast of Gateway Request message (GREQ), by Active Sources in VANET.

In the hybrid approach, the gateway does always send a GWADV message every 5 s. The main purpose of this approach (hybrid gateway discovery mechanism) is to the control overhead due to the reactive discovery scheme when there are many source nodes in the VANET. This approach also aims to reduce the overhead of the proactive protocols when the number of gateways increases.

Architecture of our proposed mechanism implemented using E-HWMP protocol. In Fig. 2, message exchanges of the E-HWMP protocol are shown.