📘 Resource Overview & Documentation Details

You are viewing the technical directory listing for QoS-Aware Routing for Connected Vehicle Networks. This engineering resource is curated under our Internet of Service division to support automation specialists, control systems engineers, and manufacturing technicians looking for verified reference materials.

🔍 Contextual Abstract & Technical Scope

Share Summary:LinkedInEmail

In connected vehicle networks, the shortest route is not necessarily the most dependable one. Vehicles move quickly, wireless links fluctuate, and traffic density can vary from one road segment to the next. These conditions make QoS-aware routing important for applications that depend on messages reaching their destination across multiple vehicle-to-vehicle or vehicle-to-roadside-unit (RSU) hops.

A proposed quality-of-service-aware routing algorithm, called QRA, addresses this challenge in Software-Defined Internet of Vehicles (SDIoV) environments. Rather than selecting a path based on distance alone, it considers both the probability that links remain connected and their signal quality. The goal is to help a network controller choose a route suited to current network conditions.

Why shortest-path routing can fall short

Software-defined networking separates network control from packet forwarding. In SDIoV, a centralized controller can use network information to help coordinate routes, while RSUs can provide local support. This architecture offers a way to adapt routing decisions as vehicle positions and wireless conditions change.

However, a controller’s global view does not by itself make every route reliable. A geometrically short path may contain weak links or road segments with limited vehicle connectivity. If a link fails during transmission, packets may be delayed or require a new route. Hop-by-hop geographic forwarding can also encounter local maxima, where a vehicle cannot find a suitable next hop closer to the destination.

For time-sensitive or emergency communications, these issues matter: a route should reflect link conditions and connectivity, not simply the number of hops or physical distance. Route discovery and RSU selection also affect the amount of control traffic exchanged with the network.

How QoS-aware routing evaluates candidate paths

QRA represents an urban road network as a graph. Intersections are nodes, and the road segments connecting them are links. The controller evaluates candidate paths using two principal measures:

  • Probability of connectivity: An estimate of the likelihood that communication can be maintained over the road segments in a route. The route-level value accounts for the connectivity of its individual segments.
  • Signal-to-interference-plus-noise ratio (SINR): A measure of received signal quality relative to interference and noise. The algorithm applies an SINR threshold so that routes must meet a minimum link-quality condition.

These measures address different aspects of route quality. Connectivity indicates whether vehicles are likely to be available to relay packets; SINR indicates whether wireless conditions are suitable for communication. QRA combines them in an objective function to rank candidate paths, with weights controlling their relative influence. This makes route selection a multi-criteria decision rather than a simple shortest-path calculation.

The connectivity model accounts for the variable wireless environment. The underlying analysis uses a Nakagami fading model and considers interference from transmitting vehicles and RSUs. In practice, the intent is to represent the fact that received signal strength and link quality can vary, rather than treating every connection as equally dependable.

Intersection decisions account for traffic conditions

Route choice in an urban grid is not only a matter of evaluating complete paths. A packet carrier must also choose which intersection to approach next. QRA uses a multi-score function that considers two factors:

  • Greediness factor: A measure related to how close a neighboring intersection is to the destination. Intersections closer to the destination receive greater priority under this factor.
  • Traffic density: An indicator of vehicle distribution and, in turn, the potential for connectivity along nearby road segments.

Combining these inputs lets the routing process balance progress toward the destination with the likelihood that a road segment can support forwarding. Choosing the closest intersection alone could favor a direction with poor connectivity; considering traffic conditions helps inform that local decision.

Using a discrete optimization method to select a route

Once route discovery produces candidate paths, the controller must identify the best option under the selected QoS criteria. QRA uses a modified Laying Chicken Algorithm (LCA) to search this discrete set of routes. The method was adapted from an approach designed for continuous optimization problems.

In the routing application, candidate routes are feasible solutions, and the objective function assigns each a value based on connectivity and SINR. The search compares solutions and updates its preferred route as it proceeds. This gives the controller a way to select among multiple discovered paths rather than automatically relying on a fixed shortest-path method.

Route discovery involves the source vehicle sending a query through a nearby RSU. If a current route is not available, controllers can initiate discovery. Discovery packets travel through the network, recording intersections and link SINR. A route that reaches the destination while satisfying the SINR requirement can be returned to the source as a candidate. The controller then evaluates available candidates and provides a route for data forwarding.

What the simulation evaluation indicates

The algorithm was evaluated in simulated urban vehicle scenarios using NS-3 for network simulation and SUMO for traffic mobility. Its performance was compared with GPSR, a geographic routing protocol, and a centralized routing protocol. The evaluation considered packet delivery ratio, end-to-end delay, routing overhead and controller overhead, as well as changes in vehicle density and packet generation rate.

Reported results indicate that QRA achieved higher packet delivery and lower average end-to-end delay than the comparison approaches in the evaluated scenarios. The simulations also illustrate an important trade-off: discovering multiple candidate routes and evaluating them can add routing overhead. The authors report higher routing overhead than the centralized comparison protocol, while describing lower controller overhead in some comparisons through the use of local RSUs.

Performance also depends on network conditions. Greater vehicle density can improve connectivity, but it can increase channel contention and interference. Higher packet generation rates add network load, while sparse traffic can make it harder to maintain connected paths. These results are simulation findings, not a guarantee of performance in every deployment.

Practical implications for network designers

For engineers and system designers, the central lesson is to define routing objectives around application and network requirements. A route that minimizes distance may not satisfy reliability needs when wireless links are unstable. Combining link quality, connectivity and traffic-aware intersection decisions offers a more contextual basis for selecting paths.

That approach also requires attention to operational trade-offs. Candidate-route discovery consumes communication resources, and controller decisions rely on sufficiently current network information. When evaluating an SDIoV routing strategy, teams should therefore consider delivery performance alongside routing and controller overhead under the traffic conditions relevant to their use case.

Conclusion

QoS-aware routing provides an alternative to shortest-path selection for dynamic vehicle networks. QRA combines connectivity probability and SINR to rank candidate routes, uses traffic density and destination proximity to guide intersection choices, and applies a modified LCA to select among discovered paths. Its simulation results point to potential gains in packet delivery and delay, alongside additional routing overhead. For network planners, the practical priority is to balance route reliability with the control traffic and changing conditions of the intended deployment.

Access Official Technical PDF