Aodv

What Is AODV?

AODV (Ad hoc On-Demand Distance Vector) is a reactive routing protocol designed for mobile ad hoc networks (MANETs) in which nodes establish routes only when a source node actually needs to communicate with a destination, rather than maintaining tables of routes to all possible nodes at all times. Defined in IETF RFC 3561, the protocol was authored by Charles Perkins, Elizabeth Belding-Royer, and Samir Das and first published in 2003. AODV is suited to networks where node mobility causes frequent topology changes and where maintaining full proactive routing tables would consume more bandwidth and memory than the traffic warrants.

The protocol inherits its distance-vector foundation from established routing algorithms while adding destination sequence numbers, a mechanism that ensures loop-free paths and avoids the counting-to-infinity problem that afflicts classical Bellman-Ford approaches in dynamic environments. AODV has been widely studied in the research community and implemented in network simulators such as ns-3 and INET, serving as a baseline against which newer MANET and mesh routing schemes are often evaluated.

Route Discovery

When a source node requires a path to a destination for which it holds no valid route entry, it broadcasts a Route Request (RREQ) message containing the source address, a broadcast identifier, the destination address, and the source's current sequence number. Each intermediate node that receives the RREQ creates a reverse route entry pointing back toward the source, increments a hop count, and rebroadcasts the packet unless it has already seen this particular request. When the RREQ reaches either the destination itself or an intermediate node holding a sufficiently fresh route to the destination, that node generates a Route Reply (RREP) unicast back along the reverse path. Nodes along the return path install forward route entries as the RREP passes through.

Destination sequence numbers are central to freshness determination. Each node maintains a monotonically increasing sequence number for itself and records the sequence number associated with any stored route. A node considers a cached route superseded if the RREQ carries a higher destination sequence number than the one in the cache, preventing stale routes from blocking fresher alternatives discovered through a new route request cycle.

Once a route is in use, AODV relies on link-layer or network-layer notifications to detect broken links. When a forwarding node discovers that its next hop is unreachable, it generates a Route Error (RERR) message listing the destinations now unreachable through the failed link and propagates it upstream toward affected sources. Sources that receive an RERR must either issue a new RREQ if they still need the path or discard the route entry. Nodes also use hello messages, periodic broadcasts to direct neighbors, to confirm that neighboring nodes remain reachable when active routes pass through them.

This maintenance scheme keeps control traffic proportional to actual route usage rather than to the number of possible destinations, which is the property that makes AODV attractive for sparse or intermittently connected networks. A detailed simulation-based analysis of AODV route maintenance behavior is available through the ns-3 AODV model documentation, which describes the protocol implementation and parameter interactions.

An extended variant, AODV-RPL (RFC 9854), adapts the on-demand approach for low-power and lossy networks such as IEEE 802.15.4-based sensor deployments, blending AODV's route discovery mechanics with the RPL objective function framework.

Applications

AODV has applications in a range of fields, including:

  • Military and tactical communications, where infrastructure-independent mesh routing is required in contested environments
  • Disaster response networks, where deployed responders need rapid ad hoc connectivity without pre-installed access points
  • Vehicular ad hoc networks (VANETs), where vehicle-to-vehicle routing operates over rapidly changing topologies
  • Wireless sensor networks and IoT deployments, particularly in the AODV-RPL variant for low-power devices