Implementation and evaluation of efficient data dissemination algorithms for cooperative perception in vehicular networks

Autonomous vehicles require timely and reliable perception of their surrounding environment, but the sensing capability of a single vehicle is limited by occlusions, sensor range, and adverse traffic conditions. Cooperative perception mitigates these limitations by allowing connected vehicles to share locally detected objects and extend their collective awareness. This is especially important for vulnerable road users (VRUs), such as pedestrians and cyclists, whose safety depends on being detected and communicated early even when they are partially occluded or outside the direct sensing range of some vehicles. However, transmitting all available perception data can overload the wireless channel, increase latency, and generate redundant information. This thesis investigates a Value of Information (VoI) based dissemination strategy for cooperative perception in a decentralized vehicle-to-vehicle (V2V) sidelink deployment. The objective is to translate an abstract VoI-based selection model into a packet-level ns-3 simulation using the MilliCar vehicular millimeter-wave (mmWave) module and to compare it with broadcast-based dissemination baselines under the same scenario, traffic, and network assumptions. The implementation represents vehicles, obstacles, camera-specific perception messages, User Datagram Protocol (UDP) packet generation, Radio Link Control (RLC)-layer operations, and reception statistics inside ns-3. Realistic perception traffic is modeled using SELMA traces. The proposed simulation framework evaluates dissemination behavior in terms of delivered packets, latency, delivered VoI, redundancy, and obstacle coverage. By connecting semantic message selection with packet-level network behavior, this work provides a basis for assessing whether VoI-aware cooperative perception can reduce unnecessary transmissions while preserving useful information delivery under realistic vehicular communication constraints.