The cost and limited capacity of fronthaul links pose significant challenges for the deployment of ultra-dense networks (UDNs) in the next generations of mobile networks, particularly for dense architectures such as cell-free massive MIMO (CF-mMIMO). Hence, cost-effective planning of reliable fronthaul networks is crucial for the future deployment of UDNs. This thesis proposes an optimization framework for traffic-aware hybrid fronthaul network planning, aimed at minimizing total costs through an integer linear program (ILP) that incorporates wired and wireless fronthaul candidate technologies, including fiber optics, millimeter wave (mmWave), and free-space optics (FSO). Beyond cost optimization, the framework also guarantees satisfying key performance metrics across diverse network configurations. The proposed framework has been validated across different dense network topologies, including small cell systems and radio-stripes based CF-mMIMO, all adhering to the open radio access network (O-RAN) architecture. Results highlight its superiority in achieving cost-effectiveness and high performance compared to alternative deployment schemes. Specifically, fiber emerges as the dominant technology in high-traffic and in dense Distributed Units (DUs) deployment scenarios, mainly due to its superior capacity and reliability. Conversely, mmWave is favored in low-traffic areas or moderate DU densities, where shorter distances between access points (APs) and DUs mitigate signal degradation. While FSO, due to its sensitivity to environmental conditions and reliability limitations, usually plays a minimal complementary role. The results also emphasize the need for diversified fronthaul planning in UDNs, as hybrid approaches maximize infrastructure utilization, reduce redundancy, and support both current traffic demands and future scalability. Furthermore, the findings unveil critical insights and trends that are invaluable for Service Providers (SPs) during the fronthaul planning phase, enabling the design of future-proof networks capable of adapting to evolving traffic demands. This work underscores the importance of effective fronthaul planning in preparing UDNs for next-generation mobile networks.