In this talk, I will discuss about the routing layer and transport layer design issues in UAV networks equipped with the latest antenna - multi-beam smart antenna. First, in Transport Layer layer, we propose a balanced transport control strategy between the above two categories, which is based on the idea of Adaptive Batch Coding (ABC). The proposed ABC scheme resists random loss through redundant coding and copes with congestion via window shrink and retransmission. Using a cross-layer design (between transport and routing layer), both the coding scheme and the traffic allocation are adaptively adjusted according to network conditions. Second, in Routing layer, for a typical hierarchical (2-level) UAV network, we propose to use bio-inspired algorithms (based on moth and ant’s behaviors), to achieve the high-throughput, low-latency two-level routing. Particularly we use male moth’s light source pursuing pattern to handle the low-level network routing from an event node to the highly mobile ’sink’ node, and use the ant’s chemical trail maintenance principle to trace the trajectory of commander node in order to deliver low-level data to the high-level commander node (in high-level network) with the minimized delay. Additionally we propose to construct a weighted fence routing topology among high-level nodes with multi-beam antennas, to achieve a high-throughput routing in high-level network.