Abstract: The allocation of bandwidth to unlicensed users, withoutsignificantly increasing the interference on the existing licensedusers, is a challenge for Ultra Wideband (UWB) networks. Ourresearch work presents a novel Rake Optimization and Power AwareScheduling (ROPAS) architecture for UWB networks. Since UWBcommunication is rich in multipath effects, a Rake receiver is usedfor path diversity. Our idea of developing an optimized Rakereceiver in our ROPAS architecture stems from the intention ofreducing the computation complexity in terms of the number ofmultiplications and additions needed for the weight derivationattached to each finger of the Rake receiver. Our proposed work usesthe Cognitive Radio (CR) for dynamic channel allocation among therequesting users while limiting the average power transmitted ineach sub-band. In our proposed novel ROPAS architecture, dynamicchannel allocation is achieved by a CR-based cross-layer designbetween the PHY and Medium Access Control (MAC) layers.Additionally, the maximum number of parallel transmissions within aframe interval is formulated as an optimization problem. Thisoptimal decision is based on the distance parameter between atransmitter-receiver pair, bit error rate and frequency of requestby a particular application. Moreover, the optimization problemimprovises a differentiation technique among the requestingapplications by incorporating priority levels among userapplications. This provides fairness and higher throughput amongservices with varying power constraint and data rates required for aUWB network.