This dissertation focuses on the methods of reliability, scalability, and manageability in smart utility networks. Smart Utility Network is a communication infrastructure based on wireless local area communication as IEEE 802.11 (WLAN), IEEE 802.15.4 (WPAN) that is widely considered in smart grid, gas, water and factory services. Especially, to enable cost-effective and flexible last-mail smart meter network, utilization of smart utility network is widely proposed by not only academic but also industrial both and next-generation wireless meters has been developed and investigated. However, WLAN and WPAN protocols may not provide QoS of smart gird service in the large-scale last-mile smart grid network because these protocols are originally designed for local area communication between devices.
This dissertation deals with the analysis of current wireless communication and control protocols in smart utility environment and proposes new schemes to overcome scalability problem in large-scale last-mile smart grid network. The proposed schemes can be listed as IEEE 802.11s based layer 2 routing, efficient neighbor discovery in TVWS environment and Software Defined Networking based network management method. The proposed schemes are evaluated via network simulator and test-bed implementation, and results show reduced control message overhead, end-to-end transmission delay and improved end-to-end packet delivery ratio. Especially, the proposed schemes can be interoperable with standard protocols, thus it can be applied to current smart grid network system efficiently.