Eco-inspired low latency performance for smart grid applications in wireless networks

As the Smart Grid matures, the number of User Equipment (UE) required for monitory and control applications will continue to rise. This will increase the traffic in an already burdened telecommunications network. This research aims to optimally allocate radio resources to UEs such that the Smart Grid Quality of Service requirements are satisfied, with minimal effect on pre-existing telecommunication traffic caused by the imposition of Smart Grid UEs. To address this resource allocation problem, a Lotka-Volterra based resource allocation algorithm and scheduler was developed due to its ability to easily adapt to the dynamics of a telecommunications environment. Each `species' in the Lotka-Volterra scheme is modelled as either Data, Voice or Smart Grid traffic class in a Rayleigh-Fading channel. Unlike previous resource allocation algorithms, the Lotka-Volterra scheme allocated telecommunications resources to each class as a function of its growth rate. Subcarrier power allocation is non-uniform and the total power allocated to each class is based on the size of its queued traffic. By doing so, the Quality of Service requirements of the Smart Grid are satisfied, with minimal effect on pre-existing traffic. Class queue latencies are reduced by intelligent scheduling of periodic traffic and forward allocation of resources. Results also proved that for high traffic situations, the Lotka-Volterra resource allocation algorithm and scheduler provided a high degree of fairness between traffic classes, whilst ensuring minimum latency requirements of Smart Grid data