Project type: Graduation assignment

Short descriptions:

1. Energy Aware Routing study
   From a clean slate, find real world use cases that would only be feasible with battery (or energy harvested) powered end nodes and routers, e.g. humidity and temperature sensors in open field precision agriculture. Characterize the use case in terms of the network parameters (size, number of nodes, packet size etc.). Study literature for candidate energy aware routing algorithms targeted at embedded systems with limited CPU + memory. Simulate the network behavior in ns2 or ns3, identifying problems such as hot spots in the routing and, if possible, propose solutions.
2. Large ZigBee Networks
   Typically ZigBee networks are built using embedded systems with limited (RAM) memory. In practice this means that the number of nodes in a network is limited by the sizes of the routing tables in memory. The assignment is to (1) investigate the trade-offs in ZigBee networks where the routing tables are used as a "cache" and dynamic route discovery is used on a "cache miss" and (2) to characterize the network behavior (e.g. latency, additional radio traffic, node energy usage, end-2-end reliability) using simulations in ns2 or ns3.
3. Network Planning
   For WSNs in commercial buildings optimal node placement is a difficult and time consuming task, often leading to slow uptake of wireless networks in building automation despite their advantages. Assume an executable model is available that for a given floor plan calculates the signal loss from a transmitter at point to any point on that floor plan. The assignment is to develop a tool that will calculate the viability of a ZigBee network where the node positions are assigned by a human. A possible, ambitious, second step is to investigate an algorithm for (near) optimal automatic placement of the coordinator and router nodes so that transmissions from the given sensor nodes are guaranteed to succeed.

Organization: NXP

Contact: dr. R.H. Mak or prof. dr. Johan Lukkien