In a wireless sensor network (WSN) with specially deployed cluster heads (CHs), association relation between sensor nodes and the CHs is important as it affects the radio resource allocations, which further determine the overall network throughput, energy consumption, and other performance. In a lot of cases, the CHs are placed in random locations, and strong overlapping may exist between their coverage areas so that sensor nodes can choose to associate to different CHs. In this paper we first formulate two optimization problems that jointly consider sensor node association and radio resource allocations, one for maximizing the network level throughput, and another for balancing the energy consumption among the CHs. For each of the optimization problems, a heuristic scheme is designed that jointly considers both the timeline allocations of the CHs and the sensor node association. Numerical results based on computer simulation demonstrate that the proposed schemes achieve close-to-optimum performance. In addition, the schemes achieve much better throughput and energy performance than the straightforward association schemes.
I. INTRODUCTIONUbiquitous wireless sensor networks are expected to play an important role in the future society, such as for health care and environmental monitoring. A future wireless sensor network (WSN) is required to cover a wide coverage area with a large number of sensor nodes. In addition to the high throughput requirement, most of the sensor network devices are designed to be small and low-power for low data-rate transmissions, and therefore energy and power saving is important for extending lifetime of the network devices as well as lifetime of the network. The IEEE 802.15.4 standard defines the physical and medium access control (MAC) layers for low-rate, low-power and flexible wireless personal area networks (PANs). ZigBee Alliance defines three network topologies above the IEEE 802.15.4 physical and MAC layers, the cluster-tree topology, the star topology and the mesh topology. Both the star and cluster-tree topologies can use beacon frames to synchronize devices to their parent node, and thus minimize power consumption of the devices by intermittent operations. The clustertree topology has better scalability than the star topology and is more suitable for large-scale sensor networks. As a result, the cluster-tree topology is attracting increasingly more attention recently, e.g.,[1]- [3].