Degree Name

Master of Engineering - Research


School of Electrical, Computer and Telecommunications Engineering


For most applications, sensor nodes rely solely on batteries for their power supply and hence battery depletion may have a great impact on the overall network performance. Consequently, most of the current research work in wireless sensor networks (WSNs) is focusing on improving architectures and protocols with energy conservation as the main focus. Instead of using a competing approach such as used in traditional wireless networks, sensor nodes are proposed to cooperate to achieve a common goal. Multiple sensor nodes can be used to transmit and receive cooperatively and such a configuration is known as a cooperative Multiple-Input Multiple-Output (MIMO) system. Cooperative MIMO systems have been proved to reduce both transmission energy and latency in WSNs. However, most current work in WSNs considers only the energy cost for the data transmission component and neglects the energy component responsible for establishing a cooperative mechanism.

When the energy cost of cooperative mechanism establishment is considered, the total energy consumed as a function of the increasing number of cooperating nodes becomes of particular interest. In addition, most of the previous work in WSNs focused only on Space-Time Block Coding (STBC) schemes and ignored other potential MIMO schemes. In this thesis, both transmission and circuit energies for both components are included in the performance models. Three major cooperative MIMO systems, namely Beamforming (BF), Space-Time Block Coding (STBC) and Spatial Multiplexing (SM) are compared and analysed. WSNs are assumed to operate in quasi-static Rayleigh fading channels with M cooperating transmit nodes and N cooperating receive nodes. The cooperative BF scheme outperforms both the cooperative SM and STBC schemes in terms of energy efficiency and packet latency for both synchronous and asynchronous scenarios. Synchronous scenarios assume perfect synchronisation between cooperating nodes. Also, the cooperative BF outperforms cooperative SM with higher diversity gain. A cooperative BF scheme with two transmit nodes is suggested as the optimal energy efficient cooperative MIMO system with the lowest packet latency when operating below 0.4Tb clock jitter difference and below 800mW radiated power in imperfect synchronisation scenarios. Tb is the bit period which corresponds to the system bit rate.

In previous work all sensor nodes are assumed to be always on which could lead to a shorter lifetime due to energy wastage caused by idle listening and overhearing. Low duty cycle MAC protocols have been proposed to tackle this challenge in WSNs. However, most of the low duty cycle MAC protocols have been proposed for noncooperative systems. In this thesis, a new cooperative low duty cycle MAC protocol (CMAC) is proposed for two cooperative MIMO schemes: Beamforming (CMACBF) and Spatial Multiplexing (CMACSM). Performance of the proposed CMAC protocol is evaluated in terms of total energy consumption and packet latency for both synchronous and asynchronous scenarios. All the required energy components are taken into consideration in the system performance modelling and a periodic monitoring application model is used. The impact of the clock jitter, the check interval and the number of cooperative nodes on the total energy consumption and latency is investigated. The CMACBF protocol with two transmit nodes is suggested as the optimal scheme when operating at the 250 ms check interval with the clock jitter difference below 0.6Tb.