Coded Cooperation To Exploit Temporal Redundancy In Wireless Sensor Networks

Latest developments in wireless communications and micro electronics have facilitated the development of low-cost, low-power, multifunctional sensor nodes that are small in size and have scarce resources at their disposal for acquiring and processing information. These tiny devices are capable of self organizing to form a network which is known as wireless sensor network. A sensor network is composed of a large number of sensor nodes which are densely deployed in close proximity to the phenomena under observation. The position of sensor nodes can not be predetermined or predefined. This fact permits random deployment of sensor network in inaccessible areas or disaster relief operations. Hence, it is necessary that sensor network protocols and algorithms must possess self-organizing capabilities so that the nodes could observe the phenomena and could report that to the sink. Sensor nodes are provided with an onboard processor to process their observation before sending raw data to the sink node for further processing. Unique deployment and the requirements of the wireless sensor network nodes are challenging the old signal processing techniques which were mainly intended for one to many transmissions contrary to the case of many to one transmission that sensor networks bring on the table. If each sensor node is assumed to be independently sending its measurements to a common central sink node, the amount of redundant information transmitted may become very significant. This shows the waste of scarce resources namely communication and power which is not acceptable in most sensor network applications. The correlation directly depends upon the physical properties of the phenomena which can be exploited at each sensor to compress the data before transmitting it to the receiver.Distributed joint source-channel coding technique is finding its application in the ever evolving world of telecommunication, especially in the sensor network paradigm. The unique characteristics of the sensor networks are giving birth to new signal processing techniques and shifting the processing burden from the transmitter to the receiver side. The nodes of sensor networks, when deployed over particular area for specific observations, often exhibit spatial as well as temporal correlation in their observation.Cooperative communication makes use of the spatial diversity inherently present in the multi-user environment by facilitating different users with diversified channel characteristics to cooperatively transmit each others messages. It looks attractive to use correlated source coding in conjunction with cooperative communication. This alliance of the two techniques culminated in comparatively less error contaminated recovery of the signal at the sink node. In this thesis we proposed a novel coding technique based on cooperative communication and distributed source coding. We harness the benefits available with both cooperative communication and distributed source coding technologies. We name our novel technique as "Relay assisted Slepian-Wolf cooperation". In chapter three we present our scheme of "Relay assisted Slepian-Wolf compression" and chalk out a frame work for the implementation of the scheme and explain how the scheme can be useful to transmit observations of sensor nodes to the sink node by utilizing the optimum power. We present the simulation result based on the frame work proposed in the scheme. Throughout our simulations, we use regular low density parity check (LDPC) codes. And apply symmetric Slepian-Wolf compression at the both relays. That means equal compression at both relays.In chapter four, we consider the real case scenario where channel conditions are not fixed and are varying. We explain how our proposed scheme can adopt to the changing conditions of wireless channel. We first give complete frame work for the implementation and then give simulation results which show how efficiently our proposed scheme can adopt the channel variations and can maintain the reliable communication between the sensor nodes and the sink node under varying channel conditions. Finally in chapter five, we analytically calculate outage probability of our scheme. We compare the analytical results with the simulated results. The comparison of the results achieved from analytical expressions and those of from simulation complemented each other.