Compressed Sensing Based Data Gathering In Wireless Communication Networks

With the development and expansion of the state-of-the-art wireless communication networks, the traditional transform coding, in which the signal is first sampled and then compressed, faces serious challenges in dealing with huge amounts of data. For example, the sampling rate required is too high, and moreover, it is also a waste of resources in the following process of compression. Compressed sensing (CS), as a novel theory which exploits the sparsity characteristic of the original signal in signal processing and coding, can jointly execute sampling and compression at a rate much lower than the Nyquist rate. Finally, the original signal can be recovered. CS has been well used in the areas of image compression and signal processing in the past few years. Recently, CS has been attracting ever-increasing interests in the area of wireless communication networks.This thesis first introduces the theory of CS in the following three aspects:the method of spasifying the original signal, the incoherence between the measurement matrix and the transform basis, and the reconstruction algorithms. Then, the current researches on the application of CS in wireless communication networks are described. These techniques are assorted according to the general OSI (Interconnection Reference Model) network model, showing that the application of CS has brought revolutionary changes in signal detection, channel estimation, data gathering, network monitoring and other fields. In addition, the relevant drawbacks are also analyzed.The applications of traditional CS in data gathering in wireless communication network are limited by the huge transport overhead cost caused by dense measurement. To solve this problem, we take advantage of CS approach based on sparse random projection, where the projections are decided by the routing schemes. In this work, we propose CS driven random routing schemes as well as ameliorated random routing methods that are executed with sparse measurement based CS for efficient data gathering, corresponding to different networking topologies in typical wireless communication networking environment, and then analyze the relevant performances compared with those of the existing data gathering schemes, obtaining the conclusion that the proposed schemes are effective in signal reconstruction and efficient in reducing energy consumption cost during multi-hop routing.In the end, this thesis summarizes the general ways of utilizing CS in wireless communication networks, as well as the potential research challenges in the future.