انرژی کارآمد و انعطاف پذیر از دست دادن دوربین بی سیم شبکه های حسگر
Abstract: Data loss during transmission and a limited energy source are two main challenges that need to be dealt with in embedded sensor networks. These problems are even more severe in wireless camera sensor networks (WCSNs), owing to the large data size. Energy spent in idle event monitoring and communication, turn out to be the two biggest sources of energy consumption. An event-based sleep and wake-up mechanism is a suitable option for surveillance applications with long event arrival intervals. With proper use of different hardware and software functionalities an efficient event-based wake-up mechanism can be implemented. Compressive Sampling (CS) turns out to be an effective solution in reducing the transmission costs and also provides a loss resilient mechanism. It involves under-sampling the data through linear random projections which allows transmission of lesser bits than the original. The randomness in sampling makes the system tolerant to losses without requiring transmission of redundant parity bits. Both these characteristics help us on saving up on energy. The original signal can be recovered from this compressively sampled measurements using l1 optimization. However, using conventional CS on embedded WCSNs has some implementation related challenges. The processor memory and the recovery time of l1 optimization, are non-linear with respect to the data size and hence large image sizes may hinder the applicability of CS in practical cases. In this thesis, a framework for practical implementation of these energy saving strategies has been provided. Issues that affect the practical usability of CS, namely recovery time and memory usage have been discussed and the solutions have been provided, backed up by a number of experimental results. Significant improvements have been observed in the implemented schemes over traditional schemes in terms of recovery time. All the suggested schemes have been implemented on an actual Imote2 sensor node test-bed. This provides a platform for future research and testing of different aspects of WCSNs.
Keywords: Applied sciences, Compressed sampling, Energy efficiency, Erasure reslience, Wireless camera sensor networks