Abstract
Compressive sensing (CS) is a new technology in digital signal processing capable of high-resolution capture of physical signals from few measurements, which promises impressive improvements in the field of wireless sensor networks (WSNs). In this work, we extensively investigate the effectiveness of compressive sensing (CS) when real COTSresource-constrained sensor nodes are used for compression, evaluating how the different parameters can affect the energy consumption and the lifetime of the device. Using data from a real dataset, we compare an implementation of CS using dense encoding matrices, where samples are gathered at a Nyquist rate, with the reconstruction of signals sampled at a sub-Nyquist rate. The quality of recovery is addressed, and several algorithms are used for reconstruction exploiting the intra- and inter-signal correlation structures. We finally define an optimal under-sampling ratio and reconstruction algorithm capable of achieving the best reconstruction at the minimum energy spent for the compression. The results are verified against a set of different kinds of sensors on several nodes used for environmental monitoring.
Highlights
The recent evolution of sensing devices and the availability of new solutions and techniques in the area of wireless sensor networks (WSNs) have increased the expectation of WSN applications
While it is common in the literature to find papers like the two aforementioned addressing the problem of analog Compressed sensing (CS) with a focus on the hardware called analog-to-information converters (AICs), other works investigate the problem from a higher system-level prospective when the samples are not discarded by the ADC architecture, but by the device performing the sensing
We have investigated the application of CS with real COTS hardware, and using an energy consumption model, we have evaluated the impact of different kinds of measurement matrices on the power consumption
Summary
The recent evolution of sensing devices and the availability of new solutions and techniques in the area of WSN have increased the expectation of WSN applications. When natural signals have a relatively low information content, as measured by the sparsity of their spectrum, the theory of CS suggests that randomized low-rate sampling may provide an efficient alternative to high-rate uniform sampling This technique is usually referred to as analog CS, and it is a novel strategy to sample and process sparse signals at a sub-Nyquist rate [11]. Our contribution is: (i) to establish a common energy framework in which a fair comparison can be made by modeling the nodes when real signals are considered for reconstruction and real resource-constrained hardware is used to perform the compression; (ii) to investigate the impact of CS parameters for compression on nodes’ lifetime; this was only partially discussed in [12]; (iii) to investigate if low-rate CS (CS with sub-Nyquist sampling) can be exploited to reconstruct environmental signals with good quality; and (iv) to propose design parameters for low-rate CS that are able to achieve a superior reconstruction quality with the minimum energy expenditure, so as to prolong the lifetime of the whole network.
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call