In this paper, two new structures of sparse cylindrical sensor network are proposed, which are 1-D nested cylindrical sensor network (1-D NCSN) and 2-D NCSN. Considering the development of Internet of Mission Critical Things, numerous of devices need to be interconnected and sharing information. Sparse structures of sensor networks will not only reduce the cost of deployment, but also decrease the quantity of data. The reason of choosing cylindrical sensor network is that cylindrical array has been widely used in wireless communication and underwater target detection. According to the characteristic of cylindrical array, the 1-D and 2-D nested array is extended to build the 1-D and 2-D sparse cylindrical sensor networks, respectively. Comparing with the 1-D NCSN, the 2-D structure could save more elements, while it is hard to derive the beampattern expression of the 2-D NCSN. Simulation results show that the 1-D NCSN has a higher resolution than the uniform cylindrical sensor network (UCSN). In addition, through augmented matrix MUSIC approach, 2-D NCSN could detect all targets in the range, even if the number of targets is larger than the number of real elements in the network. However, both the UCSN and the 1-D NCSN could not find all the sources.