The spin-wave nanoscale reconfigurable mesh and the labeling problem

Abstract

In this article, we present a nanoscale reconfigurable mesh which is interconnected by ferromagnetic spin-wave buses. In this architecture, unlike the traditional spin-based nano structures which transmit charge, waves are transmitted. As a result, the power consumption of the proposed modules can be low. This reconfigurable mesh, while requiring the same number of switches and buses as the standard reconfigurable mesh, is capable of simultaneously transmitting N waves on each of the spin-wave buses. Because of this highly parallel feature, very fast and fault-tolerant algorithms can be designed. To illustrate the superior performance of the proposed spin-wave reconfigurable mesh, we present three fast labeling algorithms.

Efficient parallel processing with spin-wave nanoarchitectures

“…The time complexity of the labeling algorithm is O(log N), as we presented in [39]. Convex hull of a figure can be found in O(log N) as well, as shown in the next section.…”

“…A labeling algorithm presented in [39] is an application example that uses this routine several times. Here, we explain the routine and then present a new image processing application, namely finding the convex hull of a figure that uses this routine.…”

Parallel Computing with Spin Waves

“…Since the maximum value of the inputs is N, this procedure takes O(log N) time [3]. Next, each node checks its most significant bit (MSB).…”

“…In that case, the switch in row 3 and column 6, represented as s (3,6) should be set to on. Similar to any standard crossbar, a spin-wave crossbar realizes arbitrary permutations.…”

Nanoscale Image Processing

“…

This chapter has two main sections. So, this architecture represents a new approach for designing nanometer-scale computational structures, thus aiming to preserve the main advantages of wave-based computers [21][22][23][24].In this section, we study different image processing algorithms on the spinwave reconfigurable mesh architecture [22]. More specifically, how three image processing problems, namely, labeling, finding the convex hull, and finding the nearest neighbor problem, can be solved on a spin-wave reconfigurable mesh.

…”