Traveling wave solutions to a neural field model with oscillatory synaptic coupling types (Q2160602)

The article focuses on a simplified homogeneous neural field model with axonal transmission delay: \[ u_t+u=\alpha\int_RK(x-y)H(u(y,t-\frac{1}{c_0}|x-y|)-\theta)dy \] where $u=u(x,t)$ denotes the mean electric potential at position $x$ and time $t$, the kernel function $ K$ represents the coupling strength and type based on spatial positioning, the parameter $\alpha>0$ controls the global coupling strength, $\theta>0$ indicates the threshold of excitation to the network and $c_0$ is connected to the speed of action potentials in the network. The firing rate function is considered binary and is modeled by the Heaviside function $H$. Details regarding this model are to be found in the article of \textit{G. B. Ermentrout} and \textit{J. B. McLeod} [Proc. R. Soc. Edinb., Sect. A, Math. 123, No. 3, 461--478 (1993; Zbl 0797.35072)]. The aim here is to improve previous known results on the existence, uniqueness and stability of traveling front solutions for special kernel classes. In the second section three oscillatory kernel classes depending on whether $K$ satisfies some threshold conditions and speed wave conditions are introduced. Assuming the kernel function $K$ belongs to one of these classes and $0<2\theta<\alpha$, one proves the existence of a unique traveling wave front solution $u(x,t)=U(z)$, such that $U(0)=\theta$, $U'(0)>0$, $U(z)<\theta$ on $(-\infty,0)$ and $U(z)>\theta$ on $(0,\infty)$, for the transformed equation by $z=x+ \mu_0t$, $\mu_0\in (0,c_0)$ being the unique wave speed with some exponentially decaying limits. Further one indicates a sufficient condition for the spectral stability of $U$. The third section of the article is devoted to the proof of existence and uniqueness result. The steps are standard: compute a formal solution, prove the existence and uniqueness of the wave speed, prove that formal solution is a real solution. In the fourth section the spectral stability of solutions is examined. An example is shown in the fifth section. In the sixth section we find a short discussion on the connection between fronts and fast pulses arising in singularly perturbed integral differential equations. The seventh section is devoted to some conclusions.