Research On Computational Power Of Spiking Neural P Systems With Polarizations Or Anti-Spikes

Biological computing is a research field in which information is encoded into biomolecules and processed by biotechnology.Membrane computing is one of the important research fields of biological computing,the purpose is to abstract computing models from the way living cells process and interact with substances.These models are known as membrane systems or P systems.Spiking neural P(SN P,for short)systems are inspired by the information processing and interaction between neurons.It is considered as a third-generation neural network and has a broad application prospect.SN P systems with polarizations and SN P systems with anti-spikes are two important variants of SN P systems.Three types of polarizations are used in SN P systems with polarizations,and there is a question of whether the number of polarizations can be optimized.SN P systems with anti-spikes has been proved to be Turing universal in the synchronous mode,but the computational power in the local synchronous mode is unknown.In the simulation software of SN P systems,the types of accepted rules are limited,and the simulation research of the more complicated SN P systems cannot be carried out.Considering these problems,researches on polarizations number optimization,computational power in local synchronization mode,and expanding SN P systems simulation software are carried out.The main research content and results are as follows:In order to optimize polarizations number in SN P systems with polarizations,the concept of anti-spikes is introduced into the systems,and a type of SN P systems with polarizations and anti-spikes is constructed.It is proved that the SN P systems with polarizations and anti-spikes are still Turing universal with only two kinds of polarizations.On this basis,a small universal SN P systems with polarizations and anti-spikes composed of 121 neurons are constructed.Compared with the small universal SN P systems with polarizations composed of 164 neurons known in the paper,43 neurons are reduced.Considering that the computational power of SN P systems with anti-spikes in the local synchronous mode has not been studied,the concept of synapses local synchronization is proposed based on the concepts of rules synchronization and local synchronization of neurons.The computational power of SN P systems with anti-spikes in the modes of rule local synchronization,neuron local synchronization,and synapse local synchronization are studied respectively.By simulating the register machine,it is proved that the computational power of SN P systems with anti-spikes in the three modes is equivalent to the Turing machine.Gutiérrez-Naranjo et al.developed a SN P systems simulation software that can accept seven forms of rules.In order to expand the types of accept rules in the simulation,an input processing program is written and it can recognize any types of rules.Then a simulation software based on Python is developed.In order to test the correctness,a SN P system is simulated using the program,and the results show that the program can correctly simulate the computing process.