The foundation of life phenomenon is the abilities of representation, memory and behavior of a life form originated from complex motion and interaction among a great number of microscopic matters. Human natural language can describe and interpret this matter world, and also support naturally the subjective initiative of humans. Not as a computer system, the brain neurons can link more effectively with each other in 3D physical space such as in which no word is stored repeatedly, and the neurons form the matter base of consciousness activity. It is known that the radial basis function representation principle is widely used to represent color, sound, smell and gustation, this paper attempts to use a quantitative analysis method to estimate with an unbiased way the number of single wavelength color attributes that can be represented by our brain. It is also known that some simple neurons in the primary cortex are sensitive to the direction of short line, this paper introduces the representation principle of single basis inclination which can represent the inclination attribute of shorter line, and our brain can further extract more topology attributes of a shape through introducing the concept of vision event which is the evolution foundation of cognition ability. By the mechanism of semantics expansion our brain can construct more complex semantics. This paper emphasizes that both the semantics in natural language and the original semantics only perceived by our brain exist in the neural fiber plexus nodes of brain, they are twins. We humans always attempt to consider this world as a whole, just as at a certain moment the neurons activated by the brain attention form an expanded semantics, and the shifting of attention forms a consciousness stream, so the semantics is furtherly expanded, and especially it is our behavior to have shaped the complex consciousness stream structure. This paper also elucidates the principle of matter primacy, and through analyzing the development process of human intelligence, we have found that the subjective initiative in human behavior makes us distinguishing from animals. In this paper, we also explain that why the brain consumes less power and the limitations of human intelligence.  Here, following the essence of intelligence, through developing a kind of computer language and the corresponding language parser(“analysisformer”) to support a Digital-Twin system, we can dynamically establish a large vector database with practical training and simulation, then through these numerous examples of behavior and reinforcement learning method we can map the intelligence principle into an intelligence automation system such as a humanoid robot and some kind of dynamic system that requires real-time optimization.

In nature, the essence of intelligence phenomenon is the abilities of representation, memory and behavior of life form originated from the motion and interaction among a great number of microscopic matters. Using semantics in natural language can describe and interpret this physical world in various scenarios. Specially the brain neurons can link more effectively each other in 3D physical space than a computer, such as our brain doesn't save certain word in multiple places, then semantics is with the form of hierarchy structure, this also forms the matter base of consciousness activity. We emphasize semantics that can emerge and be expanded along with the time-axis exist in the neural fiber plexus nodes of brain. Therefore, following a special evolution stage of intelligence, through the reinforcement learning method based on the hierarchy structure and the time-sequence segments in semantics we can construct an intelligence automation system such as nuclear-powered submarine, and especially one humanoid robot can naturally embed into the semantics structural pattern. In addition, Appendix-A introduces my paper-the generation of semantics in natural language and the formation of brain, https://www.techrxiv.org, it gives us much deep understanding to the essence of semantics.

The human natural language is the operation system of our consciousness activity, and before it appeared we human interacted with the nature and each other typically through the image and sound, of course human beings also are the part of nature. The use of tools catalyzes the generation of natural language, in which the semantics space established in the image and sound of nature has been expanded greatly. The brain consciousness behavior depends on the attention mechanism which can extract the specified semantics, and this paper introduces the brain’s general stimulation-response mechanism that can decompose the semantics in natural language. Based on this kind of mechanism rather than computing-power and the current computer processing ability, it is sure that we are able to better simulate the brain intelligence.

We have known that in our eyes and ears, to the external stimulations there exist many cells that are respectively most sensitive to some specific center frequencies, and this paper has  discussed that the neuron system utilizes the transient response zone to represent different stimulation intensity levels, therefore we can generally use the Gaussian radial basis function model to explain that the neuron system is how to analyze the sound and color in the associated primary layer, and also can better explain some phenomena in the color cognitive behaviors. In addition, we have known that in the cerebral cortex, many small neuron groups are respectively sensitive to the different arrangement directions of line segments, and we find that the neuron system can further  extract the high-order information of a curve-line segment at the corresponding position, then as introduced later, the vison neuron system can conveniently represent a curve-line segment with an attribute-set constructed by using the polynomial radial basis function model related to the Taylor series expansion approach.

For a long-distance guard radar, MTI filter usually employs low staggered pulse repetition frequencies (PRF) modulation technique to overcome blind speed and detect the targets at low altitude in the presence of clutter. For a given ratio of the maximum-to-the minimum period, we find that there exists an optimum stagger ratios construction criterion (OSRCC) which can guarantee the first null depth minimum, and the associated estimation formula containing variable    has been derived successfully. In addition, based on more understanding to the properties of staggered PRF, we introduce a new stagger ratios optimization method that is much simple, efficient and flexible. In brief, this paper has thoroughly deconstructed this important basic technique in radar system.