| Recently,the fast development of nano-technology and synthetic biology contributes the manufacture of biologically inspired nano-machine in labs.Some typical examples are artificial cells,DNA robots,etc.Nano-network,consisting of nano-machines,has potential prospect in application of medicine,environmental,industry,military and so on.The major application of nano-network is on the area of medicine,focusing on the in vivo scene.Correspondingly,nano-network is studied as a branch of body area network.A complex task is difficult to be completed by a single nano-machine due to the limit of size,power,etc.In this case,multiple nano-machines shall work together to finish the task,thereby the communication mechanism is indispensable.Traditional electromagneticwave communication is difficult to achieve a satisfying performance considering the complexity of in vivo environment and restrain of nano-machine manufacture.Based on the background,molecular communication is proposed as a new communication scheme substituting for traditional communication.Molecular communication is a type of transmission technic which uses biological medium as channel,particles(including molecules)as information carrier,and utilizes the rule of biological metabolism for transmission.Compared with traditional electromagnetic communication,molecular communication may benefit on eliminating interference which is caused by complex environment in vivo,facilitating the implant,recycle and networking of nano-machines,and solve the problem of acquiring energy,etc.Study of molecular communication still stays in theory stage.Its research topic could be classified by the layer concept of traditional communication networks.More efforts focus on the physical layer,including channel modelling,coding,modulation,detecting of signals,design of nano-machine,design of interface,performance analysis,etc.Less efforts focus on the high layer,including protocol of routing,switching,application,etc.Channel modelling is an important topic which has not been fully investigated.Channel models build the bridge between communication engineering and biology rules,that is the research basis of molecular communication.This thesis studies channel model of molecular communication with some important biological medium.Also,this thesis chooses or improves some actual technics which fit the biological rule in nature,in order to set up the models and study their relevant technics.Detailed contents of research in this thesis are listed as following: 1)diffusive channel based on fluid medium in molecular communication;2)conduction channel based on neural signaling in molecular communication;3)hybrid channel based on fluid medium and neural signaling in molecular communication;4)conduction channel based on calcium signaling in molecular communication.Fluid medium is a common substance which occupies the most volume of body,some typical cases are blood,tissue fluid,lymph,etc.Molecules usually diffuse in fluid,and their concentrations form gradient difference in time and spatial domain,which is available for modulation,coding and transmission of information.Similar with the fluid medium,cells are another type of basic units in vivo.One typical type of cells distributing in whole body is neuron,of which the signaling mechanism is the most efficient in nano-network.For most regular cells,calcium signaling is an attractive mechanism of transferring energy and information.The communication applications in the layer of cells benefit from the channel modelling and relevant technic of neural signaling and calcium signaling.In addition,interactions between different biological medium such as the interactions between blood vessel and neurons,support the round-trip transmission of information and energy,which is the research basis of hybrid channel and maintain the biological activity in vivo.First,this thesis studies the diffusive channel based on fluid medium in molecular communication.An end-to-end channel model is first built,the slotted mechanism and the On-off key modulation are adopt.During the transmission,the Fick's law is used to model the regulation gradient of molecular concentration.Based on the gradient of molecular concentration,the receiving probability distribution is deduced,and the bit error rate is given.A classified model of molecules is proposed,which divides molecules into three parts: signal molecules,inter-symbol interference molecules and noise molecules.In order to improve the reliability performance of communication,an adaptive threshold variation algorithm is further proposed based on sensitivity theory of nature cells.The simulation results indicate that the proposed algorithm improves the performance efficiently,meaning that it's easier to gain a better performance if nanomachines inherit the adaptive study ability of nature cells.Then,this thesis studies the conduction channel based on neural signaling in molecular communication.This thesis considers a double-level neural model connected by gap junction channels,including sender and receiver terminals.In the thesis,the model is set up via modularization,considering pre-synaptic response,gap junction channel transmission and post-synaptic response.The channel capacity is deduced based on the model.Besides,this thesis study the interference of neural signaling,including the inter-symbol interference of SISO channel and co-channel interference of MISO channel.The intensity of the two interferences is quantified with discretization of output signals.The simulation results indicate that different types of interference impact neural signaling in different degree.The study of interference provides a new thought to explain the behaviour of performance degradation and conflicting in neural signaling.After that,this thesis studies the hybrid channel based on fluid medium and neural signaling in molecular communication.Based on the biological molecular interaction between blood vessel and neurons,the properties of network across different biological medium are discussed similar with traditional heterogeneous network.An one-way single-threaded model is established from blood vessel flow medium to neurons.The transmission of signals and relaying of information are considered.The mutual information between input,relaying and output terminal is deduced with the tool of information theory.The simulation results indicate that the relaying terminal could work efficiently in the hybrid channel.The hybrid channel utilizes the advantages of different biological medium to realize the molecular communication in complex scene.Finally,this thesis studies the conduction channel based on calcium signaling in molecular communication.A model of calcium signaling in cellular network is set up.In the sender terminal,hormonal stimuli are used to generate calcium signals,which transmit to adjacent cells via gap junction channels.In the receiver terminal,a decoding method integrating time and frequency domain is proposed.This thesis further designs a switching mechanism of gap junction channels for calcium signaling,based on permeability difference of calcium ions for different types of gap junction channels.Three types of noises are studied about how they affect channel switching.The decision problem of noisy channel switching is solved by a machine learning algorithm.The simulation results indicate that the proposed channel switching mechanism enables the controllable directed transmission of signals.The switch functionality of cells promotes the progress of application of drug delivery and cell computing. |
PDF Full Text Request