基于链路可靠性的路由协议（The routing protocol based on link reliability）.docVIP

基于链路可靠性的路由协议（The routing protocol based on link reliability） Link reliability model is built on the basis of the random walk model of shift. In the random moving model, the movement process node consists of a series of random length of time, the time period is called the movement cycle (movement epoch). In a period of time, speed and direction of motion of the nodes remain unchanged. Do not exercise attribute node is determined by three parameters: n, N and sigma lambda 2n deficiency. The following is the definition of the three parameters and related assumptions: The cycle length of Tin with parameters lambda n exponential distribution. The motion cycle of the node moving speed Vin is a random variable with mean variance sigma 2n n. The exercise period of node motion direction in; obey (0, 2 pi) in uniform distribution. The speed and direction of movement, and the movement of Vin from cycle length theta in is independent of each other. The nodes are not related, link failures are independent. For the two nodes, we can select one of the nodes as the reference point, with the relative displacement of another node to represent the motion relationship between the nodes, as shown in figure. The initial position of two nodes in the graph is M1 and N1, the distance between them is C1; after time Shakespeare, two node position is m2 and N2, the mutual distance is C2, random motion vector of them is Rm (T) and Rn (t). If the node n as a reference point, time node after t ns position is not relative to the node m3, node solution nodes relative to the knife relative motion vector for random: Rm, n (T) = Rm (T) - Rn (T), the relative velocity between two nodes can be expressed as: Figure 1 relative motion between two nodes If the node m and node n in a mobile speed relative time t0 V, at t moment, m node moves relative to the range of the node of the n can be used to express the circle of radius vt. According to the t0 node m and node n distance C (C = Req, Req for the maximum