Interconnections between these satellites using inter-satellites links ISL are organized to provide a significant flexibility for satellite communication networks to reconfigure in case of faults without losing data [15, 19, 20].
As a result, using ISL in satellite networks will increase the effectiveness of satellite communication, providing higher quality of services. Here, we present integrated networked reconfiguration schemes for a network of satellites interconnected via ISL. Such clusters consist of a number of satellites located around a nominal position in one or multiple orbits GEO, LEO, etc. The satellites are connected via RF Radio Frequency links and through a ground station. The satellites are also interconnected via ISL for reconfiguration purposes in case of faults.
Basically, in a network of interacting satellites the system is divided into two segments: space segment and the earth segment. The space segment consists of one or more satellites in particular orbits. Generally, the satellites in a network are located in the same orbit. In some particular networks, the satellites are located in different orbits which are referred to as multitier satellite networks. The operating orbit of satellites and their characteristics depend on the applications and services provided in the networks.
As discussed in the previous section, satellites are controlled and monitored via Telemetry Tracking and Command TTC subsystems. TTC subsystems include earth segment interfaces as well. Ground Station 7 Figure 1. Operation control center is needed to provide feedback information to the satellite network for resource management and other required control functions. There are three major types of connectivity among satellites which are point to point, point to multi-point also called broadcasting and multi-point to point .
Adding a transmission technology to each satellite, convert the broadcast system into a multipoint to multipoint one, by which each satellite in the network interacts as both sender and receiver . Frequency band is a specific range of frequencies while spectrum is a full range of all the frequencies from zero to infinity. The frequencies between 1 and 4 GHz are the optimum range of the spectrum for space-earth transmissions. A significant amount of atmospheric signal absorption occurs above 12 GHz.
At 22 GHz and 60 GHz, there are very high signal absorption due to resonance by water humidity and oxygen, respectively. Therefore, these frequencies can only be used for interlinked satellites communications . For interconnection between the satellites and the ground station or among the satellites interacting in a network, radio links are used. Satellite communications depend on the availability of radio spectrum which is limited.
In Figure 1. On the basis of the communication range of frequencies, different types of information may be sent via the network in data, voice and multimedia formats. Below 1 GHz, 9 only data packets may be sent. Between 1 GHz and 7 GHz, voice packets are transferred. Above 7 GHz, data of multimedia nature are transferrable. The data GHz 0. According to TCP principles, the sender sends a packet and waits until it receives an acknowledgement from the receiver to indicate that the packet has arrived correctly . The sender keeps a copy of the sent data with its time stamp and if the related acknowledgement receipt times out, a lost packet will be detected.
The lost packet will be resent. This is called a congestion avoidance which is identified by using TCP over satellites [27, 28]. For modeling of the network of interacting satellites in Chapter 4, TCP algorithm is used as the functional protocol in the network. Therefore, control issues in a network of interacting satellites are very important due to the operational and environmental variation they encounter in the space. The constant growth in the global communications and the need to replace older and out of commission satellites, call for design and launch of newer satellites.
As a result, there is a growing trend toward employing networks of interacting low cost small satellites. With networks of satellites, communication can be more reliably maintained [7, 8]. The designed controller will depend on the knowledge about the plant and the available data. When a control system is shared via a communication medium within several nodes beyond the operating system, then a networked control system NCS is obtained.
Exchanging data, including input, output and control parameters, among the system components through the network is the key feature of an NCS . Using the control over a network reduces the system complexity as well as allows the data to be shared efficiently. Furthermore, intelligent decisions can be made over a large physical environment such as the space where the cooperating satellites are deployed. In this study, an integrated supervisory fault tolerant control and networked reconfiguration scheme is developed to simulate the communication interactions over multiple satellites.
In this method, the reconfiguration procedure refers to the retransmission of the data related to the faulty satellite by distributing and sending it via other satellites operating in the network. Orbit transferring, redundant satellite substitution and repositioning the healthy satellites are other reconfiguration schemes which are developed and simulated in this study. Depending on the operating conditions and the size of the affected data, different reconfiguration topologies are used.
This can be done by applying supervisory control using high-level petri nets. Network performance is then evaluated using this method [30, 31]. In case of a fault in the network, the number of states will change depending on the loss conditions. To prevent these loss conditions due to faults, a networked control model of a multi-satellite communication system is developed using high-level petri nets . These growing operational constraints deserve attention in the form of monitoring, fault diagnosis and reconfigurable control .
To this extent many investigations have been conducted on satellite fault diagnosis and control. James Albus and colleagues  focused on Collaborative Tactical Behaviors for Autonomous Ground and Air Vehicles by developing a four dimensional real time control system. Their methodology resulted in detailed design requirements for perception, knowledge representation, decision making, and behavior generation processes that enable complex methods to be planned and executed by unmanned ground and air vehicles working in collaboration with manned systems.
Feng et al developed a dynamic simulation to evaluate the location update 13 efficiency in LEO satellite networks. Only LEO satellite communication networks are proposed in their modeling. Angeli developed an artificial intelligent fault diagnosis method for on-line systems by applying numerical techniques. In this method fault detection is performed by comparing the predicted behavior of a system based on qualitative models with the actual observation .
The proposed system uses computational intelligence CI to detect and isolate faults and also to infer cause of failures from the telemetry data time series history using functional models of satellite ACS. Huang et al. They investigated how to set rules for a satellite queuing system so that all the GEO satellites as users have self-interest in controlling congestion when it occurs.
Xing and colleague attempted to design and simulate an autonomous control system for satellites . In their research a satellite autonomous orbital control system was designed and the semi-major axis control, the eccentricity and the inclination angle were discussed.
Powel and Morgansen  derived performance limits for a group of autonomous space vehicles obeying a nonlinear motion control model. They performed Monte—Carlo simulations with random initial headings to analyze the communication energy required for the convergence of the discrete-time system. The results 14 implied a strategy for designing minimum communication energy algorithms for heading alignment and coordination for vehicles in which communication is energetically expensive.
Casbeer and Holsapple  investigated a column generation technique as a distributed method for solving task assignment problems with precedence constraints. With a careful division of the overall problem into small local problems, the column generation approach iteratively solves the sub-problems in a distributed way to reach an overall optimum. Owing to the complexity introduced by the precedence constraints, they conclude that it is unlikely that column generation alone would be practical for distributed task assignment but could perhaps be used in conjunction with other methods or a hierarchical design to allow sub-groups to solve smaller sized problems.
Lee and Kim developed a fault tolerant control scheme for satellite attitude control system . In their research a sliding mode control scheme with finite reaching time is proposed for a fault-tolerant satellite attitude control system in the presence of actuator faults and external disturbances. The actuator fault is modeled to reflect the degradation of the actuation effectiveness, and the solar array induced disturbance is considered as external disturbance.
The control scheme is designed to perform the rest-to-rest maneuver of a satellite system with the degradation of the actuation effectiveness, and the stability analysis is performed using the Lyapunov theorem. Also numerical simulations are conducted and the results are compared to verify the performance of the proposed fault-tolerant control scheme. Neural network control methods are widely used by many researchers for different purposes of fault diagnosis and control [16, ]. Li proposed a neural network based fault tolerant controller for mobile robots . The effectiveness of his proposed method is illustrated by performing the simulation of a circular trajectory tracking control.
Semsar-Kazerooni and Khorasani investigated the optimal consensus algorithms for cooperating team of agents subject to partial information . The objectives of their work were the design and development of controllers for a team of agents that accomplish consensus for agents' output in both leaderless and modified leader-follower architectures. According to the literature, much of the research work on fault detection and control methods of satellite networks has been conducted by assuming only computer type of problem or a limited mechanical failure, but not for a combination of them.
Also intelligent fault prognosis and control recovery scheme have not been investigated for a network of cooperating satellites [16, 17, 49]. Furthermore; the monitoring and controlling of non-measurable faults hidden faults is still an open research field . Furthermore, some networks of cooperating satellites are used for multi-purpose missions. Also, despite the time-invariant attributes of air-borne networks, the topology of satellites networks can vary with time. Satellite networks have a vast range of applications in many key industries and as such must be monitored, controlled and managed, in order to make such a complex system operate efficiently and securely.
Actually, the system must have the ability to adapt itself to the required changes of an application in the presence of fault or emergency cases, and reconfigure accordingly . The proposed research specifically intends to analysis and development of: 1 a fault detection and reliability analysis strategy for the network; 2 a supervisory control system to monitor the behavior of the network in order to meet the performance criteria; and 3 networked reconfiguration schemes without disturbing the stability of the system.
This research is carried out to develop discrete event models using high-level petri nets. A fault tolerant reconfiguration approach is used to formulate such solutions with SPNs. These methods help to reconfigure the network to full functional performance conditions. Previous studies have been on individual satellites to detect and control their faults and reconfiguration issues have not been addressed. The developed methodology is based on Stochastic Petri Nets. Chapter 2 also presents the basic supervisory controller to determine the uncontrollable and unobservable state transitions.
Synthesizing a controller with such transitions in the network is then explained and the deadlock conditions discussed. To be more specific, Chapter 2 develops a fault diagnosis and supervisory control system of multi-satellite interactions using SPNs. Vulnerability, uncertainty and probability of the presence of a fault in the network are then assessed using SPN models. To enable the system to operate in fault tolerant conditions, a set of reconfiguration models in a faulty network are developed using CPNs, in Chapter 3, Chapter 4 and Chapter 5 to meet the objectives 2 and 3 of Section 1.
Also the required performance parameters in a satellite network are developed in these chapters to evaluate network functions. Chapter 3 is devoted to development of fault tolerant reconfiguration schemes for networks subject to partial failure that meets objective 3. The proposed reconfiguration method enables the faulty network to operate in full performance condition.
A stability analysis of the faultless system is shown in this chapter. The reconfiguration method is verified and investigated by measuring the performance parameters in terms of mean delay and throughput. The results are compared to that of an analytical method in the absence of faults. The fault tolerant control method of Chapter 4 is further developed in Chapter 5 to reconfigure the networks subject to general types of failures and accomplish objective 3.
The sensitivity of the performance measures to network input parameters is determined in the absence and presence of faults in the networks. The results are then discussed and compared. Chapter 6 concludes and remarks on the findings and proposes ideas for future work to improve the implementation and further refine of the developed techniques. Supervisors are used to ensure that the behavior of the network, which needs to be controlled and reconfigured, does not violate a set of required operational conditions. The supervisory control and reconfiguration actions are based on system observations.
International journal of stochastic modelling and applications
The methodology developed here is divided into two stages. In the first stage a network of interacting satellites is modeled using Stochastic Petri Nets SPNs and controllability and observability characteristics of the network are analyzed. Then in the second stage, three major indicators are developed to quantify the vulnerability, uncertainty and failure probability characteristics of the system. The results of the proposed methodology are then presented. Some references best explain these references are . More recent applications are reported in .
One method to model complex systems such as multi-satellite interactions is to use petri nets. Petri nets have a simple mathematical representation by employing linear matrix algebra making them particularly useful for analysis and design of discrete event systems .
Petri nets are divided into low and high-level forms [60, 61]. The advantage of using high-level petri nets in a satellite network is its both state and action oriented nature which describes the states of the system and the transitions events which cause changes in the states. Therefore, for a complex system such as a multi-satellite interaction, the networked control performance can be modeled, using much of the existing real conditions.
A petri net is a multi-graph, since it allows multiple arcs from one node of the graph to another. Since the nodes of the graph can be partitioned into two sets places and transitions , such that each arc is directed from an element of one set to an element of the other set, it is a bipartite directed multi-graph. As shown in Figure 2. A marking is an assignment of tokens dots to the places of a petri net.
A transition can be caused, i. Figure 2. In a petri net model, the conditions are modeled as places and events as transitions. The occurrence of an event in the physical plant is modeled by the triggering firing of a transition . Deadlock : A deadlock in petri net is a transition which cannot be fired. A transition is live at level i, if every transition is live at this level.
Reachability tree  represents the infinite number of markings which result from loops. Frontier nodes are those which have not yet been processed. The root of the reachability tree is set equal to the initial marking as a frontier node. As long as frontier nodes remain, they are processed by the following algorithm: Let x be the frontier node to be processed. Node x is redefined as an interior node; node z becomes a frontier node.
Each matrix is m rows one for each transition by n columns one for each place. The system incidence matrix and the reached marking after firing the enabled transitions are shown respectively in Equations 2. SPN as a simulation method is an attractive graphically-oriented modeling framework well-suited to path generation on computer [63, 64]. System states change when events occur. Stochastic changes occur at random times. By modeling a stochastic process, X t is assumed as the state of the system at time t which is a random variable.
The network of interacting satellites is considered as a discrete event stochastic system by defining appropriate system states . To model the supervisory controllers using petri nets, controllability and observability specifications have to be checked to prevent forbidden connections in the model . Many researchers have used petri nets as a tool for modeling and synthesizing the control laws for different types of discrete event systems [13, 60, 61, 66, 67].
The inequality constraint in relation 2. According to this definition, an uncontrollable transition can be observed. But the inverse case is not applicable which means that the arcs cannot be connected from controller places to the uncontrollable transitions. This requirement is checked in petri net models using Equation 2.
There cannot be any connection between the controller places and the 26 unobservable transitions. Therefore, the unobservable transitions are also uncontrollable. This requirement is checked using Equation 2. Therefore, by checking these requirements the PN model would be applicable in the real time systems for on-line reconfiguration and handling the uncontrollability and unobservability conditions.
Here the definition of each term is provided: Vulnerability: Is defined as a quantity which indicates how many times an event has occurred within a time interval . Uncertainty: Is defined as the number of failed components at time t . Here we have developed formulas to calculate these indicators using SPN models. Vulnerability: Is defined as a quantity which indicates how many times an event modeled by a PN transition, has occurred in that interval.
Firing rate is a probabilistic delay after which the transitions are fired and is determined by a random variable. Equation 2. Probability: By means of logical or algebraic functions of the number of tokens in the PN places, an output condition can be specified, for instance the number of tokens in the defined failed condition place. The simulation details and quantification results are presented in Section 2. It contains 10 places and 9 transitions.
Each of the places P1, P5 and P8 show the independent operational conditions of each satellite in its orbit. By firing transitions T1, T4 and T7, each satellite sends an access S1 S2 S3 GS 30 request to the ground station at known time stamps which may result in waiting condition of the satellites to access the ground station shown by places P2, P6 and P9. Place P4 shows the ground station availability where its buffer size is considered to be equal to the number of initial tokens in it. In this model place P4 is marked with two tokens, which means that the ground station is able to communicate with only two satellites at the same time.
These places denote the condition when each satellite is communicating with the ground station. In the satellite network of Figure 2. So, transition T2 is uncontrollable. All the effective parameters influencing the network status can be modeled through SPN concepts and autonomously managed. In Section 2. In the following sections of this chapter, an SPN model is developed to measure the three indicators in a network of satellites.
As explained for the networked satellites modeled in Section 2. The model is shown in Figure 2. The stochastic properties of the satellite network are determined by allocating a firing rate to transition T1. Failure Modeling- Taking into account the failure and repair of each satellite, the network operation is modeled by the SPN of Figure 2. In Table 2. T2 and T9 are assumed as immediate transitions uncontrollable. P2 Satellite waiting for access to GS. P3 Satellite interacting with GS. P4 GS free. P5 GS failed. P6 Satellite failed. After synthesizing the developed petri net model to see whether it meets the system requirements according to the controllability rules, the observability conditions are checked as well using Equation 2.
If any obstructions appear in a part of the identified transitions T1 to T11, the transitions become observable. Therefore, on the basis of Equation 2. With the initial marking M1 shown in Figure 2. By examining the data reported in Table 2. According to Table 2. Substituting the numerical values of the firing rates given in Table 2. According to the Probability definition in Section 2. The Vulnerability in the time interval of [0, The stay time in each state and the return time to each state, are given in Figure 2. According to the Vulnerability definition in Section 2.
For the specified SPN of Figure 2. For the given time interval, marking no. Referring to Figure 2. According to Figure 2. With reference to the Uncertainty definition in Section 2. In Figure 2. It shows that state 20 is the most probable event which may happen in the system. So the least reliable section at this time step is the ground station which failure is the most probable event. Table 2. A petri net supervisory controller is then designed for the networked satellites. Reliability indicators are also integrated in terms of vulnerability, uncertainty and probability in the network using SPN models.
The reliability of the developed supervisory control and fault detection model is assessed for a network of three satellites using SPNs. The information provided in Section 2. The faults and their occurrences lead us to vulnerability, uncertainty and probability criteria which indicate the performance of the system. This procedure addresses objectives 1 and 2.
- Christmas Memories.
- Stochastic journal;
- Network Theory (Part 2);
- Online available;
Partial failure addresses the faults in part of the satellite communication system. Although losing the connection with some components in partial failure, it is still able to communicate with the other ones.
International journal of stochastic modelling and applications
Full failure refers to the condition when a component in the network satellite is fully failed to operate. Using the physical specifications of space objects, a procedure is developed in 3. A Colored Petri Net approach is used to analyze the system . A CPN represents a graphical language for modeling and simulation of concurrent and non-deterministic systems and analyzing their performance and properties.
Figure 3. Colored Petri Nets allow tokens to have a data value assigned to them. The assignment is called a token color. The color can be assigned arbitrarily to the places in CPNs in a way that each place contains tokens of one type. This type is called color set of the place. To distinguish the same data packet contents from one another, token colors with different time stamps are used. By performing simulation of the Colored Petri Net models, it is possible to describe different states and investigate the behavior of the system .
Therefore, for a complex system such as a multi-satellite network, the networked control performance can be simulated using much of the real conditions. Within a network, the satellites and their related payloads are the only non-terrestrial space segment elements of the system. The non-terrestrial state of the satellite communication networks means that the repair of a failed space segment of the satellites or their subsystems is not generally possible. There are several reconfiguration scenarios to enhance availability of the network when one of the interacting satellites or its corresponding transponder fails.
Using these reconfiguration procedures ensure that the network remain available following a failure event. Chapter 3, Chapter 4 and Chapter 5 explain the concepts we have developed to reconfigure network of interacting satellites to a fault tolerant mode. The proposed reconfiguration simulation results are presented at the end of each chapter. Time is divided into slots of duration equal to the packet transmission time.
Under normal faultless condition, each satellite communicates through a space network transmission protocol with the ground station. According to the applied communication protocol, each satellite will know whether its transmitted packet has been successfully received by the ground station, if it receives an acknowledgement. Packets sent to satellite i may be rejected due to three main reasons: - Full buffer of the satellite which may result in retransmission after a delay; and satellite-ground station connection faults which result in retransmission of the data through other satellites, for which reconfigurations are modeled in this chapter.
This in fact emulates the arrival of external data packets to the network, NCL- Network Confidence Level which defines the efficiency of the network transmission, Although the recovery procedure of failed satellites in a network is very costly and time consuming, the proposed reconfiguration models prevent the cluster from failing, and reconfigure the network to an acceptable level. This includes a time delay in the receiving and sending functions of the faulty satellite. The proposed reconfiguration procedure provides the possibility of uninterrupted services in the network with some additional delay time.
The related simulation results are presented at the end of this chapter. With the proposed reconfiguration model, there are relatively a few places involved, although they are repeated many times through the collection of sub-pages which constitute the specifications. A fusion place means that its multiple appearances are to be treated as if there were only one place. The next phase in the simulation is to add transitions, whose occurrences, as explained in Section 2. These events are identified at a level of abstraction that corresponds to an informal description.
The arcs and transition conditions java inscriptions 50 defined for a transition in CPN tools , are then added to relate the events and the states conditions which represent objects or data. To satisfy the required desirable conditions, variables are defined so that transitions can be referred to the initial conditions. The places and transitions are defined to be easily related to the reconfiguration specifications.
They are given suggestive names to indicate their brief functional description. To retain the defined places and transitions corresponding to some obvious notion of states or events, which are the graphical aspects of CPN, complex text inscriptions have been used. This is useful for following the graphical elements identified in the CPN model, because the number of graphical objects increases exponentially.
The idea of hierarchical modeling is also used with substitution transitions. A hierarchical CPN model allows representing a complex simulation through a simplified net that gives a broad overview of the system . So, in the process of CPN modeling a coarse-grained structure for the CPN is used by substituting the double line box transitions in a top level with more pages which bring details into the model.
For example all the transitions shown in Figure 3. Using the CPN hierarchical techniques, a model of the network is developed. The first module consists of satellite sub-systems and their interconnections and is different for each reconfiguration procedure presented in this and the next two chapters. So this module is 52 explained separately for each reconfiguration scenario investigated in Chapter 3, Chapter 4 and Chapter 5. The last two modules are the same for all reconfiguration schemes.
Therefore they are presented in this section. As it was discussed in section 1. Data packets are sent from each satellite via the network to the ground station. Also receipt of acknowledgements occurs only when there are identical acknowledgments from the receiver. The transmission network performance is also considered by modeling the loss of data packets in the system. The loss of acknowledgements is modeled in a similar way. S a te llite sS a te llite sS a te llite s. Network Transmission Capacity Data Packet Transmission Module Restricts the number of data packets which are allowed to pass through the network at once.
Satellites Ground Station Module List of all available satellites cooperating in the network. Receive Confirm Data Packet Transmission Module Satellites receive the related acknowledgements from the ground station through the network. Remove Data Data Packet Transmission Module Satellite removes the identical data from the buffer after receiving its acknowledgement. Initiation Ground Station Module Initiating list of all available satellites cooperating in the network.
The satellites CPN modules are developed in Chapter 3, Chapter 4 and Chapter 5 for full failures, referring to the type of reconfiguration method which is different in each chapter. Table 3. The first module is the satellite composition system consisting of data initiation and updating, fault detection and reconfiguration sub-modules which define the reconfiguration protocol for the network. The second module is the ground station consisting of the satellites availability information and data type processing sub-modules. The data packet transmission is the third module that determines the packet transmissions and run the acknowledgement protocol.
The entire model is explained in the next section.
PDF Stochastic Petri Nets: Modelling Stability Simulation (Springer Series in Operations Research
In this figure, the identical module of operation for each satellite in the network is shown. As it is shown in Figure 3. Each satellite receives and transmits data packets in the size of an allocated buffer which are a channelized share of the total network capacity. This is for the stable operation when the satellites communicate through the ground station in the absence of any fault. It reconfigures the network to attain performance at an acceptable level and avoid loss of data. The fault detection sub-module is shown in red color in Figure 3. The procedure sends the identified data from the faulty satellite to the other satellites via interconnection links to the ground station.
This process continues until the failed satellite is corrected and returns to the normal operation.
If the failed satellite has been damaged beyond repair depending on the failure types and available solutions , it will no longer return to the network and in a long run it may be replaced by a new satellite. Therefore, the network will continue its operation using the reconfiguration mode as defined. The data, to be retransmitted, is distributed between the other satellites according to their buffer capacity.
The reconfiguration sub-model is shown in blue color in Figure 3. Substitution of modules shown in Figure 3. On the basis of the integrated model, each satellite in the network is substituted with the major satellite composition module shown in Figure 3. The number of tokens in this place is identical to the number of data packets which are allowed to be transmitted via the network per round trip time.
Therefore, dividing the number of tokens by the total round trip time of the packets in the network determines the network transmission rate. Fault Detection Satellite Composition System Module Detecting faulty satellite which cannot connect to the ground station. Data Packets Generate Satellite Composition System Module Generating and updating data packet contents of faulty satellite to be sent to the other satellites. COMM Satellite Composition System Module Generating a list of all available satellites in the network and the ID and content of the messages has not been sent due to the failed connection.
Send Data Packets Satellite Composition System Module Sending data packets stored on-board of faulty satellite to the other satellites. It also makes it possible to compare reconfiguration options as discussed in Section 3. According to , the analytical method to calculate throughput is simplified in Equation 3.
NCL is the network confidence level as defined in Section 3. Network delay effects are challenging control problems in the systems such as multi satellite networks. The time to read a component measurement and to send the related data to a receiver through the network depends on network characteristics such as the topology and routing schemes. The delay is intensified when a data loss occurs during a transmission. Delays not only degrade the performance of a network-based control, but they can also destabilize the system. The delay time decreases with the increase in the buffer size.
A significant improvement is seen when the buffer size is increased from two to four and only a marginal improvement is achieved when the buffer size is increased from four to six. By increasing the buffer size, throughput increases as well. The packet generation rate is considered to be constant and equal to 0. Here, for the condition of this simulation, it occurs after time slots , for all the three buffer sizes. In Figure 3. As the number of satellites increases in a network, an improvement in throughput is observed.
Therefore, from the results obtained for the faultless scenario, it is concluded that for a required performance in terms of throughput and delay, a minimal configuration can be considered in terms of the number of satellites in the network, the buffer size on-board the satellites, and the network transmission capacity. Simulation output data exhibit stochastic behavior, and therefore appropriate statistical analysis must be used both to design and to interpret simulation experiments .
In this section, the performance of the reconfiguration procedure discussed in 3. First, in Section 3. A network of n satellites communicating with each other and a ground station is considered Figure 3. A number of additional experiments were conducted for different system parameters which also exhibited similar behaviors.
The effectiveness of the reconfiguration protocol and the level of performance degradation are then evaluated. To investigate the effects of satellite failures, first, a network of three satellites is considered with a buffer size of four, transmission network capacity of 0. The network performance is studied in terms of throughput and transmission delay when one of the satellites partially fails to connect to the ground station. The results are then compared to the faultless condition. It is very important to note that no data is lost, but the quality of service drops significantly.
The results show that by applying the proposed reconfiguration procedure, This percentage is almost the same slightly different for the lowest and the highest buffer sizes failures. No data is lost due to these failures. It is concluded that if one satellite fails, the network with higher number of satellites respond better to reconfiguration. If n-1 satellites fail, networks with smaller number of satellites have a better response to the reconfiguration. This happens because of the smaller throughput distribution in larger networks. So in a larger network, if only one component fails, a smaller section of data is missed and redistributed, comparing to the smaller networks.
Examination of the plots in Figure 3. As seen, the throughput is not very sensitive to packet arrival rate, but it highly depends on the number of failed satellites, network transmission capacity and on-board buffer size. Systematic intelligent networked control schemes were simulated to reconfigure the network to normal operational conditions.
The faultless condition scheme was verified through a CPN simulation analysis which results in stable performance. The performance of the proposed reconfiguration protocol is assessed in terms of mean delay time and throughput for various specifications and conditions. These performance measures are obtained as a function of buffer sizes on-board the satellites, network transmission In the performance analyses conducted in Section 3.
CPN provides means of detailed modelling and produces simulation output that satellite network operators can use to design the network and appropriate reconfiguration protocols depending on the intended mission of the satellite network. A methodology is developed in Section 4. The satellite networks to be reconfigured are considered to include a backup satellite or to have specific topologies. To analyze the system 79 Colored Petri Net models are constructed . Simulations are then conducted and the results reported and discussed. To explain this we consider a satellite in a circular orbit at an altitude h.
The linear speed of the satellite is related to its orbital altitude and is given by Equation 4. Instead, its orbit becomes elliptical in the same plane . Figure 4. With calculated speed changes one circular orbit can be changed to another in two steps. Increasing the altitude of a satellite helps to cover a larger area of the earth. This concept is also used in Section 5. The process is completely reversible if the steps are carried out in the reverse order.
In Equations 4. Changing the Relative Position of a Satellite in the Same Orbit- This procedure is used in reconfiguration models for scenarios 2 and 3 Sections 4. To change the relative position of a satellite in the same orbit, the period of the satellite must be altered. For instance, consider two satellites in the same circular orbit, the distance both angular and linear between them remains constant.
To change the relative position between the two satellites, one of them is temporarily moved to a higher or lower orbit elliptical and subsequently returned to the original circular orbit at a new position. Equations 4. This method is used in Sections 4. How fast a satellite can maneuver depends on the amount of propellant it carries to fire its thrusters for Altitude and Orbit Control Purposes.
There are practical limits to the amount of the propellant which can be carried on-board of a satellite, since it increases the total mass at launch. Currently, there are new thruster technologies which can be used rather than the so called conventional thrusters chemical propellant. Below are the properties of different thrusters which can be used for different purposes : Conventional Thrusters- The power source is a chemical reaction. Electric Arc-jet Thrusters- Use an arc-jet to preheat the propellant before burning to improve the efficiency.
Ion Thrusters- Use electric ion reaction which is the main alternative for conventional thrusters. The low thrust results in a longer time to complete the maneuver. They can produce a high thrust force T. This reconfiguration does not guarantee an uninterrupted operation of the network but ensure that the services can be restored following a failure event.
Depending on the type of the network and the satellite, a degraded performance may result. Both of these methods can be applied to reconfigure the multi-satellite interactions to its full functional conditions. The most appropriate reconfiguration method is determined according to some key parameters such as network topology, network operator speed, cost, etc.
For example, networks with relatively small number of ground station antennas can consider the use of in-orbit backup satellite located at a different orbital position as long as it is visible from all ground stations. First, the backup satellite is moved to a higher or lower orbit to change its period, and then after appropriate time has passed, it is maneuvered to its desired position relative to the other satellites in its original orbit with the same period. This reconfiguration procedure is applicable if the satellites obey one of the two covering concepts as explained in Section 4.
The footprint of the satellites is divided into radio cells spot beams each of which corresponding to a beam of a satellite antenna. Referring to Figure 4. So the earth users experience two types of handovers: beam to beam and satellite to satellite. For instance, in Figure 3. For EFC coverage type, as represented in Figure 4. When the time period ends, each beam is assigned to an adjacent cell on the ground. Sze, N. December 26, Ellingwood, B. Introduction The aim of the study is to validate the concepts of the development of the traffic safety system, the definition of strategies to reduce the accident rate in road transport under modern conditions.
Paper Publication All accepted papers must be The International System Safety Society is a non-profit organization supporting safety professionals worldwide. The RSSR conference began in addressing critical problems faced by the modern railway — how to deliver reliable service to passengers and to freight operators, while maintaining very high levels of safety. He received a Ph. The results show that the proposed method significantly improves the accuracy and efficiency of time-dependent reliability analysis. Maintenance data collection for subsea systems: A critical look at terms and information used for prediction of down time.
Zuo, "Multi-state degradation analysis for a condition monitored device with unobservable states" This increases the reliability. Pete Rotella. The system safety concept focuses on the application of systems engineering and systems management to the process of hazard, safety and risk analysis. All workshops will be held during the afternoon on December 11, May 16, Camera-ready due: June 12, The procedure for registration for the workshops will be made available on this website.
Pre-Conference Workshops We are pleased to offer three pre-conference workshops.
- A Talent for Living: Josephine Pinckney and the Charleston Literary Tradition?
- Stochastic Petri Nets: Modelling, Stability, Simulation.
- Organometallic Chemistry?
- Aims of Education!
ICSEEA is attended by government representative, industrial company, international researchers, scientists, and practitioners in the fields of sustainable energy. The proceedings of this conference will be available for purchase through Curran Associates. It aims to provide a good platform for scholars and researchers in the field of energy and environmental protection to discuss the latest developments and achievements, work out good solutions, and make contributions.
Verbania, lakeside view. After the conference, extended versions of selected contributions will be considered for publication in a Special Issue of the International Journal of Critical Computer-Based Systems. Zhengwei Hu and Xiaoping Du. This conference aims to bring together professionals, government, and academics in China and Internationally to share advancements in process safety. The 2nd international conference on Advanced Communication Systems and Information Security ACOSIS is a forum for scientists, engineers, and practitioners to present their latest research results, ideas, developments, and applications in all areas of communication systems and information security.
Journal of Radioanalytical and Nuclear Chemistry :1, Hsueh and A. RW- th International Conference on Science, Technology, Engineering and Management ICSTEM is a prestigious event organized with a motivation to provide an excellent international platform for the academicians, researchers, engineers, industrial participants and budding students around the world to SHARE their research findings with the global experts.
Today, this issue is particularly tough and important. Pulugurtha, Srinivas S. June , Mr. Published by Elsevier B. October , November , With a wide range of individual and corporate members, the Society is affiliated with major corporations, educational institutions and other agencies.