Research paper Regarding Scheduling and fault detection and diagnosis in embedded systems

1.     A safety concept for a wind power mixed-criticality embedded system based on multicore partitioning

In this paper, the author has demonstrated the safety concept for the wind power embedded system that are completely based on the multicore partitioning. Technique fault-tolerance can be reviewed for the power electronic system and electric drive system, from the viewpoint of industrial applications. The second part of the article includes the preview of the existing methods and applications on the technique of fault diagnosis within an appropriate framework and using the updated references (Perez, Gonzalez, Trujillo, Trapman, & Garate, 2014).

2.      Fault isolation by test scheduling for embedded systems using a probabilistic approach

The author Kadi (2018) has given proper information about the fault isolation through the help of test scheduling of different embedded systems with the help of probabilistic approach. And the other technique is the signal based fault diagnosis method in this method, the explicit input and output used to measure the signs in the fault diagnosis; the model has a wide range of applications to monitor diagnose the power converters and the induction motors in the real-time.

The process of extraction can be classified as quantitsative or qualitative in nature. The qualitative method emerges as the branch of artificial intelligence, which is the basic rule system to present the human expertise in the 1960s, and this is performed the online way to monitor the data in terms of past experiences (Aït-Kadi, Simeu-Abazi, & Arous, 2018).

3.      Security-aware mapping and scheduling with hardware co-processors for flexray-based distributed embedded systems

This paper is deals with the security awareness for mapping and scheduling the main processors that are completely based on distributed embedded systems. This paper is highlighting the real-time application problems related to embedded systems. It can be seen that in the cyber-physical control system implemented in such systems

These areas are composed of interconnected areas and also heterogeneous processing elements. Through this system, it will be quite easy to handle any kind of problem that is present in embedded systems (Gu, Han, Zeng, & Zhao, 2016). 

4.      Adaptive fault diagnosis algorithm for controller area network

in this paper, the author has explained some algorithms that are related the embedded systems. This proposed system will minimize errors that are regarding computational overheads by just remove the ideal subset of linearization

This paper is completely related to the implementation of the intra-tasks vertex and its main impact on the system. Moreover, the control execution order is presented in a proper way to carry out the vertex level assignment

Through such systems, it can be quite easy to meet the requirement. But the main problem is that such systems must be parallelized so making sequences will be quite easy (Kelkar & Kamal, 2014).

5.      The ACROSS MPSoC–A new generation of multi-core processors designed for safety–critical embedded systems

In this paper the author has gave proper information about the multi core processors that are related to the safety critical embedded systems. In this paper, the author had developed some analysis techniques for responding to the worst time for executing the intra-task vertex. For that case, there is a need to apply a proper system that can easily respond with the required time

The system is executed properly and it is a sequence execution that is started from one point and ends at one end. In the last step, there is work conserving scheduling of the system. All of these systems are explained through the help of algorithms (Salloum & Martin Elshuber, 2013).

6.      Fault Diagnosis with Model-Based and Signal based Approaches

In this research study the goal of the survey study is to give a detailed overview of the real-time fault and fault-tolerant control with the specific attention on the results which are posted in the last decades. The paper was first presented in 1991 in this research the basic concepts of the fault-tolerant had been introduced to analyze the acceptability of artificial intelligence. And further in 1997 and overview can be held which was given by the system development view, a contributed review can be done by the researchers in which the main issues of the fault-tolerance were presented by the control system in the same year. Some results can also be reviewed along with the fault diagnosis and the model-based tolerant were also be offered along with the brief discussion on the data analysis, which is taken out on the fault-tolerant control of nonlinear system. Technique fault-tolerance can be reviewed for the power electronic system and electric drive system, from the viewpoint of industrial applications. (Stetter, 2019) During the last four decades the outcome of the study was fruitful; a number of survey papers were written on this topic.

The second part of the article includes the preview of the existing methods and applications on the technique of fault diagnosis within an appropriate framework and using the updated references. And one of the processes is a model-based diagnostic method. The model is based on the diagnosed algorithms which were created to measure the consistency between the model-predicted outputs and the output of the sound system. The method is further divided into four categories: deterministic fault, stockist fault, fault diagnosis for hybrid and discrete-event system, fault-diagnosis for the distributed system, and network. And the other technique is the signal based fault diagnosis method in this method, the explicit input and output used to measure the signs in the fault diagnosis; the model has a wide range of applications to monitor diagnose the power converters and the induction motors in the real-time. And signal based survey method classified in the following frequency domain, time domain, and time-frequency domains. In the last the study carried out all the overview of the techniques and the applications and the whole outline of the fault-diagnosis. The overview gives all the information about the model of fault diagnosis including the detailed brief of the two approaches. (Ding, 2012)

7.      Fault-Diagnosis with knowledge-based and Hybrid/Active Approaches

In this research study, the goal of the survey study is to give a detailed overview of the fault-diagnosis, which is based on the hybrid active approaches. The research is helping since been in the 1980’s. For the model fault diagnosis, the approaches describe the system model which explains the relationship between the system variables which are available for the designers, for the signal based models the signs variables can be expended. The consistency of the observed behavior and the operating system based on the classification of the aid which is leading by the decision of fault diagnosis. The process of extraction can be classified as quantitsative or qualitative in nature. The qualitative method emerges as the branch of artificial intelligence, which is the basic rule system to present the human expertise in the 1960s, and this is performed the online way to monitor the data in terms of past experiences. In early 1980s and 1990s this technique was applied to the different systems which were related to the engineering system such as gas turbine, chemical process and the automobiles as well.

 The quantitative knowledge-based methods mainly develop to diagnose the problem-solving pattern and recognition problems. The quantitative diagnosis can further divide into two portions statistical analysis and non-statistical analysis. In the hybrid method we studied almost 220 techniques and their applications in the last decades by paying attention to the readers from various communities of industries and different social societies to get fastst access to the recent developments that were made in this field. Network and distributed fault diagnosis techniques and their applications may bring positivity in the modern industries, and they spread the wireless communication system through this the industries get their data online. The integration of the data provides a trend to obtained better monitoring for real-time and monitor their performances. (Jianbin Xiong, 2018)

In comparison with the previous approaches this approach of fault diagnosis is more mature and helpful, and this may get the theoretical and anticipated results of the applications. In the last, it is unfortunately not possible to summarize the all the existing publications because of limitations, and in the third part, the survey paper focusing on the underway techniques of the fault-tolearnt control. And an overlook on the future development of faul-tolerant diagnosis has also been presented. In the last the active fault diagnosis, analysis redundancy, fault tolerance and hybrid fault diagnosis can also be covered in this research study. (Saiful Omar, 2018) 

8.      Optimal Scheduling for precedence-constrained task graphs on Heterogeneous Distributed systems with shared buses

In this paper, the author is discussing the making schedule of precedence-constrained task graphs that are implemented on the distributed system that contain shared busses. This paper is highlighting the real-time application problems related to embedded systems. It can be seen that in the cyber-physical control system implemented in such systems. All of these systems are represented as the PTGs, and these graphs are mainly implemented in these heterogeneous areas according to the demand of the system.

 It can be noted that there is a need for optimal scheduling for implementing heterogeneous areas. This is because such a system is providing a lot of advantages to the embedded systems. The performance of the system is improved and also enhanced when it is connected to different channels.

In this paper, the author had proposed a solution that is completely based on integer linear programming solutions that can be used for the scheduling of PTG for the distributed areas. These areas are composed of interconnected areas and also heterogeneous processing elements. Through this system, it will be quite easy to handle any kind of problem that is present in embedded systems.  In the real world, the automatic cruise controller is playing a major role in demonstrating the accountability of the system.

The reason is that in the past, PTGs systems are not perfect and also contain some drawbacks. This is because such systems were based on list scheduling. The author had presented complete information about the PTGs systems, and also, its issues are discussed in that paper. There are many types of research that are working on the scheduling of a multiprocessor system. The main reason is that through this, they can easily apply a different kind of approaches related to linear programming.

The author had presented some models for implementing the PTGs systems. The first one ware related to the platform model, and it is explained with the help of examples. It can be seen that the problem of the optimal schedule has been solved, and it is completely based on the heterogeneous processing nodes that are connected with each other to form an inter-connected with the help of shared buses.

This technology is implemented through the help of interconnected shared buses. For making a perfect system for the PTG programming, there is a need to propose a new system that is ILO formulation because it can easily map from one node to another node easily without any difficulty. This proposed system will minimize errors that are regarding computational overheads by just remove the ideal subset of linearization. The author had also presented a complete experimental analysis of the PTGs systems for checking the efficacy of the scheme (Roy, Sayani Sinha, Devaraj, & Sarkar, 2019).

9.      Intra-Task Priority Assignment in Real-Time scheduling of DAG tasks on multi-cores

In this paper, the author Qingqiang had presented a complete analysis of the intra-task priority assignment. It can be noted that there are a lot of problems related to real-time scheduling and analysis. The reason is that it is quite difficult to model it as the directed acyclic graphs. Such graphs have been studied by different researchers. But there is no proper work has been proposed related to the executing of the eligible vertices for the DAG task. This paper is completely related to the implementation of the intra-tasks vertex and its main impact on the system. Moreover, the control execution order is presented in a proper way to carry out the vertex level assignment.

In this paper, the author had developed some analysis techniques for responding to the worst time for executing the intra-task vertex. For that case, there is a need to apply a proper system that can easily respond with the required time. This new system is also able to enhance the schedulability of the system in an efficient way. this proposed technique is then further analyzed for getting proper results about such tasks. there are some experiments about the system also presented in a proper way.

It can be seen that there are some multi-core platforms that are related to finding the real meantime of the system. Through such systems, it can be quite easy to meet the requirement. But the main problem is that such systems must be parallelized so making sequences will be quite easy. This is the reason why there are a lot of challenges that are refraining this system from making real-time designs. In this paper, there is comprehensive information on the DAG task has been presented with a proper example.

For that, in the start, a complete system model is presented by multi-core platforms that contain M cores that are identical to each other. There are different vertexes attached in the form of a node to make a complete DAG system. Then after this, its runtime behavior is tested properly with some examples. The system is executed properly and it is a sequence execution that is started from one point and ends at one end. In the last step, there is work conserving scheduling of the system. All of these systems are explained through the help of algorithms.

It can be seen that if priorities are assigned to the vertices of DAG, then it will be quite simple for deriving high-quality responses that are time-bound and also arranged in the parallel from. The author had presented complete dynamic programming about DAG example just for explaining the priority assignment algorithms. The results are showing that there are stills some problems related to the time that can be resolved in the future by analyzing its relationship with response time and efficiency (He, Jiang, Guan, & Guo, 2019).

10.  Multi-variant Scheduling of critical time-triggered communication in the incremental development process: Application to FlexRay

In this paper, the author had provided proper information on the problems related to the incremental development process that can be used for analyzing the embedded systems. There are many car manufacturers that are a present portfolio of models. Such models contain different kinds of variants. Then due to such variants, it will become one of the formidable task for the car manufactures to manage such variants. For that case, there is a need for such a system that can easily keep these variants as close as possible. It is one of the most difficult task, and the main reason is that there is a need for enhancement in communication protocols. If any vehicle is using the same protocol as the other one, then it will be quite difficult to identify the same signal. For this, there is a need for a system that will simultaneously schedule that signal from the car. Moreover, it can be noted that the new variants are designed in such a way that they can easily able to maintain the backward compatibility with the other one. This backward compatibility in the vehicle will help to minimize the problem related to expenses and also the physical condition of the cars.

They are also lowering the components of fine-tuning of the vehicle. For the perfect solution, the car manufacturers are using the same platform, and the reason is that it is providing signal traceability and also less expansive. In this paper, the author had presented a robust and efficient algorithm for scheduling and also for making a proper connection for efficient internal communication. All of these algorithms are quite beneficial for scheduling the internal communication of the variants of vehicles. Through the help of these algorithms, it will be quite easy to check the compatibility of the system. This is because such algorithms are presenting variant management, and also they are taking care of the backward compatibility present in the vehicles.

The author had presented the required problem in cars in an efficient way. as time increases, the car manufacturers have to manage different models. And as time passes, that will become one of the huge problems for them to manage model variants.

In this paper, there is complete information about the implementation and designs of the algorithms that can be used for solving the variant problems in the vehicles. To solve this problem, the researcher has presented an algorithm named a FlexRay static segment. This can be implemented for enabling the time-triggered communication between the different models of the car. FlexRay algorithm is explained in a proper way in that paper. This is the bus that is standardized in ISO 17458. 

The communication is implemented in different cycles, and each of its cycles consists of a 6-bit id that is also known as cycleid. This algorithm is tested in a proper way in this paper. This paper is also solving the problem related to the scheduling of the time-triggered internal communication for different kinds of vehicle models. This problem can be solved by implementing shared constrains and this is done through the help of FlexRay bus. The experimental results are discussed in a proper way that is completely based on its analysis that is completely according to the required time. There are stills some problems related to the bottleneck and its limitations, but there is a linear relationship between the different parameters of the bus. The good point was that the relationship between the iteration of incremental scheduling and bandwidth occupancy is reliable (Dvorak & Hanzalek, 2019).

11.  Incremental flow Scheduling and routing in time-sensitive software-defined networks

According to the author Naresh(2018), he has given sound knowledge on the time-sensitive networks. It can be noted that there are a lot of networking architectures that are developed for providing real-time information to a different network. But for that, there is a need for time-sensitive triggering that can easily help the user to maintain the traffic of the networks. Nowadays, there is a lot of time-sensitive applications implemented in the industries. The main principle that is presented behind these technologies is the precise transmission of the networks that are based on time.

In this paper, the author had presented a proper system that can easily handle the time-sensitive problem. It can be noted that these schedules are changing and operated offline and also remained fixed. Through this, it will be quite difficult to add and remove any network applications. In this paper, there is comprehensive information about such algorithms that are used for incremental addition for the time triggered. This system is completely based on software-defined networking that can easily transfer messages in real-time.

This system is also providing reliability for the time-triggered flows and its transmission to the other networks. These algorithms are also useful for IoT applications. At the start of the paper, the author has described the cyber-physical systems in detail. The scheduling model has been presented in a proper way. These algorithms are solving the scheduling problem through the use of ILP and also compute transmission schedules. The results show that this system is able to achieve the second setup time for scheduling time sensitive problems. But there is still some requirement of improvement in the system (Nayak, Durr, & Rothermel, 2018).

12.  Incremental Distributed and concurrent Scheduling in System-of-systems with Real-Time requirements

This paper is discussing the scheduling problem in the systems that are required real-time results. It can be noted that the System-of-System consists of large scale embedded systems. Such systems are defined by the managerial and operational impendence of the specific system. In this paper, there is brief information about the conceptual models that are explained with scheduling algorithms. These algorithms are also involved in solving the real-time problem in the System-of-System. Moreover, the requirement of real-time support is extremely high in many sectors where there is a requirement of safety. Such systems are considered as military, medical and industrial. For such systems, there is a requirement of computation scheduling that can be implemented easily through the help of different systems that contain less knowledge and also no control on SoS.

But such systems are involved in reducing the time. For that case, there are some concurrent scheduling activities that are involved in dealing with such a problem in the system. In that paper, there is a brief introduction about the high level and upgraded scheduling algorithms for System-of-systems so that the problem related to real-time can be solved easily.  There are many huge companies that are working on incremental scheduling problems that are present in such a system.

In the start of the paper, there is some information about the conceptual model of this system. This model was drawn according to the physical and logical viewpoints. For the formulation of the allocation problem and dynamic scheduling, these structural models are playing a major role. The author has also given complete and comprehensive information on the incremental and distributed scheduling of the system. There is also detail information on the incremental scheduling that will help the SoS system to transfer real-time data in a proper way. For that case, complete communication parameters are modified and explained. It can be noted that such scheduling is driven through the help of the DAG application. The algorithms are explained in a proper way of minimizing real-time issues. The author has also presented complete information about the allocation in this system. For the scheduling, the author has described two main levels. The first level consists of a high-level allocation of the application in a proper way. the next level is consists of a low level. In the next section, the author has presented the scheduling model of the system by applying a different kinds of steps.

The main problem is related to the scheduling of the constraints for minimizing the time problem. he has also defined the incremental scheduling constraints for the algorithm. In that algorithm, there is complete information about the ecosystem of the SoS. The result of these experiments is excluded by testing three main SoS separately. For that case, every SoS may contain different applications. Through the help of CPLEX 12.6.1, the computation times are obtained in an efficient way. This system can be used for the local scheduling for each system so that good results can be achieved easily. It can be noted that the importance of SoS that contains real-time constraints has been increasing at a fast rate. The main reason is that there is a requirement for making a proper schedule for an end to end communication. Now the main thing was that the end to end communication constraints are required to span the constituent system to be more accurate and also contain managerial independence in the system.

But this problem has been solved by introducing incremental and distributed scheduling. Through this system, there is a huge reduction in the computation time. Moreover, this system is also coping with the main data from different users but contains some limitations. The author has also mentioned some limitations regarding this system that will be resolved in future studies (Roman Obermaisser, 2015).

13.  Scheduling and Mapping in an Incremental Design Methodology for distributed real-time embedded system

In this paper, the author Paul (2014) has presented proper information on the scheduling and mapping for the incremental design of the embedded systems. The author has given comprehensive detail on mapping and scheduling on the distribution of embedded systems in real-time. There are many researchers that are focusing on minimizing the cost of the system. This can be done through the help of an incremental design process. The author has given proper information about this process in detail.

Moreover, this system is also involved in making the system more accurate by reducing its cost. The author is also implementing new functions that are completing the time required for the system. There are also some systems that are implemented that contain little bit changes for real-time results. There is also a huge chance of adding high-quality functions for completing the required functions. Furthermore, it can be noted that complex embedded systems are lacking in the real-time evaluation of the results through this. The system is facing a lot of problems regarding this issue.

In this paper, the author was working on the distributed real-time embedded systems that are operated in different organizations. The architecture of the system mainly consists of two main parts the first one is related to the TDMA round, and the next one is the cycle of two rounds. There is also a TTP controller that is present on every node of the system that is providing real-time protocols to the system. This controller is also increasing the rate of communication with the help of shared values. Moreover, the author has also presented the proper software architecture of the system. It contains a lot of nodes, and each of them contains Kernel. There is also a proper schedule for each Kernel.

      The author has also presented a mapping and scheduling strategy in a comprehensive way. Then in the last part of the paper, the author has also given proper information about the evaluation of its cost in a proper way. The author has demonstrated a complete system in a proper way with the help of an incremental design approach. This design process is also satisfying the main requirement of the system that is completely related to real time applications. He has also explained two main designs according to their corresponding matrices. All of these solutions are presented through the help of transformation that is completely based on the heuristic. These algorithms are also solving the problem related to time. This is because the GH system is extremely fast and can easily convey the message at low cost and high efficiency. Moreover, SH systems are able to produce quality results. The author was unable to solve the issue of architecture selection (Pop, Eles, & Peng, 2014).

References of Scheduling and fault detection and diagnosis in embedded systems

Aït-Kadi, D., Simeu-Abazi, Z., & Arous, A. (2018). Fault isolation by test scheduling for embedded systems using a probabilistic approach. Journal of Intelligent Manufacturing.

Ding, S. X. (2012). Model-Based Fault Diagnosis Techniques. Springer Science & Business Media.

Dvorak, J., & Hanzalek, Z. (2019). Multi-Variant Scheduling of Critical Time-Triggered Communication in Incremental Development Process: Application to FlexRay. IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY,.

Gu, Z., Han, G., Zeng, H., & Zhao, Q. (2016). Security-aware mapping and scheduling with hardware co-processors for flexray-based distributed embedded systems. EEE Transactions on parallel and distributed systems.

He, Q., Jiang, X., Guan, N., & Guo, Z. (2019). Intra-Task Priority Assignment in Real-Time Scheduling of DAG Tasks on Multi-Cores. IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS.

Jianbin Xiong, C. L. (2018). Fault Diagnosis Method Based on Improved Evidence Reasoning. Infinite Study.

Kelkar, S., & Kamal, R. (2014). Adaptive fault diagnosis algorithm for controller area network. IEEE transactions on Industrial Electronics.

Nayak, N. G., Durr, F., & Rothermel, K. (2018). Incremental Flow Scheduling and Routing in Time-Sensitive Software-Defined Networks. IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS.

Perez, J., Gonzalez, D., Trujillo, S., Trapman, T., & Garate, J. M. (2014). A safety concept for a wind power mixed-criticality embedded system based on multicore partitioning. In 11th International TUV Rheinland Symposium “Functional Safety in Industrial Applications.

Pop, P., Eles, P., & Peng, Z. (2014). Scheduling and Mapping in an Incremental Design Methodology for Distributed Real-Time Embedded Systems. IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS,.

Roman Obermaisser, A. M. (2015). Incremental, Distributed and Concurrent Scheduling in Systems-of-Systems with Real-Time Requirements. IEEE International Conference on Computer and Information Technology; Ubiquitous Computing and Communications; Dependable, Autonomic and Secure Computing; Pervasive Intelligence and Computing.

Roy, S. K., Sayani Sinha, K. M., Devaraj, R., & Sarkar, A. (2019). Optimal Scheduling of Precedence-constrained Task Graphs on Heterogeneous Distributed Systems with Shared Buses. IEEE 22nd International Symposium on Real-Time Distributed Computing .

Saiful Omar, W. S.-A. (2018). Computational Intelligence in Information Systems. Springer.

Salloum, C. E., & Martin Elshuber, O. H. (2013). The ACROSS MPSoC–A new generation of multi-core processors designed for safety–critical embedded systems. Microprocessors and Microsystems .

Stetter, R. (2019). Fault-Tolerant Design and Control of Automated Vehicles and Processes. Springer.