ForSyDe.Atom is a formal framework for modeling and simulating heterogeneous embedded and cyber-physical systems (CPS) at a high level of abstraction, whose "spiritual parent" is the ForSyDe modeling framework [Sander04]. In ForSyDe, heterogeneous systems are modeled as networks of processes communicating through signals. In ForSyDe, processes alone capture the timing semantics of execution and synchronization according to a certain model of computation (MoC). The shallow implementation of ForSyDe, forsyde-shallow provides libraries of higher-order functions for instantiating processes called process constructors for several MoCs.

The forsyde-atom project started as a proof-of-concept for the atom-based approach to CPS introduced in [Ungureanu17]. This approach extends the ideas of the tagged signal model by systematically deconstructing processes to their basic semantics and recreating them using a minimal language of primitive building blocks called atoms. It also expands the scope of this model by exploiting more aspects of cyber-physical systems than just timing, but also adding primitives for parallelism, behavior extensions, probabilistic distribution, etc., each in its own interacting environment called layer.

The API documentation is meant to be a comprehensive reference guide for an interested user, with enough details, pictures and examples to ease the understanding of each API element. It is not meant to be an introduction to the ForSyDe-Atom methodology nor a modeling manual. For proper introductory material we recommend consulting the following:

• [Ungureanu20a] is a peer-reviewed comprehensive introduction to the methodology and the scientific reasoning behind this modeling framework. It defines the main modeling concepts proposed by ForSyDe-Atom: atom, pattern and layer.
• [Ungureanu20b] is a collection of step-by-step tutorials and simple case studies.
• the Documentation page from the ForSyDe web site contains up-to-date pointers to further useful material.

ForSyDe.Atom is a collection of libraries, each representing a layer, which is an own DSL for describing one modeling aspect in a CPS. Each module under ForSyDe.Atom , e.g. ForSyDe.Atom.MoC represents a layer, defines the main atoms and the generic patterns for the respective layer. Each module under a layer, e.g. ForSyDe.Atom.MoC.SY describes a sub-domain of that layer and defines the actual semantics of atoms (i.e. overloads atom functions), as well as other specific patterns and utilities. Due to deliberate name clashes and to improve readability, each module needs to be imported with a (maybe qualified) alias:

import ForSyDe.Atom.MoC    as MoC
import ForSyDe.Atom.MoC.SY as SY

-- MoC.comb11 /= SY.comb11

By default, the current module, ForSyDe.Atom, only re-exports a couple of generic utilities for working with tuples and for plotting signals.

Following are the layers provided by the current version of ForSyDe-Atom. Click on any of the links for more documentation:

• ForSyDe.Atom.MoC, a DSL for capturing the semantics of computation and concurrency according to a model of computation.
• ForSyDe.Atom.Skel a DSL for describing structured parallelism.
• ForSyDe.Atom.Probability, a DSL for describing numerical values as probability
• distributions, e.g. Gaussian.
• ForSyDe.Atom.ExB, a DSL for extending the pool of values with logic symbols with well-kown semantics (e.g. absent values).

IMPORTANT!!! All multi-argument functions and utilities provided by the forsyde-atom API are named along the lines of functionMN where M represents the number of curried inputs (i.e. a1 -> a2 -> ... -> aM), while N represents the number of tupled outputs (i.e. (b1,b2,...,bN)). For brevity, we only write documentation for functions with 2 inputs and 2 outputs (i.e. function22), while all the other available ones are mentioned as a regex (i.e. function[1-4][1-4]). In case the provided functions are not sufficient, feel free to implement your own patterns following the examples in the source code.

Here are gathered pointers to documents referenced throughout the API documentation.

[Bonna19] Bonna, R., Loubach, D. S., Ungureanu, G., & Sander, I. (2019). Modeling and simulation of dynamic applications using scenario-aware dataflow. ACM Transactions on Design Automation of Electronic Systems (TODAES), 24(5), 1-29.

[Buck93] Buck, J. T., & Lee, E. A. (1993). Scheduling dynamic dataflow graphs with bounded memory using the token flow model. In 1993 IEEE international conference on acoustics, speech, and signal processing (Vol. 1, pp. 429-432). IEEE.

[Cassandras09] Cassandras, C. G., & Lafortune, S. (2009). Introduction to discrete event systems. Springer Science & Business Media.

[Fujimoto00] Fujimoto, R. M. (2000). Parallel and distributed simulation systems (Vol. 300). New York: Wiley.

[Halbwachs91] Halbwachs, N., Caspi, P., Raymond, P., & Pilaud, D. (1991). The synchronous data flow programming language LUSTRE. Proceedings of the IEEE, 79(9), 1305-1320.

[Gorlatch03] Fischer, J., Gorlatch, S., & Bischof, H. (2003). Foundations of data-parallel skeletons. In Patterns and skeletons for parallel and distributed computing (pp. 1-27). Springer London.

[Kahn76] Kahn, G., & MacQueen, D. (1976). Coroutines and networks of parallel processes.

[Lee87] Lee, E. A., & Messerschmitt, D. G. (1987). Synchronous data flow. Proceedings of the IEEE, 75(9), 1235-1245.

[Lee98] Lee, E. A., & Sangiovanni-Vincentelli, A. (1998). A framework for comparing models of computation. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 17(12), 1217-1229.

[Lohstroh19] Lohstroh, M., Romeo, Í. Í., Goens, A., Derler, P., Castrillon, J., Lee, E. A., & Sangiovanni-Vincentelli, A. (2019). Reactors: A deterministic model for composable reactive systems. In Cyber Physical Systems. Model-Based Design (pp. 59-85). Springer, Cham.

[Reekie95] Reekie, H. J. (1995). Realtime signal processing: Dataflow, visual, and functional programming.

[Sander04] Sander, I., & Jantsch, A. (2004). System modeling and transformational design refinement in ForSyDe [Formal System Design]. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 23(1), 17-32.

[Skillicorn05] Skillicorn, D. B. (2005). Foundations of parallel programming (No. 6). Cambridge University Press.

[Stuijk11] Stuijk, S., Geilen, M., Theelen, B., & Basten, T. (2011, July). Scenario-aware dataflow: Modeling, analysis and implementation of dynamic applications. In 2011 International Conference on Embedded Computer Systems: Architectures, Modeling and Simulation (pp. 404-411). IEEE.

[Ungureanu17] Ungureanu, G., & Sander, I., A layered formal framework for modeling of cyber-physical systems, in 2017 Design, Automation & Test in Europe Conference & Exhibition (DATE), 2017, pp. 1715–1720.

[Ungureanu18] Ungureanu, G., de Medeiros, J. E. G. & Sander, I., Bridging discrete and continuous time models with Atoms, in 2018 Design, Automation & Test in Europe Conference & Exhibition (DATE), 2018, pp. 277-280

[Ungureanu20a] Ungureanu, G., et. al. , ForSyDe-Atom: Taming complexity in cyber-physical system design with layers, under review.

[Ungureanu20b] Ungureanu, G., ForSyDe-Atom: User manual, version 2, 2020