No abstract

In this contribution we present the first method for model checking on nonlinear analog systems. Based on digital CTL model checking algorithms and results in hybrid model checking, we have developed a concept to adapt these ideas to analog systems. Using an automatic state space subdivision method the continuous state space is transfered into a discrete model. In doing this, the most challenging task is to retain the essential nonlinear behavior of the analog system. To describe analog specification properties, an extension to the CTL language is needed. Two small examples show the properties and advantages of this new method and the capability of the implemented prototype tool.

No abstract

The presented work inside this thesis aims to raise the degree of automation in analog circuit design. Therefore, a framework was developed to provide the necessary mechanisms in order to carry out a fully automated analog circuit synthesis, i.e., the construction of an analog circuit fulfilling all previously defined (electrical) specifications.Nowadays, analog circuit design in general is a very time consuming process compared to a digital design flow. Due to its discrete nature, the digital design process is highly automated and thus very efficient compared to analog circuit design. In modern Very-Large-Scale integration (VLSI) circuits the analog parts are mostly just a small portion of the overall chip area. Although this small portion is known to consume a major part of the needed workforce. Paired with product cycles which constantly get shorter, the time needed to develop the analog parts of an integrated circuit (IC) becomes a determinant factor. Apart from this, the ongoing progress in semiconductor processing technologies promises more speed with less power consumption on smaller areas, forcing the IC developers to keep track with the technology nodes in order to maintain competitiveness. Analog circuitry exhibits the inherent property of being hard to reuse, as porting from one technology node to another imposes critical changes for operating conditions (e.g., supply voltage) -mostly leading to a full redesign for most of the analog modules. This productivity gap between digital and analog design resembles the primary motivation for this thesis.Due to the availability of commercial sizing tools, this work deliberately focuses on the construction of circuit topologies in distinction to parameter synthesis, which can be obtained with a dedicated sizing tool. The focus on circuit construction allows the development of a framework which allows a full design space exploration. This thesis describes the needed concepts and methods to realize a deterministic, explorative analog synthesis framework. Despite this, a reference implementation is presented, which demonstrates the applicability in current analog design flows. circuit (c) a circuit representing a physical realization circuits (C) an unordered set of circuits design space (DS) design space seen as crossproduct of P S and ST S devices (M ) an unordered set of devices equivalence class (eqcls) an unordered set of objects sharing the combination set of invariant properties equivalence classes (EQCLS(c)) set of equivalence classes associated with the circuit c graph (G) a representation of object vertices connected through edges ground (V GN D ) name of the reference net or simply ground hash (hash(x)) hash associated with object x integer (N) integer numbers invariant properties (EQP ROP S(x)) set of invariant properties associated with x invariant property (p) a property associated with a specific IV i invariants (IV) an unordered set of invariants label (L(x)) label associated with object x nets (N ) an unordered set of nets parameter space ...

This paper proposes a novel modular architecture for Electrical Energy Storages (EESs), consisting of multiple seriesconnected cells. In contrast to state-of-the-art architectures, the presented approach significantly improves the energy utilization, safety, and availability of EESs. For this purpose, each cell is equipped with a circuit that enables an individual control within a homogeneous architecture. One major advantage of our approach is a direct and concurrent charge transfer between each cell of the EES using inductors. To enable a system-level modeling and performance analysis of the architecture, a detailed investigation of the components and their interaction with the Pulse Width Modulation (PWM) control was performed at transistor-level. At system-level, we propose a control algorithm for the charge transfer that aims at minimizing the energy loss and balancing time. The results give evidence of the significant advantages of our architecture over existing passive and active balancing methods in terms of energy efficiency and charge equalization time.