After a brief introduction to low-power VLSI design, the design space of ASIP instruction set architectures (ISAs) is introduced with a special focus on important features for digital signal processing. Based on the degrees of freedom offered by this design space, a consistent ASIP design flow is proposed: this design flow starts with a given application and uses incremental optimization of the ASIP hardware, of ASIP coprocessors and of the ASIP software by using a top-down approach and by applying application-specific modifications on all levels of design hierarchy. A broad range of real-world signal processing applications serves as vehicle to illustrate each design decision and provides a hands-on approach to ASIP design. Finally, two complete case studies demonstrate the feasibility and the efficiency of the proposed methodology and quantitatively evaluate the benefits of ASIPs in an industrial context.

Today more than 90% of all programmable processors are employed in embedded systems. This number is actually not surprising, contemplating that in a typical home you might find one or two PCs equipped with high-performance standard processors, and probably dozens of embedded systems, including electronic entertainment, household, and telecom devices, each of them equipped with one or more embedded processors. The question arises why programmable processors are so popular in embedded system design. The answer lies in the fact that they help to narrow the gap between chip capacity and designer productivity. Embedded processors cores are nothing but one step further towards improved design reuse, just along the lines of standard cells in logic synthesis and macrocells in RTL synthesis in earlier times of IC design. Additionally, programmable processors permit to migrate functionality from hardware to software, resulting in an even improved reuse factor as well as greatly increased flexibility.

The LISA processor design platform (LPDP) presented in Architecture Exploration for Embedded Processors with LISA addresses recent design challenges and results in highly satisfactory solutions. The LPDP covers all major high-level phases of embedded processor design and is capable of automatically generating almost all required software development tools from processor models in the LISA language. It supports a profiling-based, stepwise refinement of processor models down to cycle-accurate and even RTL synthesis models. Moreover, it elegantly avoids model inconsistencies otherwise omnipresent in traditional design flows.

The next step in design reuse is already in sight: SoC platforms, i.e., partially pre-designed multi-processor templates that can be quickly tuned towards given applications thereby guaranteeing a high degree of hardware/software reuse in system-level design. Consequently, the LPDP approach goes even beyond processor architecture design. The LPDP solution explicitly addresses SoC integration issues by offering comfortable APIs for external simulation environments as well as clever solutions for the problem of both efficient and user-friendly heterogeneous multiprocessor debugging.

Integrated System-Level Modeling of Network-on-Chip Enabled Multi-Processor Platforms first gives a comprehensive update on recent developments in the area of SoC platforms and ESL design methodologies. The main contribution is the rigorous definition of a framework for modeling at the timing approximate level of abstraction. Subsequently this book presents a set of tools for the creation and exploration of timing approximate SoC platform models.

Computerarchitecturepresentlyfacesanunprecedentedrevolution: Thestep from monolithic processors towards multi-core ICs, motivated by the ever - creasingneedforpowerandenergyef ciencyinnanoelectronics. Whetheryou prefer to call it MPSoC (multi-processor system-on-chip) or CMP (chip mul- processor), no doubt this revolution affects large domains of both computer science and electronics, and it poses many new interdisciplinary challenges. For instance, ef cient programming models and tools for MPSoC are largely an open issue: "Multi-core platforms are a reality - but where is the software support" (R. Lauwereins, IMEC). Solving it will require enormous research efforts as well as the education of a whole new breed of software engineers that bring the results from universities into industrial practice. Atthesametime, thedesignofcomplexMPSoCarchitecturesisanextremely time-consuming task, particularly in the wireless and multimedia application domains, where heterogeneous architectures are predominant. Due to the - ploding NRE and mask costs most companies are now following a platform approach: Invest a large (but one-time) design effort into a proper core - chitecture, and create easy-to-design derivatives for new standards or product features. Needless to say, only the most ef cient MPSoC platforms have a real chance to enjoy a multi-year lifetime on the highly competitive semiconductor market for embedded systems.

New software tools and a sophisticated methodology above RTL are required to answer the challenges of designing an optimized application specific processor (ASIP). This book offers an automated and fully integrated implementation flow and compares it to common implementation practice. It provides case-studies that emphasize that neither the architectural advantages nor the design space of ASIPs are sacrificed for an automated implementation.

Increasing complexity of modern embedded systems demands system designers to ramp up their design productivity without compromising performance goals. This is promoted by modern Electronic System Level (ESL) techniques. Language-driven Exploration and Implementation of Partially Re-configurable ASIPs addresses an important segment of the ESL area by modeling partially re-configurable processors via high-level Architecture Description Language (ADL). This approach also hints an imminent evolution in the area of re-configurable system design.

Computerarchitecturepresentlyfacesanunprecedentedrevolution: Thestep from monolithic processors towards multi-core ICs, motivated by the ever - creasingneedforpowerandenergyef ciencyinnanoelectronics. Whetheryou prefer to call it MPSoC (multi-processor system-on-chip) or CMP (chip mul- processor), no doubt this revolution affects large domains of both computer science and electronics, and it poses many new interdisciplinary challenges. For instance, ef cient programming models and tools for MPSoC are largely an open issue: "Multi-core platforms are a reality - but where is the software support" (R. Lauwereins, IMEC). Solving it will require enormous research efforts as well as the education of a whole new breed of software engineers that bring the results from universities into industrial practice. Atthesametime, thedesignofcomplexMPSoCarchitecturesisanextremely time-consuming task, particularly in the wireless and multimedia application domains, where heterogeneous architectures are predominant. Due to the - ploding NRE and mask costs most companies are now following a platform approach: Invest a large (but one-time) design effort into a proper core - chitecture, and create easy-to-design derivatives for new standards or product features. Needless to say, only the most ef cient MPSoC platforms have a real chance to enjoy a multi-year lifetime on the highly competitive semiconductor market for embedded systems.

For the first time advances in semiconductor manufacturing do not lead to a corresponding increase in performance. At 65 nm and below it is predicted that only a small portion of performance increase will be attributed to shrinking geometries while the lion share is due to innovative processor architectures. To substantiate this assertion it is instructive to look at major drivers of the semiconductor industry: wireless communications and multimedia. Both areas are characterized by an exponentially increasing demand of computational power, which cannot be provided in an energy-efficient manner by traditional processor architectures. Todaya (TM)s applications in wireless communications and multimedia require highly specialized and optimized architectures.

New software tools and a sophisticated methodology above RTL are required to answer the challenges of designing an optimized application specific processor (ASIP). This book offers an automated and fully integrated implementation flow and compares it to common implementation practice. Case-studies emphasise that neither the architectural advantages nor the design space of ASIPs are sacrificed for an automated implementation. Realizing a building block which fulfils the requirements on programmability and computational power is now efficiently possible for the first time.

Optimized ASIP Synthesis from Architecture Description Language Models inspires hardware designers as well as application engineers to design powerful ASIPs that will make their SoC designs unique.

After a brief introduction to low-power VLSI design, the design space of ASIP instruction set architectures (ISAs) is introduced with a special focus on important features for digital signal processing. Based on the degrees of freedom offered by this design space, a consistent ASIP design flow is proposed: this design flow starts with a given application and uses incremental optimization of the ASIP hardware, of ASIP coprocessors and of the ASIP software by using a top-down approach and by applying application-specific modifications on all levels of design hierarchy. A broad range of real-world signal processing applications serves as vehicle to illustrate each design decision and provides a hands-on approach to ASIP design. Finally, two complete case studies demonstrate the feasibility and the efficiency of the proposed methodology and quantitatively evaluate the benefits of ASIPs in an industrial context.

Today more than 90% of all programmable processors are employed in embedded systems. This number is actually not surprising, contemplating that in a typical home you might find one or two PCs equipped with high-performance standard processors, and probably dozens of embedded systems, including electronic entertainment, household, and telecom devices, each of them equipped with one or more embedded processors. The question arises why programmable processors are so popular in embedded system design. The answer lies in the fact that they help to narrow the gap between chip capacity and designer productivity. Embedded processors cores are nothing but one step further towards improved design reuse, just along the lines of standard cells in logic synthesis and macrocells in RTL synthesis in earlier times of IC design. Additionally, programmable processors permit to migrate functionality from hardware to software, resulting in an even improved reuse factor as well as greatly increased flexibility.

The LISA processor design platform (LPDP) presented in Architecture Exploration for Embedded Processors with LISA addresses recent design challenges and results in highly satisfactory solutions. The LPDP covers all major high-level phases of embedded processor design and is capable of automatically generating almost all required software development tools from processor models in the LISA language. It supports a profiling-based, stepwise refinement of processor models down to cycle-accurate and even RTL synthesis models. Moreover, it elegantly avoids model inconsistencies otherwise omnipresent in traditional design flows.

The next step in design reuse is already in sight: SoC platforms, i.e., partially pre-designed multi-processor templates that can be quickly tuned towards given applications thereby guaranteeing a high degree of hardware/software reuse in system-level design. Consequently, the LPDP approach goes even beyond processor architecture design. The LPDP solution explicitly addresses SoC integration issues by offering comfortable APIs for external simulation environments as well as clever solutions for the problem of both efficient and user-friendly heterogeneous multiprocessor debugging.

Integrated System-Level Modeling of Network-on-Chip Enabled Multi-Processor Platforms first gives a comprehensive update on recent developments in the area of SoC platforms and ESL design methodologies. The main contribution is the rigorous definition of a framework for modeling at the timing approximate level of abstraction. Subsequently this book presents a set of tools for the creation and exploration of timing approximate SoC platform models.

Increasing complexity of modern embedded systems demands system designers to ramp up their design productivity without compromising performance goals. This is promoted by modern Electronic System Level (ESL) techniques. Language-driven Exploration and Implementation of Partially Re-configurable ASIPs addresses an important segment of the ESL area by modeling partially re-configurable processors via high-level Architecture Description Language (ADL). This approach also hints an imminent evolution in the area of re-configurable system design.