Search Results

A new network solution has been developed and implemented by Ford Audio engineering to support several upcoming distributed audio product programs. Organized around the seven layers of the OSI model and implemented primarily in software, ACP is based on the physical layer of SAE J1708 (same as Ford's DCL) and a portion of the SAE J2178 message structure. ACP's UART-based protocol depends heavily on a conversation structure that utilizes message coupling to provide reliable distributed control functionality. The use of controlled network traffic, a collision avoidance technique, and a structured message language enhance the efficiency of this new network realization.

The structure of network conversation is an important adjunct to the process of in-vehicle data transfer. Beyond the physical layer, data link, address, and message structures, the utilization of network conversation is one of the system integrator's key tools in implementing network-based products. To use networks for both open-loop and closed-loop distributed functions, many system designers are implementing in software some of the elements of the OSI Session Layer. This paper examines the characteristics and timing aspects of network dialogue and their strong relationship with the performance of distributed operating systems.

Ford's Audio Control Protocol (ACP) is a UART-based protocol which does not use bitwise arbitration. Instead, software algorithms provide collision avoidance and collision resolution. The collision resolution method is non-deterministic and whether or not all collisions can be resolved within a specified time interval is an important ACP protocol issue. To focus on this technical issue, an executable mathematical model of ACP's data link layer was constructed to characterize all transmit, receive, and timing elements of the protocol. Given the performance requirement that all collisions must resolve within a specified time interval, this model was used to analyze the protocol's byte-level collision behavior. This paper examines the performance characteristics of ACP's collision resolution algorithm, and recommends improvements.

Because of the large software and networking architecture contained within the overall organization of OSEK, most beginners will find learning OSEK to be rather difficult. Additionally, if one starts with little or no experience in developing a distributed embedded system using small area network technology, the learning curve will be even longer. This technical paper aims at introducing the essential concepts behind one of OSEK's three major components - the OSEK Communication. The primary focus is to help the beginner understand the basic technical features of OSEK COM.