Concurrent token ring (CTR), a multiple-access protocol for ring-type area networks, is proposed. In CTR, message frames are partially removed by the destination node. A truncated portion of the frame, which contains the source and destination addresses, is configured by the destination into a conditional token which is allowed to circulate back to the source for removal. The conditional token informs 'downstream' nodes of busy data paths and allows them to establish concurrent message transmission if resources are free. Simulation results indicate that the CTR protocol has a higher throughput and shorter transmission delay than the IEEE 802.5 token ring.>

The authors introduce a communication subnet architecture, called Hybrid Meshnet, for local area networks (LANs) that is designed to achieve a high degree of network performance and flexibility. These objectives are achieved by utilizing a dual-channel structure. A token ring is used for short or prioritized data transmissions. A complementary mesh network with arbitrary data links is used for relatively high volume and low-priority data transmission. A distributed network control and routing mechanism allocates the communication network nodes. Hybrid Meshnet supports a LAN system in which the communication structure may easily be modified and changed by the insertion or deletion of data links.>

A novel destination partial removal protocol, DRUID (destination removal without intermediate node delay) ring, is proposed for ring-type networks. In the DRUID ring, a node may initiate message transmission if its output link is free. Every node will relay the message immediately if it is idle when the message arrives. At the same time, the node also receives and buffers all incoming messages. Upon decoding the destination address, the destination node will truncate the message frame to a cut frame by eliminating the information field. This cut frame will continue to circulate around the ring until it arrives at a busy node where it is buffered, identified, and removed. The message frame will be completely removed by the destination node if it is busy when the message arrives. The protocol supports high concurrency while the loop latency is minimized.>

This paper presents a hybrid architecture for local area networks (LANs) which combines the advantages of token ring and mesh networks while remaining relatively simple. Each node may have two communication channels. The first is a token ring and is used for short or prioritized message data transmissions. The other consists of an arbitrary collection of data links, which are used for relatively high volume and low priority data messages. With this architecture, a LAN can be installed in the form of token ring and can be upgraded and “tuned” to its application environment by the insertion of node-to-node data links.

FISHNET (facilities integrated in a shared habitat network) addresses the problem of effectively integrating a wide variety of computers, terminals, memory devices and computer peripherals in a local environment. High performance is achieved by effectively separating the high volume node-to-node data communications and the low-volume network control information. Data messages travel through point-to-point data links. Control messages, which set up data paths and coordinate network activities, use an independent multidrop control link which connects all devices. The fully distributed control mechanism allows a high degree of concurrent data transmission.>

Control algorithms for accurate and fast moving robots require computations based on the complete dynamic model of the robot. These computations are time consuming due to the high degree of coupling between robot links and the nonlinear dependence of the robot's dynamic characteristics on orientation of the links. A method for processing the dynamic control equations in real-time by using parallel computation is described. The recursive Newton-Euler equations are distributed over multiple computing elements, one for each joint. Concurrency is achieved by substituting "predicted" values for the actual values of variables involved in the recursive equations. The method has been simulated and compared to other approaches to this problem such as simplification of the dynamic equations. The experimental results have shown that the errors introduced by prediction are relatively small and compare very favorably to other techniques. The computer architecture and algorithms presented offer a practical way to implement the control algorithms with relatively inexpensive computing devices.