A novel approach for 3D object recognition is presented. This approach is model-based, and assumes either 3D or 2/sup 1///sub 2/D scene acquisition. Transformation detection is accomplished along with an object identification (six degrees of freedom, three rotational and three translational, are assumed). The diffusion-like simulation recently introduced as a means for characterization of shape is used in the extraction of point features. The point features represent regions on the object's surface that are extreme in curvature (i.e. concavities and convexities). Object matching is carried out by examining the correspondence between the object's set of point features and the model's set of point features, using an alignment strategy. Triangles are constructed between all possible triples of object's point features, and then are aligned to candidate corresponding triangles of the model's point features. 2/sup 1///sub 2/ range images are transformed into a volumetric representation through a parallel projection onto the 3-D space. The resultant volume is suitable for processing by the diffusion-like simulation.>

A least upper bound is derived on the amount of adjustment of virtual token passing (VTP) time needed to assure collision-free access control in VTP networks (i.e. networks that use 'time-out' or scheduling function-based access protocols) and in which nodes change their spatial configuration due to motion (although always stay within range and in line-of-sight of each other). Since the new bound is a function of network geographical size, as well as of node maximal speeds and the time that passed since the previous adjustment, it allows VTP times that are shorter than those found in previous publications, especially when intervals between adjustments grow larger. For most VTP networks, with large mixed populations of mobile and stationary users, the average VTP time (which is a major factor in performance of the access protocol) allowed by the new bound is shorter than that of any configuration-independent protocol.>

We introduce a method for broadcasting control information (such as the information essential for correct operation of SOSAM and other scheduling-function-based access protocols) in stationary networks with "hidden" nodes (multihop networks). Control information is transmitted as short bit- parallel control messages on a separate control channel whose capacity is shared among the bits of a control message using code division multiplexing (CDM). The CDM method takes advantage of spread-spectrum signal properties that allow, in particular, high accuracy of time-of-arrival measurement and relatively easy separation of multipath copies of a control message.

We describe a distributed method for keeping down “token-passing” overhead in a mobile multi-access network, by performing local corrections in the token-passing sequence as they become necessary due to changes in node spatial configuration. These corrections involve only a subset of the nodes in the network, thus reducing the required computational effort. The method consists of three distributed protocols. The first is for detecting deterioration and identifying a subset of nodes that constitutes a “problem area”, the second is to provide each node with the distances (propagation delays) to all other nodes, and the third is to use this topological information to construct a minimal spanning tree, from which a good “token passing” sequence can be derived. The second and third protocols may also have applications other than the one described here, such as position location, routing and broadcasting.

We present a theoretical description and analysis of a collision-free channel-access protocol for a shared channel with "mobile" nodes that are all within range and in line-of-sight of each other, in arbitrarily changing spatial (one-, two-or three-dimensional) configurations.