249 results about "Space partitioning" patented technology

The systems and methods of the invention are directed to portfolio optimization and related techniques. For example, the invention provides a method for multi-objective portfolio optimization for use in investment decisions based on competing objectives and a plurality of constraints constituting a portfolio problem in a space, the method comprising: generating a set of solutions of portfolio allocations in the space, the space having a plurality of dimensions; selecting a first dimension from the plurality of dimensions, the first dimension being a dimension under consideration; dividing the space into bins based on each dimension in the plurality of dimensions other than the dimension under consideration; determining a respective point in each bin with the most extreme value in the dimension under consideration; determining, based on the point in each bin with the most extreme value, whether other points in the space are dominant or dominated; and removing the dominated points from further consideration, so as to result in a reduced set of solutions, the reduced set of solutions being used in investment decisioning.

A system, method, and computer program product to automatically eliminate the distribution information available for reconstruction from a disguised dataset. The invention flattens input numerical values into a substantially uniformly distributed dataset, then maps the uniformly distributed dataset into equivalent data in a target distribution. The invention allows the incremental encryption of new values in an encrypted database while leaving existing encrypted values unchanged. The flattening comprises (1) partitioning, (2) mapping, and (3) saving auxiliary information about the data processing, which is encrypted and not updated. The partitioning is MDL based, and includes a growth phase for dividing a space into fine partitions and a prune phase for merging some partitions together.

An apparatus and method of processing object data (e.g., a scene having one or more objects) logically divide the object space into a plurality of contiguous (or noncontiguous) portions. At least two portions each have object data for such object(s). The apparatus and method also determine a positional priority of the at least two portions of the object space for a given viewpoint. Next, the apparatus and method merge rendered object data after at least some of the object data is rendered. This merging is based upon the positional priority of the at least two portions.

An apparatus and a method for detecting from an image a particular subject corresponding to multiple views of the subject by dividing a particular subject space into a plurality of subject subspaces and further dividing a subject subspace into subject subspaces representing multiple views; configuring a tree-structured detector wherein the tree structure has a root node that covers all subject subspaces and has a plurality of branches, each branch corresponding to a child node that covers at least one subject subspace; training each node to determine which nodes in the adjacent lower layer the images of the subject in the corresponding nodes should be sent.

A method for partitioning a memory mass storage device is disclosed. The partition task is performed by the controller within the memory mass storage device. Firstly, the controller partitions the logical space of the memory storage device into multiple areas, each area belonging to a particular drive. Secondly, the controller partitions the logical space of the memory storage device into a public area and a security area, both areas belonging to the same drive. Finally, the controller partitions the logical space of the memory storage device into multiple areas, which include public areas and security areas and belong to multiple drives.

The invention provides a bounding box and space partitioning-based virtual object collision detection method. The method includes the following steps that: virtual object collision pre-detection is performed on two irregular virtual objects; region segmentation is performed on a region to be detected; intersection testing is performed in the sub regions segmented from the region to be detected; virtual object collision detection is performed by using a point vector set representing a moving virtual object and a triangular surface representing a virtual object which does not require assembly at present; and if the virtual objects intersect with each other, the virtual objects collide with each other, otherwise, the virtual objects do not collide with each other. According to the bounding box and space partitioning-based virtual object collision detection method of the invention, a collision detection range of a space is narrowed through using the spatial correlation of the virtual objects, so that time consumption can be reduced, and at the same time, the detection efficiency of the method and the geometric accuracy of collision detection are greatly improved; a bounding box is reduced into the triangular surface and the points, so that the misjudgment of the collision detection can be decreased; and a collision detection process is refined to the interference between the triangular surface and the points, and separate detection is carried out, and therefore, detection efficiency can be greatly improved.

An error detection and correction system extracts editing patterns and derives correction rules from them by observing differences between draft documents and corresponding edited documents, and/or by observing editing operations performed on the draft documents to produce the edited documents. The system develops classifiers that partition the space of all possible contexts into equivalence classes and assigns one or more correction rules to each such class). Once the system has been trained, it may be used to detect and (optionally) correct errors in new draft documents. When presented with a draft document, the system identifies first content (e.g., text) in the draft document and identifies a context of the first content. The system identifies a correction rule based on the first content and the first context. The system may use a classifier to identify the correction rule. The system applies the correction rule to the first content to produce second content.

A system and method for producing a selectable view of an object space include: a) dividing the object space into n object sections to be imaged; b) providing at least n cameras, where the cameras are configured such that each object section is associated with at least one unique camera configured to image substantially only that object section; and c) imaging each of the object sections with its unique camera, so as to create at least one image of each object section, where the images of the object sections are combined to create a substantially continuous composite mosaic of the object space, where a view of a portion of the mosaic is selectably provided to a user based on selection instructions from the user, and where at least one of the view, the mosaic, and the images of the object sections is sent to the user via an information network, such as a cable network. The view may be provided in 3D to the user via a head-mounted display, and the selection instructions may include a physical orientation of the display.

A system and method for enabling highly scalable multi-node event distribution networks through the use of summary-based routing, particularly event distribution networks using a content-based publish/subscribe model to distribute information. By allowing event routers to use imprecise summaries of the subscriptions hosted by matcher nodes, an event router can eliminate itself as a bottleneck thus improving overall event distribution network throughput even though the use of imprecise summaries results in some false positive event traffic. False positive event traffic is reduced by using a filter set partitioning that provides for good subscription set locality at each matcher node, while at the same time avoiding overloading any one matcher node. Good subscription set locality is maintained by routing new subscriptions to a matcher node with a subscription summary that best covers the new subscription. Where event space partitioning is desirable, an over-partitioning scheme is described that enables load balancing without repartitioning.

A copying-type garbage collector operates in multiple concurrent threads. Each thread evacuates potentially reachable objects from the from space to the to space in a depth-first manner: if a thread has evacuated an object containing references to any from-space objects, it evacuates all of that object's descendants before it evacuates any other reachable objects. To keep track of descendants that must be evacuated before non-descendants can be, the thread places objects containing references to non-evacuated objects into a linked list maintained by pointers that it installs in the from-space locations from which the objects on the list were evacuated. Additionally, it divides the to space into local-allocation buffers (“LABs”) to which respective threads exclusively evacuate objects, and each thread maintains a LAB stack representing all the LABs it has filled that still contain references to unevacuated from-space objects. When a thread has completed evacuating the descendants of evacuees in all of its LABs, it “steals” work from other threads. It may do so, for instance, by processing a reference in an object belonging to another thread's list, by transferring to its own list one or more objects from another thread's list, or by transferring to its own LAB stack one or more LABs from another thread's LAB stack.

The invention discloses a storage protecting controller and a method for improving safety of an SOC (system on chip). The method includes the steps: dividing a program code space of a processor into a plurality of code spaces, and setting an entry address corresponding to each code space; dividing a storage space into a plurality of storage protecting areas, individually setting access authority attributes of the processor to the storage protecting areas when each code space executes program codes; judging whether a program pointer of the processor skips or not and whether skip is abnormal or not, and generating skip abnormal indication if the skip is abnormal; and monitoring whether access of a current bus to the storage space is abnormal or not, and stopping access of the bus if access of the current bus to the storage space is abnormal, so that the program pointer of the processor is prevented from skipping from one code space to another code space to execute a program, and abnormal access to the storage protecting areas is prevented. The storage protecting controller can effectively improve program code running safety, and can be widely applied to SOC chips of various types.

Separation of a color into values for colorants in colorant space can be done easily when color correction and color separation are done in a separate step. To perform color separation the color can be defined as a set of values in a trajectory space wherein trajectories are defined comprising points for which ink behavior during color separation is known. Colorant values for the color can be determined using simple interpolation techniques based upon values of the color located in trajectory space, known values and corresponding colorant values of the trajectory points. Trajectories are preferably color axes dividing the trajectory space into tetrahedra. Obtained colorant values may be further corrected using techniques such as colorant replacement or colorant splitting. A color correction step can be done previous to the color separation step. The combined process can be calibrated using a test target.

We disclose a transformation-based method for indexing high-dimensional data to support similarity search. The method, iDistance, partitions the data into clusters either based on some clustering strategies or simple data space partitioning strategies. The data in each cluster can be described based on their similarity with respect to a reference point, and hence they can be transformed into a single dimensional space based on such relative similarity. This allows us to index the data points using a B<+>-tree structure and perform similarity search using range search strategy. As such, the method is well suited for integration into existing DBMSs. We also study two data partitioning strategies, and several methods on selection of reference points. We conducted extensive experiments to evaluate iDistance, and our results demonstrate its effectiveness.

Limited range physical entities located in a topological space are dynamically grouped by partitioning the space into grids for different functional domains, a group being associated with each grid that corresponds to a unique functional domain and a unique topological space. A range of effectiveness for each entity is determined and compared with other groups. Each entity is then assigned to be a member of a particular group. The assigning step may comprise calculating a membership value for each entity, the membership value being equal to the ratio of the intersecting space to total space within a group.

A mobile phone is provided. The mobile phone includes a first case, a frame defining a space for elements of the mobile phone, the frame together with the first case dividing the space into a plurality of chambers, and a seal interposed between the frame and the first case for independently sealing more than one of the plurality of chambers.

Software-Defined Networking (“SDN”) enables flexible flow control by caching policy rules at OpenFlow switches. Compared with exact-match rule caching, wildcard rule caching can better preserve the flow table space at switches. However, one of the challenges for wildcard rule caching is the dependency between rules, which is generated by caching wildcard rules overlapped in field space with different priorities. Failure to handle the rule dependency may lead to wrong matching decisions for newly arrived flows, or may introduce high storage overhead in flow table memory. A wildcard rule caching system, which may be used for SDN partitions the field space into logical structures called buckets, and caches buckets along with all the associated wildcard rules. Doing so resolves rule dependency while using control network bandwidth efficiently. Further, controller processing load and flow setup latency are reduced.

The invention relates to a rapid network packet classification method based on network traffic statistic information, which belongs to the technical field of filtration and monitoring of network traffic. The method comprises the steps of: determining the space partition of various domains of a header and a rule mapping table thereof according to a classification rule set, and establishing an address lookup table, a port lookup table and a rule lookup table; recording the times that the header of a network packet appears in the space partition of the domains and calculating prior distribution; establishing a letter search tree of the domains according to the prior distribution; performing continuous matching on the received header of the network packet according to the letter search tree and the lookup tables; and re-calculating the prior distribution and updating the letter search tree at update time, and continuing to match the received network traffic. The method uses the classification rule set and heuristic information of the network traffic from two different levels of network packet classification, strengthens the adaptability of the classification method, and improves average classification efficiency. The method has quick lookup speed and strong adaptability, can be realized on a plurality of platforms, and is applicable to the filtration and monitoring of high-performance network traffic.

The invention discloses a nonlinear kernelled adaptive prediction method which comprises the following six steps: data preprocessing, subspace subdivision, subspace adaptive fitting control, subspace connection, new sample prediction and prediction output. The method is characterized in that after data is preprocessed, the integral space of the data is subdivided into a plurality of successive subspaces; the optimal kernels and parameters are adaptively selected on each subspace by using an intelligent sliding controller based on particle swarm support vector regression so as to form optimal local fitting hypersurfaces; then the optimal local fitting hypersurfaces are connected by using a three-point Lagrange interpolation method so as to form a final total-space regression prediction function; and finally, new data is predicted and output. By using the method disclosed by the invention, the characteristic space subdivision of data is realized, and the prediction speed for large-scale multi-attribute nonlinear data is improved; and the kernel functions adapting to the data distribution are screened adaptively by using the intelligent sliding controller based on particle swarm support vector regression, thereby optimizing integral parameters, and ensuring the precision and accuracy of fitting prediction.

The invention discloses a dynamic storage method of a Flash memory, which comprises the following steps: a. the Flash storage space is divided into a plurality of logical partitions, and the logical partitions can memory at least one type of data record; b. the storage space held by each piece of data record is taken as a virtual sector, and a virtual sector distribution list is arranged for each virtual sector; and c. the data record is read at one terminal of the logical partition in sequence, and the virtual sector distribution list corresponding to the data record is read at the other terminal of the logical partition in sequence; therefore, the dynamic storage method effectively solves the technical problem of low utilization rate of the Flash storage space in the static storage method.

The invention provides a bad block management method for a flash memory. The method comprises the following steps of: dividing the storage space of the flash memory into a first area comprising a plurality of blocks and a second area comprising a plurality of blocks; performing bad block detection on the flash memory to acquire information about bad blocks; establishing a mapping relation between the bad blocks in the first area and normal blocks in the second area according to the acquired information about the bad blocks, and generating a block mapping quick access table comprising the information about the established mapping relation; and respectively storing the generated block mapping quick access table into a plurality of bad block management information blocks in the second area, wherein a plurality of bad block management information blocks are any normal blocks which are positioned in the second area and do not have the mapping relation with the bad blocks in the first area.

A data input means inputs boundary data of an object to a computer, a data converting means converts the boundary data into a triangle patch having a phase, an associating means divides a space into rectangular parallelepiped cells having boundary planes intersecting perpendicularly and associates the cell with a triangle to be included in the cell, a dividing/arranging means divides a triangle patch having a phase and floating in the space at cell faces and keeps all triangles arranged within and on the boundaries of cells, a ridge line integrating means integrates ridges that do not alter the phase, a cell assigning means assigns each triangle and its vertex to a cell with reference to index data of the vertex, and a labeling means sets an attribute value of each cell.

The invention belongs to the technical field of computer graphics processing, and relates to the tridimensional visual model simplification technology. The tridimensional visual model simplification technology uses the method of combining octree and peak abstract of defined new peak weights. Firstly, a model space is divided into a series of leaf node zones through the octree, and all of peaks are dispersed in the zones; subsequently, the peak abstract algorithm with defined new peak weights is adopted for deleting the peaks, thus simplifying a three-dimensional model. By adopting the method of combining the octree and the peak abstract of defined new peak weights provided by the invention, the simplification quality and the simplification speed of the three-dimensional model are effectively improved.