Abstract: Systems and methods for locating the shooter of supersonic projectiles are described. The system uses at least five, preferably seven, spaced acoustic sensors. Sensor signals are detected for shockwaves and muzzle blast, wherein muzzle blast detection can be either incomplete coming from less than 4 sensor channels, or inconclusive due to lack of signal strength. Shooter range can be determined by an iterative computation and/or a genetic algorithm by minimizing a cost function that includes timing information from both shockwave and muzzle signal channels. Disambiguation is significantly improved over shockwave-only measurements.

Abstract: A quantum cryptographic device provides authentication services over the optical (quantum) channel and the public channel. In one implementation, polarizers generate optical pulses that have a polarization state based on a bit from a first bit sequence. A polarization rotator further rotates the polarization basis of the optical pulse by a rotation angle specified by one or more bits of a second bit sequence. A receiving device receives the modulated optical pulses, demodulates the pulses, and may determine whether the optical channel can be authenticated. In an alternate implementation, phase modulation, instead of polarization modulation, is used to similarly modulate the optical pulses.

Abstract: A method, apparatus, and article of manufacture for reducing broadcast traffic in a wireless backbone network. In operation, a wireless access point receives an address resolution protocol (ARP) message from a network device. The wireless access point determines the destination of the ARP message and attempts to identify the Ethernet address associated with the destination network device. If the Ethernet address cannot be determined, the access point sends an ARP request message to the other wireless access points on the network, requesting them to transmit ARP packets to the network device. When the desired network device comes within range of the network, the access points discontinue transmitting ARP packets and a message acknowledging the device's presence is transmitted to the original access point. This method of having access points periodically broadcast ARP packets to network devices reduces ARP broadcasting across the wireless backbone, thereby increasing network throughput.

Abstract: A method and apparatus for building a training set for an automated speech recognition-based system, which determines the statistically optimal number of frequently requested responses to automate in order to achieve a desired automation rate. The invention may be used to select the appropriate tokens and responses to train the system and to achieve a desired “phrase coverage” for all of the many different ways human beings may phrase a request that calls for one of a plurality of frequently-requested responses. The invention also determines the statistically optimal number of tokens (spoken requests) required to train a speech recognition-based system to achieve the desired phrase coverage and optimal allocation of tokens over the set of responses that are to be automated.

Abstract: A real-time signal processing engine robustly detects, localizes, tracks and classifies ground targets based on radar signals from a multistatic radar system. The system differentiates between different targets based on an optimized cost function, which can include the total returned normalized pulse energy. The local transmitters/receivers can communicate with each other via the transmitted radar signals.

Abstract: Methods are described which enable the operator of a network such as an Internet protocol (IP) network to allocate ingress and egress channel capacity for the use by different services or for use by different service providers. The methods can prevent a service provider or a service from overusing network capacity. The methods also enable the network operator to temporarily allocate unused capacity if demand for a service exceeds previously allocated capacity. Further a method is described whereby capacity on a Voice over Internet protocol (VoIP) network is sold to service providers with guarantees for the number of available channels, where the channel capacity guarantees could be global, or by any type of subdivision. In addition to having a committed channel capacity, the service provider could be allowed to exceed the committed capacity whenever excess capacity is available on the network.

Abstract: The invention relates to a transport protocol and associated methods and stack architectures for improving the energy efficiency of transmitting packets through an ad hoc network. The protocol controls transmissions by taking into account per-packet energy limits, per-node loss tolerances, and/or minimum availability rates determined based on path quality measurements collected by packets traversing the network and application reliability requirements associated with various applications.

Abstract: A speech recognition system receives an audio signal and detects various features of the audio signal. For example, the system classifies the audio signal into speech and non-speech portions, genders of speakers corresponding to the speech portions, and channel bandwidths used by the speakers. The system detects speaker turns based on changes in the speakers and assigns labels to the speaker turns. The system verifies the genders of the speakers and the channel bandwidths used by the speakers and identifies one or more languages associated with the audio signal. The system recognizes the speech portions of the audio signal based on the various features of the audio signal.

Abstract: The invention relates to communications devices for reduced energy communications in an ad hoc network. The communication device includes a first low powered transceiver for initiating communications with other communications devices and a second transceiver for transmitting data messages to the other communications devices once communication is initiated. The communication device also includes a communications control processor for determining times at which the other communications devices will be available to receive communications based on scheduling data received from those communication devices.

Abstract: A system (126-129) detects transmission of potentially malicious packets. The system (126-129) receives packets and generates hash values corresponding to each of the packets. The system (126-129) may then compare the generated hash values to hash values corresponding to prior packets. The system (126-129) determines that one of the packets is a potentially malicious packet when the generated hash value corresponding to the one packet matches one of the hash values corresponding to one of the prior packets and the one prior packet was received within a predetermined amount of time of the one packet. The system (126-129) may also facilitate the tracing of the path taken by a potentially malicious packet. In this case, the system (126-129) may receive a message that identifies a potentially malicious packet, generate hash values from the potentially malicious packet, and determine whether one or more of the generated hash values match hash values corresponding to previously-received packets.

Abstract: Genetically adaptive neural network systems and methods provide environmentally adaptable classification algorithms for use, among other things, in multi-static active sonar classification. Classification training occurs in-situ with data acquired at the onset of data collection to improve the classification of sonar energy detections in difficult littoral environments. Accordingly, in-situ training sets are developed while the training process is supervised and refined. Candidate weights vectors evolve through genetic-based search procedures, and the fitness of candidate weight vectors is evaluated. Feature vectors of interest may be classified using multiple neural networks and statistical averaging techniques to provide accurate and reliable signal classification.

Abstract: Methods and systems are disclosed in which information is securely transmitted in a network comprising untrusted network devices. Setup messages are sent to a networking device, and based on the setup messages, light directing devices are configured to direct light along a path from an origin endpoint to a terminal endpoint, thus providing a path through the network. Through the path, a stream of light information is sent using a plurality of light pulses to carry out quantum-cryptographic key distribution.

Abstract: A method (508) and system (226) provide directional medium access control for wireless nodes (102) in a wireless network (100), each of the nodes (102) having a transceiver (228) and a plurality of antennas (232, 234, 236, 238, 240) in selective communication with the transceiver (228). Packets (202) are transmitted from a transmitting node (106) of the nodes (102) when a channel (104) is sensed as being free in a beam direction of a selected one of the antennas (232, 234, 236, 238, 240), and in accordance with a pre-determined transmission mode (304) for the data packet (202). The transmitting node (106) is periodically placed in a forced idle state, thereby being prevented from transmitting data packets (202), but being able to receive and respond to data packets (202) from other nodes (102).

Abstract: A network processor [200] performs Cyclic Redundancy Check (CRC) operations using specialized hardware circuits [308-308]. The network processor [200] includes a plurality of hardwired CRC polynomials that are used to implement the CRC operations. A CRC instruction selects which polynomial to use when performing the CRC operation.

Abstract: A method of determining routing information among nodes [210] in a network [200] may include obtaining a set of times of arrival [410] and corresponding source nodes [420] for chunks of data in the network [200] and assigning weights to a set of chunks of data [520-550] based on respective times between the set of chunks of data [520-550] and one chunk of data [510]. A set of routing information [600/800] among the nodes in the network [200] may be updated based on the assigned weights.

Abstract: A method, a device and a system for communications to and from a retro-reflector device (302) is provided. The retro-reflector device (302) receives a first frame (400) encoded in an input beam (106). The retro-reflector device (302) creates and sends a second frame (420) in a first reflected beam (108) formed by the retro-reflector device (302) reflecting the input beam (106) along a path closely aligned with a path of the input beam. At least one of the first frame (400) and the second frame (420) includes medium access control information. In some implementations, the first frame (400) may include a data throughput rate (404, 406), a preamble (402) and an error correction code (412).

Abstract: A data-display system employs a display in which the representations of data objects are caused to move on the display in order to convey information about the represented data objects. In one example, icons in a link-analysis display that represent data objects satisfying a selection criterion are made to execute distinctive motion. In another example, three-dimensional models of moving bodies in whose features components of respective data objects are encoded are projected onto a screen plane, and the resultant values are used to generate the display.

Abstract: A method (38) and system (56) evaluate the accuracy of an Automatic Location Identification (ALI) system (36) deployed within an environment (20) and configured to location a wireless communication device (24) originating an emergency call (22) through a wireless communication network (26). The method (38) includes a subprocesses that identify a validation region (144) in which a service area (106) of a public service answering point (PSAP) (32) and an RF coverage area (129) overlap, classify sub-regions within the validation region (144) according to a predetermined set of test scenarios (148) representing unique calling environments, and select test points (200) within the validation region (144) from which test calls, that simulate emergency calls, will be performed.

Abstract: A system protects against loss of communication during network attacks. In a first implementation, a system (120) models the behavior of normal users in a network in response to an application of a first packet filtering technique. The system (120) receives a group of packets from a first user subsequent to the application of the first packet filtering technique and creates one or more models reflecting the behavior of the first user based on the received packets. In another implementation, a system (130) receives a stream of packets subsequent to a filtering technique being applied, partitions the packets into groups, where each group corresponds to more than one packet, and classifies each group of packets as a normal group or an attack group using one or more models. Each model reflects a normal response to an application of the filtering technique. The system (130) forwards groups classified as normal groups, thus preventing network attacks from choking off all communication in the network.

Abstract: A method for operating a user communication device (18a, 18b, 19a, 19b), and a program and user communication device that operate in accordance with that method. An interface (2d, 2b, 23, 22) of the communication device (18a, 18b, 19a, 19b) is operated to enter identifiers identifying respective calling sources (18a, 18b, 19a, 19b) from which call signals may be received. The interface also is operated for entering into the device (18a, 18b, 19a, 19b) electrical signals representing corresponding audible signals that are to be individually generated in response to calls being received from the respective calling sources (18a, 18b, 19a, 19b). The identifiers are stored in a memory (2c, 24) in association with respective ones of the electrical signals.