Abstract: A system and method are described for providing least-cost call routing for mobile wireless devices having access to voice-over-IP (VoIP) infrastructure, such as may be operated by an enterprise having global presence. Using a local wireless connection, such as a wireless local area network (WLAN), a software agent on the mobile wireless device registers with a common, enterprise-wide registrar, which, in turn, informs a VoIP gateway in the device's home region of the device's new location. The home region VoIP gateway will then re-route incoming calls for the device to a VoIP gateway serving the region in which the device is currently located. The local VoIP gateway will then route the call to the device via the WLAN. Seamless routing occurs even when the wireless device obtains a local phone number in a foreign network. Outgoing calls from the device can be routed via the VoIP infrastructure or the foreign network based on availability of service and/or cost.

Abstract: A method and system for communicating the identity of devices detected in a wireless communication system. The devices connect to a mobile wireless network comprising a wireless access network and a core Internet Protocol (IP) network. The identities of devices so detected are specified by a combination of device and network-level identities and is communicated to a management server. In an exemplary embodiment, network elements of the core IP network are used to detect and convey the identity of wireless devices to a management server.

Abstract: In an internet protocol television (IPTV) system, fast channel change (FCC) transactions that experience late multicast joins are handled specially, thereby allowing the system to be provisioned for the vast majority of multicast joins that are within an expected range. During an FCC transaction, the subscriber interface device, such as a set-top box (STB), can determine how long it can wait to join the multicast stream of the new channel before a gap can be predicted to occur in the handover from the FCC unicast to the multicast. If the STB has not joined the multicast stream by the determined time, a recovery procedure is invoked in which the STB sends a RESTART request to the FCC server which, in turn, increases the rate of the FCC unicast as if the FCC transaction were restarted, thus allowing the handover to occur seamlessly. This solution reduces the demands of each FCC transaction, allowing for greater scalability and reduced cost.

Abstract: For wavelength division multiplexing (WDM) communications, apparatus and methods are provided for performing tone-based optical channel monitoring that is less sensitive to stimulated Raman scattering (SRS). In tone-based optical channel monitoring, in which WDM channels are modulated with one or more tones, detecting and measuring the tone power is commonly used as a measure of signal power in each channel. In WDM systems with long fiber spans and high signal powers, however, SRS tends to transfer energy from shorter wavelengths to longer wavelengths, whereby the tones are no longer accurately indicative of the signal power of the individual wavelength channels. Apparatus and methods are provided which reduce the effect of SRS by filtering the monitored WDM signals into sub-bands and detecting each sub-band independently.

Abstract: An optical apparatus including a 360-degree star coupler with derivative structure(s) and applications to optical imaging, optical communications and optical spectroscopy.

Abstract: A self-polarization diversity technique to combat PMD in a direct-detection optical OFDM system. This technique does not require any dynamic polarization control, and can simultaneous compensate PMD in a WDM system with one device. Simulation results show that this technique virtually completely eliminates the PMD impairments in direct-detection optical OFDM systems.

Abstract: Duobinary and NRZ modulation of an X-Gb/s optical signal is achieved with a lumped element InP Mach-Zehnder device configured to operate at X/k-Gb/s where k>1 and arranged in a push-pull configuration.

Abstract: Methods and apparatus are described for generating and receiving amplitude and differential-phase encoded signals in which the number of phase states at a given amplitude level is always less than or equal to that at a higher amplitude level and at least two amplitude levels have different numbers of phase states.

Abstract: A signal identification method for efficiently identifying Dense Wavelength Division Multiplexing (DWDM) channel modulation formats, signal powers, and center frequency detuning of signals that may be used in conjunction with telecom-grade monitoring equipment. The method utilizes least-square curve fitting estimates applied to a set of curves, each curve is characteristic of a modulation format and rate. For each least-square estimate an error value is calculated, and a curve fit with the least error is selected as the identified signal format for a signal.

Abstract: Methods and apparatus for aligning the transmitters of two or more bidirectional ports of an integrated circuit (IC), particularly an application-specific IC (ASIC) or field-programmable gate array (FPGA). Misalignment of two or more transmitters is determined by the IC itself without the use of external test equipment. Receivers of the bidirectional ports whose transmitters are to be aligned are used by the IC to detect misalignment. Any misalignment of the receivers is also determined and either eliminated or taken into account when aligning the associated transmitters. Variants for ICs with and without internal loop-back capability and for ICs with and without differential outputs are described.

Abstract: A low-profile real-time x-ray detector has a visual sensor that provides a wide angle visual image of a location on an inspected part that is being imaged by x-rays. Fish eye lenses in the visual sensor view a reflection of a part being inspected from a highly reflective surface mirror on an x-ray sensitive surface. The visual sensor may provide an exact visual replica of the area on the inspected part that is being imaged by x-rays.

Abstract: A method for accurately estimating a pose of the human head in natural scenes utilizing positions of the prominent facial features relative to the position of the head. A high-dimensional, randomly sparse representation of a human face, using a simplified facial feature model transforms a raw face image into sets of vectors representing the fits of the face to a random, sparse set of model configurations. The transformation collects salient features of the face image which are useful to estimate the pose, while suppressing irrelevant variations of face appearance. The relation between the sparse representation of the pose is learned using Support Vector Regression (SVR). The sparse representation, combined with the SVR learning is then used to estimate a pose of facial images.

Abstract: A wrapper organization and architecture for networks on a chip employing an optimized switch arrangement with virtual output queuing and a backpressure mechanism for congestion control.

Abstract: A fully automatic, computationally efficient segmentation method of video employing sequential clustering of sparse image features. Both edge and corner features of a video scene are employed to capture an outline of foreground objects and the feature clustering is built on motion models which work on any type of object and moving/static camera in which two motion layers are assumed due to camera and/or foreground and the depth difference between the foreground and background. Sequential linear regression is applied to the sequences and the instantaneous replacements of image features in order to compute affine motion parameters for foreground and background layers and consider temporal smoothness simultaneously. The Foreground layer is then extracted based upon sparse feature clustering which is time efficient and refined incrementally using Kalman filtering.

Abstract: A reusable device for protecting spooled materials such as wire, cable, fiber or others during their transport, storage, and warehousing. The device is wrapped circumferentially around the spooled material, secured, and is sufficiently impact resistant to prevent unintended damage to the spooled material. Keepers, disposed lengthwise along the device, serve to prevent the spooled material from becoming unwound along the ends of the spool. An external surface provides a surface for advertising or other written information.

Abstract: A method for human face detection that detects faces independently of their particular poses and simultaneously estimates those poses. Our method exhibits an immunity to variations in skin color, eyeglasses, facial hair, lighting, scale and facial expressions, and others. In operation, we train a convolutional neural network to map face images to points on a face manifold, and non-face images to points far away from that manifold, wherein that manifold is parameterized by facial pose. Conceptually, we view a pose parameter as a latent variable, which may be inferred through an energy-minimization process. To train systems based upon our inventive method, we derive a new type of discriminative loss function that is tailored to such detection tasks. Our method enables a multi-view detector that can detect faces in a variety of poses, for example, looking left or right (yaw axis), up or down (pitch axis), or tilting left or right (roll axis).

Abstract: Dispersion compensator apparatus comprising 1XM, MXN and NXN couplers, where the coupling ratios of the MXN and NXN couplers are selected such that the dispersion compensator provides a desired amount of dispersion compensation.