Abstract: A technique is presented for compressing data which leverages the frequency of an escape symbol for better compression. The prefix of a data string is evaluated and the probability of all characters that might succeed it is predicted in tabular form. Symbols are designated “Hit” or “Miss” based upon whether they are in the table. A binary tree is generated by partitioning nodes into Zero and One groups based on a single bit value. A partition bit is chosen to maximize the difference of probability sums of Hit symbols in Zero and One groups, with exceptions for partitions having non Hit symbols in one of the groups. A probability value is assigned to each branch, based on the probabilities of Hit and Miss symbols. Encoding or decoding a symbol is facilitated by encoding or decoding the branch probabilities on the shortest path from the root to the leaf node containing the symbol using arithmetic encoding or decoding method.

Abstract: The present invention relates to a data processing system and method and, more particularly, to a software TDM bus emulator which emulates the operation of a TDM bus. Often computer aided telephony voice processing applications that require specialised processing services such as DTMF tone detection, facsimile service and voice recognition or synthesis services. Typically, these services are provided via dedicated hardware digital signal processors that communicate using a SONET bus. However, not all computer aided telephony servers have the full range of such specialised hardware resources. Accordingly, the present invention provides a data processing method in which a TDM bus is emulated to periodically execute, using system resources, a number of software data processing resources, which communicate with each other to form a data processing chain. The results of the data processing chain are ultimately returned to the voice application.

Abstract: A method and apparatus for detection of a port name in a loop network is provided, particularly a loop network in the form of a Fibre Channel Arbitrated Loop (FC-AL). The loop network (100) has a plurality of devices (120) each device (120) having at least one port (211, 212) on the loop network (100). The method includes determining which ports (211, 212) are populated with devices (120) for which the unique port name (WWPN) is not known. The populated ports are then all bypassed and a mode is entered on the loop network (100) in which idle frames are transmitted around the loop network (100). One port is un-bypassed at a time and the port name from the un-bypassed port is received and recorded. The port name is received from the un-bypassed port in a Loop Initialisation Select Master (LISM) frame transmitted by the un-bypassed port.

Abstract: A loop network hub including a hub port with a loop initialization insertion mechanism. The loop initialization insertion mechanism causes a hub port which detects a new node port connection to automatically begin generating loop initialization data. A hub port continues to generate loop initialization data until that hub port receives a loop initialization sequence. The loop initialization data propagates around the loop of the hub, halting ordinary processing. In this way, the entire loop is cleared. Upon receiving a loop initialization sequence, the hub port originating the loop initialization data inserts the new node port into the loop. At this point, loop initialization begins and each node port in the loop network obtains a unique loop network address.

Abstract: A method for distributing participants of a conference handled by a first conference unit in a packet switched network is disclosed. This is provided by, when a first condition is reached, locating a second conference unit satisfying a second condition, reserving conference call capacity in said second conference unit, transferring one or more of said participants from the first to the second conference unit and preferably entering said first conference unit as a participant in said second conference unit, wherein data streams of possible conference participants still attached to the first conference unit are mixed in the first conference unit and communicated as one data stream to the second conference unit. The invention is particularly useful when applied to the SIP protocol in a 3GPP network.

Abstract: A method of scheduling transmissions in an ad-hoc wireless network is invoked when a frame is received by a medium access control layer of a receiving node. The method includes determining a next node to transmit based on a schedule and a bandwidth allocation. When the next node is different from a node that sent the received frame, the current burst is terminated and the next node is invited to transmit the next burst. When the next node is the node that sent the received frame, the current burst is allowed to continue to be transmitted. Such an approach is suitable for use in applications that require transmissions with a specified quality of service (QOS).

Abstract: A network management architecture is disclosed having a network backbone and one or more macro service areas. Each macro service area includes one or more basic service areas, which each include an access point and a plurality of associated clients. A management system is employed for connecting the access points to the network backbone. The management system regulates access of the clients to the network so as to control a desired network parameter, which can relate to throughput, quality of service, or any other desired network parameter.

Abstract: The present invention provides a data transfer method for use in a CDMA system operating in the TDD mode. A data signal is transmitted in the form of a data stream of data bursts of different users between a first station and a second station via a multiplicity of antennas, and a reference signal is transmitted in connection with the specific data burst. In a first step, a reference signal is transmitted from the second station to the first station. In a second step, a channel estimation is performed in the first station based on the reference signal that is received from the second station. In a third step, the first station transmits a data burst to the second station via the multiplicity of antennas, it being possible to modify the transmission signals of various antennas having various amplitude and phase factors in accordance with the channel estimation.

Abstract: The present invention comprises aggregating a plurality of physical channels to a single logical channel, a scalable virtual channel (SVC). This may be implemented in a point to multipoint communications system. The present invention enables a variety of receivers, some receivers being able to receive only one physical channel, and other receivers being able to receive a plurality of physical channels as a SVC, with at least one same channel being used for both types of receivers.

Abstract: A mobile device is equipped with components to facilitate communication with another user, as well as controlling pointing on a computing device proximally disposed adjacent to the mobile device. The communication may be audio or textual. The mobile device includes a movement mechanism to generate movement signals, which are processed into pointing control signals and transmitted to the proximally disposed computing device. The movement mechanism may be an electro-mechanical mechanism, an optical-electro-mechanical mechanism, or an optical mechanism, integrated with the device, or disposed on an interchangeable cover removably mated with a core body of the mobile device. The pointing control signals may be transmitted as radio frequency signal employing a radio frequency signal transmitter, or infrared signal employing an infrared signal transmitter. Similarly, the transmitter may be disposed on an interchangeable cover removably mated with a core body of the mobile device.

Abstract: Control information, which is transmitted through a circuit different from the one used for the transmission of a plurality of pieces of information to be multiplexed, is analyzed. When the analysis results in a determined intra-TTI data length of information having longer Transmission Time Interval, it is determined whether a range of values that TFCI may take is limited or not. If the determination provides a positive result, the intra-TTI data length of information having a longest Transmission Time Interval which has been determined by the latest TFCI decoding is used as a base for limiting TFCI candidates to be decoded next, so as to decode the TFCI.

Abstract: There is provided a supplemental channel assigning apparatus and method in a mobile communication system. In the supplemental channel assigning method, a mobile station receives a plurality of supplemental channel assignment messages successively from a base station on an existing traffic channel, each of the channel assignment messages having the fields of a start time, a duration, a sequence number for message identification, and a channel identifier for channel identification, and stores the received channel assignment messages in a memory according to the durations and sequence numbers of the channel assignment messages.

Abstract: The present invention reduces network operator memory requirements, administrative overhead, and operational costs by enabling a solution for SS7/IP communication that uses a single domain (e164.arpa) in ENUM for numbering schemes such as E.164, E.212, and E.214. Employing the present invention enables the SUA to deliver SCCP-user messages to the destination node using Global Title Information based on IMSI or E.164 numbers. Network operators populate ENUM databases with MAP URIs associated with mobility services such as MSC, HLR, and VLR. End point service node IP addresses, associated with a set of services, are stored in ENUM corresponding to the SAPC belonging to the local operator for a given PLMN. These end node IP addresses then are returned in the ENUM/DNS response to an E.164-based query.

Abstract: In a complete network-in-a-box system acting as an enterprise network demarcation point, packets such as voice, video and data packets, are routed over common network connections, such as LAN and WAN. The packets are mapped from a public address field (such as an IP address) and port number to a private address field and port number, the mapping process typically being handled by a NAT (Network Address Translation). The packets are also prioritized, by marking the packets for priority queuing and routing, and configuring the bandwidths of the WAN traffic and the voice traffic to predetermined quantities and configuring the address fields of the voice devices. Simultaneous transmission of the various packets can be limited to predetermined quantities, typically by utilizing a CAC (Client Access Control).

Abstract: A recording system for recording calls between callers and agents of a transaction processing system includes an Internet Protocol (IP) gateway, an interface configured to operatively couple the IP gateway to the transaction processing system, a communication network operatively coupled to the IP gateway, where the IP gateway is configured to convert a call received from the transaction processing system into IP protocol data, and provide the IP protocol data to the communication network. Also included is a memory storage system operatively coupled to the processor and configured to store the IP protocol data, a processor operatively coupled to the communication network and configured to receive the IP protocol data, and a switch controlled by the agent and settable between a record-on mode and record-off mode such that the recording system records the call if the switch is set to the record-on mode and does not record the call if the switch is set to the record-off mode.

Abstract: Disclosed is a system and a methodology enabling a first provider party to provide to a second recipient party software and services to enable the recipient party to embed in an Internet e-mail of the recipient party directed to an addressee, an HTML hyperlink file which includes a telecommunication initiating trigger, sending the e-mail to the addressee via the Internet, the addressee upon receiving the e-mail actuating the hyperlink via its trigger and, responsive to the trigger actuation, connecting the addressee via the Internet to a web page located on an Internet server, wherein the web page includes a request for a telephone number of the addressee.

Abstract: The invention relates to a signaling method for the transmission of signaling information within the frame of a voice transmission or multimedia transmission via an IP network of the SIP-T type, which is connected to at least one public telecommunication network, whereby a first ISUP message, coded in a national format and a second ISUP message, coded in the international ITU-T format, are transmitted with the SIP-T messages.

Abstract: A module and kit for coupling at least one telephone service signal to at least one telephone device over a wiring. The network includes: a wiring having at least two conductors for carrying multiple time-domain multiplexed digitized voice channels; an exchange side device coupled to the wiring and operative to couple at least one telephone service signal to at least one digitized voice channel; and at least one subscriber side device coupled to the wiring and operative to couple the at least one telephone device to at least one digitized voice channel.

Abstract: A broadband packet switch that handles all packets that arrive within the same frame time by simultaneously bit-pipelining the packets into different sections of the one-stop packet buffer through an input switch. Each packet remains in is selected section until its eventual exit from the buffer through an output switch. Access to the memory storage in which the packet buffer resides is not through a memory bus or buses, thereby engendering massive parallel access.

Abstract: A content addressable merged queue (camQ) architecture for high-speed switch fabrics reduces the memory requirement for crossbar switch input and output queues using memory cells and age tag comparators. CamQ emulates VOQ FIFO for each supporting priority, eliminating HOL blocking. Multiple QoS levels are supported cost effectively at higher traffic bandwidth limits. Content addressable memory (CAM) cells store payload destinations, which can be addressed by cell priorities. Once a priority for QoS is decided, all the cells with the selected priority in the payload can make connection requests to destination ports directly through the CAM structure. An age tag is assigned to incoming cells and fast age tag comparators provide FCFS features by selecting the oldest cell. Small memory sizes prevent the bottlenecking in ingress and egress queues. A CIOQ crossbar has a fast switching speed, emulating a FIFO output queue switch. Age and priority are interleaved to schedule switching.

Abstract: An apparatus and method for switching data packet flows by assigning schedules to guaranteed delay and bandwidth traffic. Scheduled bandwidth is subtracted from the available bandwidth, and the remaining unscheduled bandwidth is available for standard “best-efforts” packet transmissions, or other guaranteed packet flows. Guaranteed bandwidth is assigned on a schedule basis, and packet flows are assigned schedules which are mutually acceptable to both transmitters and receivers. Flows transmit and expect to receive packets in accordance with their assigned schedules. Schedules are negotiated between packet transmitter applications and packet receiver applications. During scheduled periods, packets sent by the packet transmitter are directly and immediately forwarded to the assigned packet receiver.

Abstract: A node includes input ports that are configured to receive packets from other nodes or from devices coupled to the node and output ports that are configured to send packets to other nodes or devices. Scheduling logic may control how packets that are received via the input ports are routed to the output ports. An input port may receive a packet that has multiple destinations or recipients. The scheduling logic may be configured to route this multi-destination packet to at least one of the output ports if at least one of the packet's recipients has an input buffer available to receive the packet.

Abstract: A hardware search engine facility is provided to allow CPU search and update of a Forwarding Table CAM under the control of software running on the CPU. The hardware search engine provides one or more comparand-mask pairs which allow for a match, exclusion or magnitude comparison on specific entry values and/or the option to ignore or “don't care” certain bits of the entry. Control registers may be set in software to specify a start address and stop address in the CAM for the search. An indication of valid or invalid entries may be provided as well. Once the search is initiated by software, the search engine will read the entries sequentially starting from the programmed start address. It will perform a compare using the comparand-mask pair and attempt to identify a match. The locations in the CAM which match the search criteria may be put into a CPU-accessible memory. If the memory fills up before it can be read by the software, the search may be halted until the memory is emptied.

Abstract: A cross stack port aggregation method and system associates a destination index with a received packet when it is provided to devices in the stack. Each device utilizes the destination index to access a descriptor identifying ports in the device included in the port aggregation group. An index generated from packet address data is used to select a bit in a group mask unique to each port.

Abstract: A method and apparatus for processing Ethernet data frames in a media access control (MAC) sublayer of an Ethernet passive optical network (PON) are provided.

Abstract: A high-speed router and method for operation of the core of such a router are disclosed. The disclosure describes routing packets from core input ports to core output ports by aggregating or queuing packets at router core ingress ports in queues designated for common router core egress ports. A scheduler selects a set of queues, up to one per ingress port, for switching through the router core at each epoch (an epoch is a time slice). When the epoch for a given set of queues arrives, data from each queue is stranded, with one strand sent to each of multiple switch fabric cards. The switch fabric cards operate in parallel to switch the strands from that queue to a common egress port (as configured by the scheduler), where the strands are recombined to reconstruct the original queue data. This architecture can be made fault tolerant, can be made to degrade gracefully when one or more switch fabric cards goes down, and can support increased traffic simply by expanding the number of switch fabric cards.

Abstract: An advanced data structure allows lookup based upon the most significant 16 bits and the following variable number of K bits of the IP destination address. This 16/K scheme requires less than 2 MB memory to store the whole routing tables of present day backbone routers. A 16/Kc version utilizes bitmaps to compress the table to less than 0.5 MB. For the 16/K data structure each route lookup requires at most 2 memory accesses while the 16/Kc requires at most 3 memory accesses. By configuring the processor properly and developing a few customized instructions to accelerate route lookup, one can achieve 85 million lookups per second (MLPS) in the typical case with the processor running at 200 MHz. Further, the lookup method can be implemented using pipelining techniques to perform three lookups for three incoming packets simultaneously. Using such techniques, 100 MLPS performance can be achieved.

Abstract: A process of filtering packet data in a network device is disclosed. A descriptor table is accessed to obtain a first descriptor and a first field is extracted from a packet based on the first descriptor. An action table is accessed based on a value of the first field of the packet. Then an action from the action table is executed when a valid bit from the action table for the action is set or an action from a miss action table is executed when a valid bit from the action table for the action is not set.

Abstract: A method is provided for determining a global mask value for filtering incoming messages in a CAN environment having a predetermined number of match registers, where the global mask value has one or more bits. A set of incoming messages to be identified is established. Each of the one or more bits in the global mask value is set to a first digital value. The following steps are repeated until the set of incoming messages to be identified fits within a predetermined number of match registers: adding a don't_care bit to the global mask value and determining a best position for the added don't_care bit. The don't_care bit is represented by a second digital value in the best position.

Abstract: A router for supporting differentiated qualities of service (QoS), and a fast Internet protocol (IP) packet classification method performed in the router, are provided. The router hierarchically divides a memory having flow tables into a fast internal cache memory and an external typical memory. The internal cache memory stores recently searched table entries. The router preferentially searches the flow tables of the internal cache memory. Only when the internal cache memory searching fails, the flow tables of the external memory are searched. Consequently, the frequency of interactions between a packet classifier and the external memory decreases to improve the speed of packet classification.

Abstract: Systems and methods for cell based wrapped wave front arbiter (WWFA) with bandwidth reservation are disclosed. The method for reserving bandwidth of a given priority using the WWFA for arbitrating bandwidth among virtual links between input and output ports, each virtual link supporting one or more priorities and corresponding to an arbitration unit (AU) of the WWFA generally comprises performing at least one arbitration pass of a wave front of the WWFA where AUs having a reserved bandwidth request of a given priority and reserved bandwidth credit of the given priority compete for the bandwidth associated with the wave front, and where AUs not having a reserved bandwidth request of the given priority and reserved bandwidth credit of the given priority do not compete for the bandwidth, and performing at least one subsequent arbitration pass where AUs having a reserved bandwidth request of the given priority compete for the bandwidth.

Abstract: The invention provides a chassis, a scalable router that includes a plurality of chassis and a method of upgrading a router. Each chassis includes a plurality of processing modules and a programmable interconnection module. Data connections are provided between each processing module on each chassis and the interconnection module on that chassis, and a data connection is provided between the interconnection module on each chassis and the interconnection module on at least one other chassis. In this way, the inter-chassis and intra-chassis connections pass through the programmable interconnection module, which may be used for concentrating the location of resources, such as opto-electronic and electro-optical converters, in a single card or unit. Moreover, the programmable nature of the switch fabric in each interconnection module is amenable to reconfiguration so as to support a change in that chassis' interconnection pattern that may be required for expanding the router's capacity.

Abstract: A network unit comprises a multiplicity of ports for transmitting and receiving data packets, a bridging facility and a routing facility. The network unit has a multiplicity of mutually exclusive states which includes an initial state from which said network unit can transition to either a slave state; or a waiting state from which in the absence of receipt from elsewhere of a broadcast advertisement including an advertised address corresponding to an address of the unit, the unit transitions to a master routing state in which the routing facility is enabled and the unit broadcasts a corresponding advertisement including an address of the unit. The unit transitions from the waiting state, when it receives from elsewhere (i.e., another unit) such a broadcast advertisement, to a master bridging state in which the routing facility is disabled and the unit provides commands for disabling a routing facility of other similar units to which the network is connected.

Abstract: A common BGP framework is provided to achieve VPN information discovery in network based VPNs using modifications to existing BGP-4 multiprotocol extension path attributes. The framework allow each VPN to implement different OSI layer-2 or layer-3 protocols, and to implement different OSI layer-2 or layer-3 protocols from those used by the backbone of the shared network. The extensions added to BGP-4 apply both to Virtual Router (VR) models of VPN systems and to non-VR models.

Abstract: A system for reducing clock speed and power consumption in a network chip. The system has a core that transmits and receives signals at a first clock speed. A receive buffer is in communication with the core and configured to transmit the signals to the core at the first clock speed. A transmit buffer is in communication with the core and configured to receive signals from the core at the first clock speed. A sync is configured to receive signals in the receive buffer at a second clock speed and to transmit the signals from the transmit buffer at the second clock speed. The sync is in communication with the transmit buffer and the receive buffer.

Abstract: Methods and systems for processing data communicated over a network. In one aspect, an exemplary embodiment includes processing a first group of network packets in a first processor which executes a first network protocol stack, where the first group of network packets are communicated through a first network interface port, and processing a second group of network packets in a second processor which executes a second network protocol stack, where the second group of network packets is communicated through the first network interface port. Other methods and systems are also described.

Abstract: A local-area network is provided where a plurality of OFDM stations and DSSS/CCK-only stations coexist. A beacon signal having a contention-free-period (CFP) is transmitted from an access point to the plurality of stations. The OFDM-modulated data is transmitted to the plurality of stations before the CFP ends, and after the CFP period ends, the DSSS/CCK-modulated data is transmitted, thereby preventing collisions between the stations.

Abstract: Using low PF values in conjunction with traffic-adapted contention windows leads to substantial decreases in delay and jitter. In general, adaptation to traffic reduces contention or delay: opening up the contention window in congestion and closing it on relief. Residual backoff adaptation provides for the reduction of the already decremented backoff values of stations that interrupted the backoff countdown process due to a transmission. It is good to adapt both the contention window and the residual backoff in order to avoid jitter. Otherwise, if the contention window is reduced but residual backoffs stay unchanged, new arrivals will enjoy shorter backoff delays than older ones, resulting in greater jitter. Adjusting both preserves the relative ordering of backoff counter values, which implies also some form of age ordering. Different adjustments can be applied to different priority traffic.

Abstract: Methods and apparatus are described relating to a crossbar which is operable to route data from any of a first number of input channels to any of a second number of output channels according to routing control information. Each combination of an input channel and an output channel corresponds to one of a plurality of links. The crossbar circuitry is operable to route the data in a deterministic manner on each of the links thereby preserving a partial ordering represented by the routing control information. Events on different links are uncorrelated.

Abstract: Methods and apparatus are described relating to a crossbar which is operable to route data from any of a first number of input channels to any of a second number of output channels according to routing control information. Each combination of an input channel and an output channel corresponds to one of a plurality of links. The crossbar circuitry is operable to route the data in a deterministic manner on each of the links thereby preserving a partial ordering represented by the routing control information. Events on different links are uncorrelated.

Abstract: Network load is reduced by combining packets by considering the MTU of a transmission path. When a routing processing unit detects the transmission of a large number of packets addressed to the same destination network, a routing information gathering unit determines the MTU of a transmission path along which the packets are to be transmitted. A combining unit assembles a combined packet by combining packets up to a length that does not exceed the MTU of the transmission path, and sends out the combined packet.

Abstract: A router stores a value that indicates the maximum number of channels that can be output to a group, and the number of channels that are output to the group. When a member of the group indicates a channel change selection from a first channel to a second channel, the router drops the first channel before outputting the second channel when the maximum number of channels and the number of channels are equal.

Abstract: The present invention provides a system and method for enabling the synchronization of a switch and an interface device. Based on a comparison of cell sequence numbers included in cells received from the interface device to a current cell time within the switch, cell time adjustment information can be transmitted to the interface device. The cell time adjustment information is used by the interface device to determine the cell sequence number that is included in cells transmitted to the switch.

Abstract: The object of the invention is a method for emulating the two-way information flow of a real application in a system, which comprises a first network element (101) and a second network element (102), as well as a packet network (103) between said first network element and said second network element. The method comprises transmitting a set of packets from the first network element (101) through the packet network (103) to the second network element (102). At the second network element, at least part of the transmitted packets are received. In a return transmission, a packet arrived at the second network element is transmitted back to the first network element only after a selected delay from the receiving of the packet at the second network element. The object of the invention is also a system, a network element and a computer program product implementing the method.

Abstract: A method and apparatus is described for the automatic delay compensation in space diversity radio transmissions. The method includes the steps of: a) receiving a first analog signal and a second analog signal, a possible delay being between the first and second signals; b) sampling the first and the second analog signals to obtain a first digital signal and a second digital signal, respectively; c) sending the digital signals to respective equalizers, and the steps of d) digitally delaying one of the first digital signal and the second digital signal by a period equal to an integral multiple of the sampling period, and e) recoverying, at the equalization phase, the residual difference between the imposed delay and the actual one.

Abstract: An improved digital interface circuit that allows a plurality of data streams and other digital information to be output over a single channel. The digital interface circuit includes a plurality of data inputs, at least one control input, at least one clock input, and a single serial bit output. The digital interface circuit receives respective input data streams at the data inputs, receives digital control information at the control input, and receives a clock signal at the clock input. The control information is an N-bit data stream having a data rate of 1/N times the rate of the input data streams (N?1). The digital interface circuit generates a frame synchronization signal for providing framing for the N-bit data stream, and time-multiplexes the data and control information over the single serial bit output.

Abstract: A multiple-fiber, stretched, fused and cleaved coupler designed to act as an output window to a fiber laser array or fiber-coupled laser array, in which the stretch, length and/or position of the fibers is chosen facilitate the in-phase oscillation of the lasers in the array. The in-phase oscillation of the lasers is facilitated by making one or more of the fibers interferometrically dark and having a different propagation constant than the other fibers. The other fibers, the interferometrically lit ones, are made to have fairly uniform intensity under the same collimated illumination.

Abstract: A method and apparatus are provided to modulate a beam of laser light. An acousto-optic modulator (AOM) is positioned to receive continuous-wave (CW) laser light at an angle approximating the Bragg angle. Fine tuning of the modulation is then used to compensate for any variations in the actual angle from the Bragg angle. A photodiode provides a feedback signal that may be used to tune or adjust the operation of the AOM. The photodiode is positioned to receive laser light at the first order diffraction and provide a signal indicative of the magnitude of the first order diffraction laser light produced by the AOM. The photodiode signal is coupled to an RF frequency generator so that the RF frequency may be adjusted to vary the acoustic wave in the AOM to compensate for any variations in the Bragg angle.

Abstract: A buried heterostructure quantum cascade laser structure uses reverse biased junction to achieve current blocking. Doping and ridge width of the structure may be adjusted to provide effective mode discrimination.

Abstract: In a semiconductor laser element, a lower cladding layer, a lower optical waveguide layer, an InGaAs compressive-strain quantum-well active layer, an upper optical waveguide layer, and an upper cladding layer are formed in this order in a stripe-shaped region on a substrate. A current-blocking layer is formed on both sides of the compressive-strain quantum-well active layer so that the compressive-strain quantum-well active layer is sandwiched between two portions of the current-blocking layer, and trenches extending along the direction of the laser resonator are formed through the current-blocking layer. Instead of providing the trenches, the widths of the layers formed above the substrate are reduced so as to form a ridge structure.