PON (Passive Optical Network) technologies are applied to FTTX in order to provide high-speed Internet services. Currently, TDM-PON (Time Division Multiplexing PON) has been deployed widely. However, TDM-PON faces difficulty to control upstream traffic with optical level multiplexing in case of transmission rate grager than 10 Gbps because of performance limitation of the optical burst multiplexing. In order to release the limitation, we have proposed SDM-PON (Subcarrier Digital Modulation Passive Optical Network) that does not require the optical burst multiplexing. In this paper, we improve SDM-PON mechanisms in order to increase the number of connectable ONUs (Optical Network Units) up to 32 in one OLT with realizing low power consumption. We evacuate transmission performance with simulations to confirm 32 ONUs enable to communicate simultaneously in realistic conditions.

Optical burst mode receivers are an inherent part of the Passive Optical Network (PON) and Optical Switched Access Network (OSAN). We previously implemented optical burst mode receivers with AC-coupling and DC-coupling using off-the-shelf components, and showed that AC-coupling offers better performance. We then proposed an optical burst mode receiver with a transfer function Gn(s)=1-Hn(s) (Hn(s): a Bessel filter transfer function of degree n) by generalizing AC-coupling (n=1), and showed that the proposed receivers have better performance than AC-coupling equivalents. This paper proposes an optical burst mode receiver with two transfer functions: a fast response Gn(jω/ω1) for receiving the beginning of a burst signal, and a slow response Gn(jω/ω2) for the body of the signal. The receiver switches the transfer function from Gn(jω/ω1) to Gn(jω/ω2) just after Gn(jω/ω1) successfully eliminates the DC component of the input burst. We present transfer function design examples, and show that the new proposal offers much better performance.

To economically provide higher bit rate services to users, we previously proposed the Subcarrier Digital Modulation based Passive Optical Network (SDM-PON). We also showed the feasibility of a 16-QAM SDM-PON that constantly provides the service data rate of 1 Gbps to each user, in which ONU biases the QAM signal higher so that it is entirely positive before converting into an optical signal (bias signal sending scheme). To decrease sending optical power, this paper proposes a scheme for clipping negative components of the electrical QAM signal (clipped signal sending scheme), along with three methods for regenerating the original QAM signal at the receiving side. We show that the clipped sending signal scheme, and all three regenerating methods, are effective by simulation. The resulting SDM-PON supports 8 ONUs at different distances, 10 km to 20 km, where each ONU constantly sends 1 Gbps data to OLT using 16-QAM.

This paper proposes an indoor trace tracking algorithm with UHF-band passive RFID tags. The tags are attached to the ceiling and the user carries an RFID reader. The reader captures the RFID tags, and the user's position is estimated by the coordinates of the captured tags. The user's track is drawn by connecting estimated positions over time. To decrease the estimation errors expected with the simple approach, this paper proposes a new tracking algorithm that compensates the corner effect, and takes the movement characteristics of the user into consideration.

The sampling theorem of Whittaker, Ogura, Kotelnikov, Someya, and Shannon (WOKSS's sampling theorem) appears in most books on information theory, and is well known. WOKSS's sampling theorem must exclude periodic functions. This paper first introduces a sampling theorem for periodic functions, and gives its physical meaning; this theorem is effective for functions that can be expressed as a finite Fourier series in the same way as WOKSS's theorem deals with a band-limited signal. Since many signals such as an Orthogonal Frequency Division Multiplexing (OFDM) signal can be expressed by periodic functions, the sampling theorem can play an important role in a communications area. Extending the theorem, this paper proposes a sampling theorem for periodic functions with no minus frequency component of a finite Fourier series, and shows its example. In addition, we examine the OFDM signal generation and termination by applying sampling theorems for periodic functions including the proposed one.

It is well-known that TCP has difficulty in achieving desirable levels of performance in multi-hop ad hoc networks. TCP implements a loss-based congestion control mechanism for adjusting traffic in the networks. However, when TCP reacts to a loss event in a multi-hop ad hoc network, contentions raised in the IEEE 802.11 MAC layer among nodes usually becomes excessive. This problem eventually induces packets scheduling variation at the MAC layer. As a result, the quality of service (QoS) assurances given to multimedia traffic streams, such as UDP traffic, are violated. This paper focuses on preventing the TCP congestion window (CWND) size at the network layer from becoming excessive. We propose an interference drop scheme (IDS) at the network layer to prevent excessive increases in MAC contention by dropping over-delayed packets held by local queues. Simulations show that the proposed approach yields better a TCP performance and the assurance of QoS provision to UDP flows in multi-hop ad hoc networks.

The passive optical network (PON) shares feeder transmission facilities to provide broadband services economically. To enhance PON survivability, ITU-T Rec. G.983.5 defines some protection switching methods. These methods, however, always bring about signal loss when switching is performed. Another related problem occurs when the network operator replaces one optical fiber cable with another that takes a different route due to construction activities like social infrastructure renewal. If hitless switching is available, the operator can carry out that activity anytime without impacting users, and can also offer higher-grade broadband services. Thus a hitless protection technology for PON systems is very attractive and useful. This paper proposes hitless switching schemes for protected PON systems based on the PON's ranging functionalities. We also present a specific design example for GE-PON.

The congestion control provided by Transmission Control Protocol (TCP) plays a key role in ensuring network stability and increased Quality of Service (QoS). However, in the face of the increasing number of high-speed networks such as Multi-protocol Label Switching with band-width width reservation, conventional congestion control algorithms limit the network resource utilization efficiency since they overreact to packet loss. This paper proposes a new TCP congestion control scheme that improves the goodput by using reserved bandwidth to set the slow start threshold and the congestion window after a timeout or three duplicate acknowledgments. The advantage of this algorithm is that the TCP sender recovers faster after packet loss, especially when the connections have large round trip times; it avoids Reno's overreaction to loss.

To meet the explosive increase in TCP/IP data traffic that has occurred due to the rapid growth of the Internet, the broadband passive optical network (B-PON) with dynamic bandwidth assignment (DBA) is considered to be one of the most cost-effective solutions. The DBA function enables B-PON users to share their upstream bandwidth with each other dynamically. This paper proposes a new DBA algorithm that completely satisfies the requirements specified in ITU-T G983.4. The proposed algorithm consists of three processes. The first process predicts the requested bandwidth and ensures quick response to ONT input data. The second process determines the number of cell slots (CS) of each bandwidth type in each bandwidth assignment period (BAP) and ensures complete assignment of guaranteed bandwidth, and fair assignment for non-assured bandwidth among transmission containers (T-CONT). The third process determines the CS locations of each bandwidth type in each BAP according to the number of CS obtained from the second process so as to minimize the cell delay variation (CDV). This paper also evaluates the proposed algorithm using a prototype system.

The technique of variable frames with HEC (header error control) and frame length information is being studied widely for transporting IP packets efficiently in layer 2. The HEC and frame length information are used for frame alignment. This paper investigates the frame alignment performance and finds that the conventional frame alignment method allows a misframe, the framer declares out-of-sync falsely due to bit error corruption of the header, to occur frequently. We propose a method for reducing the misframe frequency that is based on forced two-bit error correction. We show that the proposed method has a much better performance than the conventional method.

The development and system design of a new synchronous DS4 multiplexer with cross-connect function is described. This new multiplexer is expected to provide cost reductions due to more efficient use of transmission facilities, and rapid circuit provisioning. The multiplexer consists of three kinds of digital cross-connect blocks, a transmission terminating block, an intraoffice-interface block, and a controler. It is also expected to provide improved network maintenance, higher network reliability, and greater system versatility for new services.

Demand for high-speed wide-band telecommunication services is rapidly growing due to the remarkable progress in office, factory, and laboratory automation. These services are expected to be first introduced in leased circuits and then extended, to high-speed switched services. This paper presents the users' demands for such high-speed services and the NTT design concepts for the high-speed digital data transmission system (HDDTS) in NTT to provide such high-speed leased circuit service. The ISDN user-network interface is adopted. The basic high-speed pointto-point transmission service started in October 1984, and enhanced leased circuit service involving multiplexed interfaces and muitipoint circuits will be introduced in the fall of 1985.

The design objectives and test results of a new services integrated system using a packetized speech technique are presented. First, several important design parameters are investigated through theoretical analysis and computer simulation. Second, an experimental system for confirming the choice of design parameters as well as measuring the speech and data communication quality is described, and the test results of this system are discussed.

This paper describes a new integrated service network using packetized voice/data technology which was developed in consideration of the sophisticated service demands for voice, data, and facsimile communication. A major problem in constructing this network is in the local distribution network, i.e., the subscriber interface unit, the local end office, and the subscriber loop which connects them. The concepts, choice of design parameters, and laboratory test results of the packetized service integration network are briefly discussed.