The characterization of the transfer function of the power line (PL) channel is a nontrivial task that requires a truly interdisciplinary approach. Until recently, a common attribute and limitation of existing models for the PL channel transfer function lay in the phenomenological or statistical approach usually followed. This approach allows one to describe the channel only partially, e.g., as dominated by multipath-like effects, and prevents one from unveiling special properties of it. Multiconductor transmission line (MTL) theory was recently found by the authors to be a useful and accurate tool in modeling the PL transfer function while, at the same time, taking into account the wiring and grounding practices mandated by several regulatory bodies for commercial and residential premises. Crossing several layers of abstraction and following a bottom-up approach, complex circuit-level models originating from MTL theory can be manipulated and represented in terms of cascaded two-port networks (2PNs), thus allowing one to compute a priori and in a deterministic fashion the transfer function of any PL link. In the present contribution, we present additional analysis and data that validate the accuracy of the MTL approach and further justify its use in the PL channel context. Moreover, we also describe in detail the methodology to follow for modeling both grounded and ungrounded PL links in a unified framework. A consequence of the validity of the proposed modeling is that it can facilitate the process of standardization of the PL transfer function, an important step toward the availability of a commonly agreed upon (set of) channel transfer functions.

We propose and experimentally validate a novel method for transmitting several OCDMA channels in the unused bandwidth of a single WDM channel, thus allowing the OCDMA and SONET signals to share the same WDM channel. The theoretical justification of this new modulation scheme is based on a powerful property of spectrally phase-encoded OCDMA signals which is reported and proved for the first time: the operation of spectral phase encoding allows us to convey broadband signals over a disjoint (non-contiguous) frequency support. Remarkably, this property' does not impair signal orthogonality' so that it can be exploited both in synchronous and asynchronous OCDMA.

Considerable effort has recently been devoted to the determination of accurate channel models for the power line (PL) environment, for both the indoor and outdoor cases. The common denominator (and limitation) of the known and previously published models is the particular type of approach followed. In fact, until now, the PL channel (PLC) has been treated from a mere phenomenological or statistical point of view. This allows us to describe the channel only partially, e.g., as dominated by multipath-like effects, and prevents us from unveiling any of its particular properties. We report results about a new approach to the modeling of the PLC based on multiconductor transmission line (MTL) theory. The need for an MTL approach arises from the fact that indoor power cables consist of three conductors, and not just two as for classical twisted-pair and coaxial cable cases. This approach allows us to include wiring and grounding practices dictated by the United States National Electric Code (NEC). On the basis of this modeling, we reveal that the PLC is a more deterministic medium than commonly believed. Moreover, it is also possible to prove an interesting symmetry property of the PLC that was previously unknown: the PLC, regardless of its topology, exhibits the same transfer function when driven from either side, provided that the source and load impedances are the same.

We present initial results on a WDM compatible spectrally phase encoded optical CDMA (OCDMA) system that uses 16 phase-locked laser lines within an 80-GHz tunable window and an ultrahigh frequency resolution encoder/decoder. The described system differs from conventional OCDMA systems by independently phase encoding individual spectral line of a mode locked laser (MLL). We report here novel properties of phase codes and we also demonstrate compatibility of the proposed OCDMA system with conventional WDM networking by simultaneous transmission of two 2.5 Gb/s OCDMA users in one DWDM window along with six DWDM channels in other DWDM windows through a transparent reconfigurable network having 25 km node spacing.

EMI/EMC is a dominant problem encountered in high bit-rate (>100 Mb/s) transmission over unshielded twisted-pair cable (UTP) which leads to a novel set of line-code dependent tradeoffs affecting transceiver and cable plant complexity. To understand the tradeoffs, the authors examined the factors affecting RF emissions and susceptibility in both trellis-configured cable plant models and installed cable plant. An analytical model is presented that describes mode conversion by discontinuities in multi-pair UTP cable plant. Termination of the three dominant propagating modes produces consistently lower radiated emissions than other standard cable termination procedures. Cable plant parameters which predict good transmission do not assure satisfactory EMI performance. Radiated emissions from the trellis model occur in broad bands at 43, 53, 60, 70, and 80 MHz, and show a positive correlation with the transmit signal spectrum. For a given transmit level at 155 Mb/s, the emissions with MLT3 and BPR1 were 8-13 dB (10 dB typ) and 4-20 dB (13 dB typ) below NRZ levels, respectively. The authors also compared the performance of NRZ, MLT3, BPR1, and BPR4 line codes in an experimental 155 Mb/s link with 100 m of UTP5 by measuring the BER sensitivity to both injected noise and pseudo-random data sequence length. When the measured receiver penalty associated with three levels is included, MLT3, BPR1, and BPR4 offer 5-15 dB better performance over NRZ at 155 Mb/s, although implementation complexity is greater.

The authors examine the physical design issues associated with terabit/second switching systems, particularly with regard to the customer access portion of the switch. They determine the physical design requirements in the areas of backplane interconnections, integrated circuit packaging, and circuit board technology and identify areas where existing- or near-future physical design technologies are inadequate to meet the requirements of this application. A new 3D interconnection architecture that solves some of the problems encountered at the backplane level is suggested. It is also suggested that multichip module technology will help meet some of the speed and density requirements at the chip packaging level. Some of the system-level consequences of the proposed model are discussed.>