The paper suggests a practical "hybrid" synthesis methodology which integrates designer-derived analytical models for system-level description with simulation-based models at the circuit level. We show how to optimize stage-resolution to minimize the power in a pipelined ADC. Exploration (via detailed synthesis) of several ADC configurations is used to show that a 4-3-2... resolution distribution uses the least power for a 13-bit 40 MSPS converter in a 0.25 /spl mu/m CMOS process.

This paper presents a performance analysis of wireless MAC protocols in which the transmission of wireless packets is reserved through contention-based access. Compared with other pure random access protocols, this type of protocol is suitable for wireless networks with bursty packet arrivals, e.g., wireless ATM, or wireless LAN, for the reason that the carrier-sensing function cannot be efficiently implemented, thereby resulting in a very poor performance on the pure random access mechanism. The proposed analysis is based on simulation and the result shows that the efficiency of the protocol varies with different network parameters. The result also gives guidelines for optimizing the network performance, e.g., the control of the number of packets reserved in each request, the adjustment of signaling time-slots to the data time-slots ratio, and the adjustment of the threshold for rejecting a packet reservation request.

We proposed a link protocol called burst-oriented transfer with time-bounded retransmission (BTTR) for wireless real-time communications. This scheme uses a large transmission window for sending/receiving a burst of time-sensitive data, and within this window, another smaller observation window is repeatedly used for dynamic error retransmission(s). There is time limitation on each retransmission such that the burst of data can be received in a timely manner however by trading the packet loss performance for delay and throughput performance. An analysis is given in terms of the expectations of delay, throughput, and packet drop rate. The result shows that the proposed scheme can meet the delay and throughput requirement at reasonable packet loss performance.

An architectural design for high speed residential data access using the traditional IEEE 802.3 medium access protocol over the existing CATV network is proposed. The described architecture is attractive in using the existing infrastructure combined with the existing access scheme in the residential area and thus it is cost effective and well suited for home Internet access. However, the traditional CSMA/CD protocol is not suitable for the cable TV network due to its long propagation delay. Thus, we propose to combine the existing IEEE 802.3 CSMA/CD MAC with the segmented cable subnetworks so that a home user can get access to the Internet as if he is using the Ethernet at 10 Mbps, or Fast-Ethernet at 100 Mbps. Only three subcarriers in the passband cable is used and therefore it will not interfere with the traditional broadcasting channels and the future digital video channels. The segmented cable are interconnected by the defined cable bridge (CB) and a simple flow control mechanism is proposed among the CBs. The functional components and operations of each CB are also described. For a fair performance among the subnetworks, a prioritized queueing scheme is also proposed on each CB.

This paper presents a novel data placement and retrieving scheme or real-time multimedia stream playback via a scalable storage system.The multimedia streams are contiguous data packets which may be representing video, audio, or other media be retrieved simultaneously from a storage system on a real-time and on-demand basis, The major challenge in implementing the system is the support of simultaneously retrieving a big number of streams from a large-capacity storage system.This bottleneck comes from the seeking delay caused by the random movement of disk head in the storage under multiple stream requests.The proposed scheme efftciently reduces the seeking time by a very simple placement method and a retrieval scheduler and, in the mean time, continuous playback of multiple real-time streams and a scalable storage capacity are both achieved.Performance valuation of the proposed scheme is provided through theoretical analysis and experimental measure.

Inter-symbol interference (ISI) is often regarded as an important issue for high speed wireless applications. Due to the fast fading characteristic combining with the ISI, channel equalization in the wireless system often involves with dedicate design for each specific application. Multi-carrier modulation (MCM) which converts single high-rate symbol stream into parallel low-rate streams prolongs the symbol period and mitigates the ISI in nature. A multirate MCM transmission system based on octave-band filter banks was proposed to provide multimedia services with graceful variation in quality of services from high-tier mobile communication to indoor wireless communication. In this paper, the performance analysis of the multirate transmission system is studied. By considering the proposed system a wavelet domain multirate system, the system performance is modeled and simulated. Models derived in this paper is useful for further improvement of the multirate transmission system.

Personal mobile equipments are often required to be simple, inexpensive, as well as powerful in the same time, for instance, a lightweight mobile terminal with multimedia on Internet. We refer to this equipment as Mobile Multimedia Terminal (MMT). This paper presents a protocol architecture supporting the multimedia access for MMT in a wireless WAN environment. To save the processing power, each MMT has a service agent located in a server connected to the mobile switching center. Based upon this model, a transport protocol and architecture for mobile multimedia communications is proposed. The proposed protocol is intended for the handover and real-time data transportation between MMT and the MMT agent. Architectural design of the MMT and its agent will also be described.

This paper introduces a nonlinear recursive least squares decision feedback equalizer (DFE). The nonlinearity is modeled using a bilinear system. Bilinear models are attractive because of its ability to approximate a large class of nonlinear systems efficiently. Results of computer simulation experiments that compare the performance of the bilinear DFE to two other alternatives, viz., a conventional DFE and a truncated Volterra DFE, in eliminating the intersymbol interference caused by a nonlinear channel are presented. The results demonstrate the superior performance of the bilinear DFE over the competing structures at least for the cases considered.>

Although WDM-based single-hop star networks are attractive owing to their all-optical communication features, the throughput of such lightwave networks is limited due to the small number of available wavelengths. In this paper, a wavelength-reusable local lightwave network which consists of interconnected WDM star networks is proposed. Based on this architecture, the lower bound for the problems of minimizing the switching duration is derived. A transmission scheduling algorithm for this architecture to efficiently reuse the wavelengths is also proposed. The analytical result shows that the proposed scheduling algorithm always produces solutions no greater than twice of the lower bounds. Simulation results show that given the same number of users and available wavelengths, the solutions (in terms of the average switching duration and the average number of switching matrices) obtained by the proposed scheduling algorithm on the interconnected WDM networks are better than the optimal solution on a single-star WDM network. In most cases, the performance improvement achieves 20% to 45%.>