This paper presents the architecture, design and implementation of a 16 point FFT processor for a high speed wireless local area network. The 110000-transistor chip is implemented in 0.6 /spl mu/m TLM CMOS, operates worst case at 50 MHz at a supply voltage of 2.5 volts, and consumes 80 mW.

Current wireless LANs that are small enough for portable computing devices have transmission rates up to a few Mbit/s, at the lower end of that obtained in IEEE 802compliant wired LANs. These LANs can provide a useful service when the application demands and number of users are kept low. Much higher performance, from several 10’s of Mbit/s to over 100 Mbit/s, is needed to accommodate more users and multimedia traffic.

This paper summarizes aspects of the VLSI development of a high-speed wireless local area network (WLAN). The implications for system-on-a-chip designs are summarized

HIPERLAN is a high speed wireless LAN standard adopted by the ETSI. It operates in 150 MHz of spectrum in the 5 GHz band, with up to five TDD channels provided for duplex data transmission at up to 23.5 Mb/s. This paper shows that in the presence of cochannel cells and full link utilisation, HIPERLAN performs poorly when using its DCA-type clear channel assessment (CCA) scheme, and that the target performance levels are only achievable under optimal FCA-type channel assignments. Two main factors contribute to this poor performance: HIPERLAN has insufficient channels to meet C/I constraints in the presence of cochannel cells; and the CCA scheme results in poor channel distributions.

A characterisation of cable network elements to facilitate accurate simulation and analysis for the design of digital cable networks is presented. Nonlinear and frequency-dependent blocks are proposed for modelling each network element. Complex behaviour of a network can be described by a combination of these simple blocks, The technique is illustrated by application to a cable amplifier.

This paper summarizes developments aimed at providing cheap manufacturable 5 GHz data radio front-ends for the newly created U-NII and HIPERLAN bands. The result is a set of reconfigurable modules for radio architecture experimentation and a small form factor 5 GHz radio for WLAN and link applications which is economical and manufacturable in volume.

WLAN systems face technical problems similar to those encountered in outdoor wide-area radio-based systems, including the limited available bandwidth and fading noise due to multipath interference and blockage. The goal in WLAN system design is to transmit at the maximum information rate with an acceptable probability of error and minimum equipment complexity, power, and cost. Competing approaches use either infrared radiation or radio waves in the microwave or millimeter-wave bands. We have developed a radio solution at millimeter-wavelength frequencies, where the spectrum is sufficient to accommodate link speeds of hundreds of Mbps. Using a test bed with burst-mode transmission capability and an experimental 40-GHz radio, we have demonstrated a picocellular approach with a range of approximately 10 meters and link rates up to 185 Mbps. In addition, we have built a prototype modem with a raw link rate of 54 Mbps for use in a high-speed indoor WLAN demonstrator.

The third derivative of MESFET drain current behavior is useful for devise characterization. It provides information necessary to device a model to predict large-signal dynamic behavior with accuracy over an extended range of operating conditions. Extra parameters are proposed to define behavior in subthreshold and triode operating regions. A continuously differentiable form that models third-order behavior correctly describes these regions. The result is an accurate large-signal model suitable for design and analysis of distortion and intermodulation in analog circuits.<>

This paper presents a new interference model for microcellularnetworks which integrates radio propagation parameters and user terminal mobility. This model uses a parameter denoted the ’’interference to noise ratio‘‘ (INR) toobtain a simplified description of mobile link outage contours as a function ofthe location of the fixed and mobile radio ports. The INR is used todemonstrate that microcell networks are more interference limited than macrocellnetworks, and thus are more affected by user terminal mobility. Expressions arederived for the INR and user terminal cell radius distributions.It is shown that in microcell systems a significant proportion ofterminals may not be able to meet a contiguous coverage criterion, and that closermicrocell spacing can reduce rather than improve the coverage quality. Examination of cochannel and adjacent channel reuse ratios in DCA microcell systemssuggest that the closer frequency reuse is primarily responsible for thesecoverage effects. Monte Carlo simulations are used to test the analytical theory. These results may form the basis of a design methodology for microcellsystems.

The distribution functions of cell radii in a microcell system are derived in closed form for three portable terminal distributions and tested using a Monte Carlo simulation. These distributions provide information about the quality of the microcell coverage. The results suggest that cell size reduction from near-far or same-cell interference from terminals using well spaced channels can be as severe as that resulting from adjacent channel interference from terminals in nearby microcells.

This paper presents a general interference model for arbitrary cellular and microcellular networks which enables a unified treatment of all mobile link reception environments, from purely noise limited to purely inter ference limited, as a function of the ratio of interference power to receiver noise power at a receiver . This ratio is denoted ‘INR’ and given the symbol . The model enables a simplified spatial description of radio link outage contours to be derived, and it is shown that the maximum possible range for any mobile terminal is a simple function of the INR. Computer simulations are used to test the analytical theory and show that in comparison to lar ge ce l systems, microcellular networks are interference limited and exhibit lar ger variation in cell sizes. Examples show that this may present dif ficulties in engineering a microcellular system to meet a tar get coverage quality require ment. 0 Introduction The first practical cellular telecommunications system was developed by the Bell T elephone Laboratories during the 1970s [1]. The cellular concept was based upon frequency reuse via fixed channel allocations (FCA) that enabled a receiver at a cell boundary to meet the signal to interference ratio (S/I) required for good signal quality in the presence of cochannel inter ferers [2]–[6]. The capacity of cellular systems can be increased by splitting existing cells, cell sectoring, or a combination of both, but in practice there are limitations to the extent to which cells can be physically reduced in size or sectorised. The limits for most lar ge cell systems appears to be a cell radius of approximately 1 kilometre and at most 3 sectors. In many lar ge cities around the world, the existing cellular systems are beginning to reach their capacity limits. To meet the growing demand for mobile telecommunications, and to satisfy the vision of a personal communicator in every pocket, microcell architecture was developed. This architecture differs from conventional or lar ge cell architecture in three fundamental ways: The cells are typically less than 1 km in radius The mobile terminals radiate much smaller power levels There is no centralised, fixed cell planning (all channels are available in every cell)

An interference model developed for arbitrary microcellular networks, based upon a parameter called the ‘Interference to Noise Ratio’ or INR, is used to derive interference and cell radius statistics for mobile stations in a microcellular network. The theoretical INR and cell radius statistics for simple interference environments show good agreement with numerical Monte Carlo simulations. For more complex environments, it is hypothesised that the Central Limit Theorem could be applied to approximate the behaviour of an interferer ensemble, enabling the cell radius statistics for a network to be expressed in terms of the system design parameters. Simulations of microcell ensembles that support this hypothesis are presented.

This paper presents an interference model for cellular and microcellular networks. This model enables a unified treatment of all reception environments from purely noise limited to purely interference limited through a parameter denoted the ‘interference to noise ratio’ or INR. Using the model, a simplified spatial description of mobile link outage contours is derived. Computer simulations are used to test the analytical theory. It is shown that as a link becomes more interference limited, larger variations in cell sizes result.

This paper summarises the cell design issues for microcellular networks and explains how conventional large cell network design techniques may not be suitable for designing a ubiquitous microcellular network. A general interference model developed for arbitrary microcellular networks, using a parameter denoted the 'interference to noise ratio' or INR, is used to determine the microcell spacing required to maintain certain cell coverage targets as a function of mutual spill power. It is shown that a contiguous cell coverage requirement imposes significant limits upon the accumulated interference allowable in a microcell system.

The paper describes an interference model that provides a simplified description of cellular and microcellular outage contours. The model and computer simulations indicate that cell size variation increases as microcellular systems become more interference limited. Some cellular and microcellular systems are modelled.