An admission and congestion control scheme is proposed for large-scale wireless mesh access networks that stabilizes the network under heavy traffic loads within a cooperative and distributed medium access control framework. The scheme utilizes only local information available at the controlling router, and yet guarantees to deliver the data of the admitted user to the destination router unless the user is dropped over the first hop. We first show that the scheme can stabilize the network, and then demonstrate its stability properties through simulations. A high-fidelity and large-scale simulator is created that implements measurement-based models for radio signal propagation and interference calculation, and captures the stochastic network behavior resulting from random traffic arrivals, admission control, and queueing. Simulation results confirm that the scheme indeed stabilizes the network and thus provides stable per-session throughput under heavy traffic loads.

This paper presents a novel hierarchical and grid based reconfigurable optical-wireless network (GROW-Net) architecture. GROW-Net supports scalable and flexible integration of a large scale wireless mesh network in a municipal environment. Under the GROW-Net architecture, a joint evolution strategy is proposed to allow graceful upgrades in both optical backbone and wireless mesh network. To alleviate the known throughput bottleneck in mesh networks, a capacity enhancing technique based on flexible cell-splitting is proposed. The technique allocates wavelength resources to unutilized dark fibers using the unique structure of GROW-Net's reconfigurable and colorless access gateways. To analyze the effectiveness of the approach, throughput performance of the wireless mesh network is analyzed using a high fidelity simulator. The wireless mesh network in the simulation employs a distributed and cooperative medium access control protocol within a time division multiple access and time division duplex framework. Through the high fidelity simulation, results show the degree of throughput enhancement by using cell splitting method. The evolution strategy further enables the backbone to scale its bandwidth and adapt to future high throughput and very high throughput wireless technologies. The performance of the proposed backbone is demonstrated experimentally over the optical testbed.

A medium access control (MAC) protocol is developed for time division multiple access (TDMA) and time division duplex (TDD)-based large wireless mesh networks (WMNs) with infrastructure support. Utilizing control time slots, the protocol operates in a fully cooperative and distributed manner: mesh routers cooperate to negotiate resource allocation, and the messages are exchanged only among one-hop neighbors. Data time slots are dynamically distributed among mesh routers adapting to the varying network conditions. A large high-fidelity and time-driven simulator is created which implements realistic radio propagation models and interference as well as capturing the stochastic network behavior due to random traffic arrivals and queueing. A novel parallel processing technique is employed to reduce the long runtimes. Through simulations, protocol performance is determined for different resource allocation schemes including data time slot selection policies and routing strategies. Results show that the effects of different resource allocation schemes on the network performance greatly vary under wide- ranging congestion levels.

In multiple input multiple output (MIMO) systems, predicting performance prior to data transmission remains a difficult problem. This paper proposes an algorithm to predict the average bit error rate (BER) and packet error rate (PER) of a MIMO system, after decoding the error correction code (ECC). Since the algorithm makes no a-priori assumptions about the channel, it is applicable to a wide variety of random channels. In our results, we show a good match between the predicted BER and PER, and the BER, PER from simulated data transmission.

In coded OFDM systems, the transmission power of each subchannel is the same. In this letter, a power optimization scheme is presented to minimize FER or BER of a coded OFDM system under constant transmission power constraint. The power optimization is shown to be a convex optimization problem.

In practical applications, some coded symbols are unequally punctured for rate matching of terminated convolutional codes. In this letter, performance upper bounds are proposed for a terminated convolutional code punctured with an arbitrary pattern. To obtain the upper bounds, a weight enumerator is defined to represent the relation between the Hamming weight of the coded outputs and that of the selected input information bits. A method is presented to compute the weight enumerator based on a modified trellis diagram. Simulation and numerical analysis show that the proposed bounds are tight to the performance of terminated convolutional codes.

Orthogonal Frequency Division Multiplexing (OFDM) in combination with bit-interleaved coded modulation (BICM) is efficient and robust high-speed transmission technique used in IEEE 802.11a standard. In this paper, using Request To Send (RTS)/ Clear To Send (CTS) mechanism, we present a method to improve throughput performance by adaptively changing constellation size and power distribution across the sub-carriers without sacrificing throughput due to explicit feedback. Based on the theoretical analysis, this complex maximization problem can be approximately solved by a simple iterative algorithm. Simulation results show that the proposed method can significantly improves throughput performance and the performance gain is robust to channel estimation and synchronization error.

In orthogonal frequency division multiplexing (OFDM) systems, carrier frequency offset (CFO) due to mismatch of the local oscillators causes inter-carrier interference (ICI), which may result in significant performance degradation. Although, several frequency synchronization schemes were reported, there possibly remains frequency offset and it can still generate ICI and induces phase distortion of the OFDM symbols. In this paper, we propose a method to compensate both residual frequency offset (RFO) and RFO induced phase error (PE) by the Kalman filter. Our approach is based on building a simple robust state-space model and the Kalman filter is then applied to estimate and track the RFO and PE. Simulation results show that the proposed method significantly compensates the performance degradation due to RFO and almost achieves ideal packet error rate (PER) performance with lower complexity.

Using parallel processing, the execution time of wireless network simulations can be significantly decreased without compromising the fidelity of the simulation. However, incorporating details such as signal propagation, interference, and resource allocation schemes may add considerable complexity to the parallel simulation which may overshadow the gain of parallelization. This paper reports on a recent work on parallelizing a high fidelity mobile wireless network simulator where dynamic resource allocation schemes are incorporated and the physical layer characteristics are included in sufficient detail. The program was run on a modular super-computing platform with 32 processors interconnected by high-speed infiniband links. Simulation results indicate that parallelization can significantly improve the runtime performance of the simulation without simplifying the physical layer modeling; a runtime speedup of more than 11 was achieved using 16 processors for a set of simulation parameters. A set of quantitative results on the capacity of the simulated network is obtained using the parallel simulator.

The authors correct a mathematical equation contained in the correspondence "Multiple-antenna signal constellations for fading channels," previously published in the IEEE Transactions on Information Theory

In this paper we are interested in designing a robust timing synchronization algorithm for OFDM systems that utilize pilot-aided channel estimation. We first characterize the impact of timing errors on the performance of a pilot-aided OFDM system. We derive analytical expressions for average channel estimation error variance in the presence of timing errors in high delay spread fading environments. The derived expressions show that pilot-aided channel estimators are considerably sensitive to timing synchronization errors due to the impact of rotations in different bases. We then show how to utilize this sensitivity to design a robust timing synchronization algorithm, without training overhead. The proposed algorithm is a cross-block design that uses channel estimation information to improve timing synchronization. We confirm our analytical results by simulating the proposed algorithm in high delay spread fading environments. In this paper, i.e. part I, we focus on timing synchronization for low-mobility cases. The analysis and results are then extended to high-mobility applications in part II.

In this paper we consider the design of a robust timing synchronization algorithm for pilot-aided OFDM systems in high-mobility fading environments. We first analyze the impact of both mobility and timing errors on the performance of a pilot-aided OFDM system for frequency selective fading channels, by deriving an expression for channel estimation error variance. The analysis will show that, even for high levels of mobility, a pilot-aided channel estimator is considerably sensitive to timing errors, due to the impact of rotations in different bases. We then show how this sensitivity can be utilized to design a robust timing synchronization algorithm for mobile OFDM systems, without relying on synchronization training information. Theoretical results are then confirmed by simulating the performance of an OFDM system in high delay and Doppler spread fading environments. Finally, we show how the proposed mathematical framework and algorithm can be used to address timing synchronization in the presence of a frequency offset as well. The analysis of this paper is the extension of the derivations of Part I [8], the accompanying paper on the design of a robust timing synchronizer for low-mobility OFDM systems.

This letter addresses the effect of timing synchronization errors that are introduced by an erroneous detection of the start of an orthogonal frequency-division multiplexing (OFDM) symbol. Throughout this letter, the term "timing error" would refer to this type of error. Such errors degrade the performance of an OFDM receiver by introducing intercarrier interference (ICI) and intersymbol interference (ISI). They can occur due to either an erroneous initial frame synchronization or a change in the power delay profile of the channel. In this letter, we provide a mathematical analysis of the effect of timing errors on the performance of an OFDM receiver in a frequency-selective fading environment. The analysis presented in this letter is for the case that no equalization technique has been used to mitigate the introduced ICI and ISI. We find exact formulas for the power of interference terms and the resulting average signal-to-interference ratio. We further extend the analysis to the subsample level. Our results show the nonsymmetric effect of timing errors on the performance of an OFDM system. Finally, simulation results confirm the analysis. The results of this letter can be easily extended to address the effect of such errors on DMT modems.

The capacity advantage of multiple input-multiple output (MIMO) channels lays in the decomposition of the channel into several spatial sub-channels. Knowledge of the channel at the transmitter can further increase the system capacity. This effect has been theoretically demonstrated. Here we illustrate the potential of channel state information (CSI) at the transmitter using actual channel measurements of a system with 4 transmitters and 4 receivers in an indoor environment, under line-of-sight and non line-of-sight conditions, considering both infinite and finite allowable modulation sizes. We show that CSI at the transmitter is more beneficial in the case of low signal to noise ratio and in cases where the channel richness is limited. In terms of achievable spectral efficiency, 60% of the theoretical capacity can be obtained without advanced coding techniques, as long as eigen mode transmission is used, and the performance is similar for all power and rate allocation schemes that we investigate

With the rapid growth of the Internet and other Internet protocol-related applications, wireless systems have shifted toward supporting a wider variety of data services. This has placed an enormous strain on the already tight capacity of wireless systems. Many improvements have been made at the link layer in second- and third-generation wireless systems. However, further improvements in packet data admission, scheduling, and policing are necessary to maximize capacity and user satisfaction. Of the three, efficient scheduling has the greatest impact on increased system capacity. In addition, proper scheduling can greatly improve user satisfaction. We introduce an adaptive cross-layer packet scheduler, which outperforms popular packet schedulers with respect to both packet delay and user throughput.

Orthogonal frequency-division multiplexing (OFDM) is robust against frequency selective fading due to the increase of the symbol duration. However, for mobile applications channel time-variations in one OFDM symbol introduce intercarrier-interference (ICI) which degrades the performance. This becomes more severe as mobile speed, carrier frequency or OFDM symbol duration increases. As delay spread increases, symbol duration should also increase in order to maintain a near-constant channel in every frequency subband. Also, due to the high demand for bandwidth, there is a trend toward higher carrier frequencies. Therefore, to have an acceptable reception quality for the applications that experience high delay and Doppler spread, there is a need for ICI mitigation within one OFDM symbol. We introduce two new methods to mitigate ICI in an OFDM system with coherent channel estimation. Both methods use a piece-wise linear model to approximate channel time-variations. The first method extracts channel time-variations information from the cyclic prefix. The second method estimates these variations using the next symbol. We find a closed-form expression for the improvement in average signal-to-interference ratio (SIR) when our mitigation methods are applied for a narrowband time-variant channel. Finally, our simulation results show how these methods would improve the performance in a highly time-variant environment with high delay spread.

OFDM systems are sensitive to timing synchronization errors. Utilizing pilot-aided channel estimators in OFDM systems can further increase this sensitivity. This is shown in Y. Mostofi et al. (Oct. 2002) where authors have analyzed the-effect of such errors on a pilot-aided channel estimator in a fixed wireless environment. They proposed an algorithm that exploits this sensitivity to improve timing synchronization without additional training overhead. The effect of these errors and design of suitable synchronization algorithms for high mobility applications, however, have not been studied before. In this paper we extend the analysis in Y. Mostofi et al. (Oct. 2002) to high mobility environments. Timing synchronization becomes more challenging for mobile applications since power-delay profile of the channel may change-rapidly due to the sporadic birth and death of the channel paths. We find analytical expressions for channel estimation error in the presence of timing synchronization errors and mobility. We show that the sensitivity of the channel estimator can still be exploited to improve timing synchronization in high mobility environments. Then we extend the algorithm proposed in Y. Mostofi et al. (Oct. 2002) to high mobility applications. Finally simulation results show the performance of the algorithm in high delay and Doppler spread environments.

In OFDM receivers, pilot tones should be inserted among the sub-carriers in order to estimate the channel in high delay spread environments. To estimate the channel at all the sub-carriers, using the pilot tones, different frequency-domain interpolators can be used. For a given number of pilot tones, these interpolators will have different levels of complexity depending on the number of adjacent pilot tones that they use to estimate the channel at each sub-carrier. For high delay-spread channels, as the number of adjacent pilot tones used by an interpolator increases, channel estimation improves resulting in a tradeoff between performance and complexity. In this paper, we first find an analytical average error rate formula for a pilot-aided OFDM receiver in a delay spread fading environment. This analysis will provide a framework to evaluate the performance of different interpolators. As an example, we then evaluate and compare the performance of two interpolators with different levels of complexity: a trigonometric interpolator and a linear one. We derive the average error rate formulas of the interpolators and confirm that the former one has a considerably better performance as the delay spread increases. However, for flat fading channels, we show that the linear interpolator has a 3 dB gain. Finally, our simulation results support the mathematical analysis and comparison.

A new model for UHF propagation in large buildings is presented. This model relies on knowledge of the interior arrangement of the building without requiring much detail. The guiding of radiation along hallways is the most significant propagation process at distances of more than 10 m from the transmitter. The waveguide model predicts the power loss rate along the hallways, which is affected by the coupling among the propagating modes. The coupling results from the roughness of the surfaces in the building; it is predicted in an average manner using the average deviation of the walls from perfect smoothness. The model predictions are compared to measurements in an office building and to ray tracing predictions.

Orthogonal frequency division multiplexing (OFDM) is robust against inter-symbol interference (ISI) due to the increase of the symbol duration. However, for mobile applications, channel variations during one OFDM symbol introduce inter-carrier-interference (ICI), which degrades the performance. This gets more severe as mobile speed, carrier frequency or OFDM symbol duration increases. We introduce two new methods to mitigate ICI in an OFDM system with coherent channel estimation. Both methods use a piece-wise linear model to approximate channel variations. The first method extracts channel variations information from the cyclic prefix. The second method estimates these variations utilizing the next symbol. Since mathematical analysis of the whole system becomes intractable in a long delay spread environment, we provide a mathematical analysis that investigates the effect of linearization for the case of a time-variant one path channel. Then our simulation results show how these methods would improve the performance in a highly time-variant environment with a long delay spread.

A new model for UHF propagation in large buildings is presented. This model relies on knowledge of the interior arrangement of the building without requiring much detail. The guiding of radiation along hallways is the most significant propagation process at distances of more than 10 m from the transmitter. The waveguide model predicts the power loss rate along the hallways, which is affected by the coupling among the propagating modes. The coupling results from the roughness of the surfaces in the building; it is predicted in an average manner using the average deviation of the walls from perfect smoothness.

In this paper, we present performance upper bounds for convolutional codes terminated with finite input sequence. To obtain the upper bounds, we define a generalized weight enumerator representing the relation between the Hamming distance of the coded output and the Hamming distance of selected input bits of the code. The upper bounds on frame error rate (FER) and average bit error rate (BER) of selected bits are obtained from the generalized weight enumerators. We present a low complexity method to compute the weight enumerator of terminated convolutional codes based on a modified trellis. We also present a generalized weight enumerator of single error event to provide tighter performance upper bounds by removing codewords composed of multiple error events from the weight enumerator.

Multiple-input-multiple-output (MIMO) systems have the potential to achieve very high capacities, depending on the propagation environment. Capacity increases as signal correlation decreases. We present the measurements of a MIMO system under strong and weak line-of-sight conditions. The system capacity decreases as the distance from the transmitter increases. Indeed the transmitter correlation increases as the distance increases. The receiver correlation is lower than the transmitter correlation under both propagation conditions.

Channel capacity in a waveguide with mode coupling is limited by a keyhole effect caused by the propagation mechanism. The coupling among the propagating modes reduces the channel rank, and at sufficient distances between the transmitter and receiver, the channel rank converges to 1. Measurements of the received power and channel capacity along a hallway and in rooms beside it, show a decrease of capacity with distance, which is explained by the keyhole effect in the hallway.

The Bell Labs layered space-time (BLAST) architecture provides high capacity wireless communications in rich scattering environments. Training sequences are transmitted periodically to estimate the channel. We compare the performance without channel tracking to the performance with interpolation-based channel tracking. The optimal training interval, training length and the maximum throughput are found through computer simulations for both schemes as functions of the Doppler frequency and the number of antennas. We show that the throughput is significantly improved with channel tracking for fast continuous fading channels. However channel tracking increases the computation complexity for data detection. A sub-optimal detection algorithm with lower complexity is proposed and investigated.

In the context of heterogeneous networks and diverse communication devices, real person-to-person communications can be achieved with a universal personal identification (UPI) that uniquely identifies an individual and is independent of the access networks and communication devices. Hierarchical mobility management techniques (MMT) have been proposed to support UPI. Previously proposed profile replication algorithms for improving the performance of such MMT are either threshold-based or off-line. In contrast to an off-line algorithm that assumes perfect knowledge of user calling and mobility statistics, we propose in this paper on-line replication algorithms that decide whether to distribute new replicas or delete existing replicas after serving each request, all based on the input sequence seen so far. The performance of our on-line profile replication algorithms is studied via large scale network simulations.

The paper compares performance of finite impulse response (FIR) adaptive linear equalizers based on the recursive least-squares (RLS) and least mean square (LMS) algorithms in nonstationary uncorrelated scattering wireless channels. Simulation results, in terms of steady-state mean-square estimation error (MSE) and average bit-error rate (BER) metrics, are found for the frequency-selective Rayleigh fading wireless channel experienced in a mobile ad hoc network where nodes are lognormally shadowed from each other. For the nonstationary channel models considered, RLS is always found to outperform LMS.

Waveguide theory is suggested for the analysis of UHF radiation in urban microcellular environments, where both the transmitter and the receiver are below rooftop level. The delay profile is calculated from the steady state power distribution among the modes of the waveguide. The delay profile is very asymmetrical because most of the power is contained in the low order (faster) modes, and because the differences between the group velocities of the modes tend to be smaller for the low order modes.

Fading is fundamentally different in ad hoc and cellular mobile wireless networks. Herein we show a way to simulate small-scale and large-scale fading on links in an ad hoc network and present simulation results for a multipath shadowed outdoor environment. We study fundamental discrepancies between ad hoc and cellular networks and find that link performance is worse for the ad hoc case, but, more interestingly, the performance gap shrinks with "increased mobility".

We have analyzed the behavior of a multiple element antenna system in an indoor environment based on the measurements taken in the Lucent Crawford Hill building with a system of 12 transmitters and 15 receivers. The parameters that we investigated were the distance between the transmitting and the receiving arrays, and the polarization of the transmitter/receiver elements. Our measurements show that the capacity of single polarization systems (only vertical or only horizontal) decays with distance in the presence of a strong line-of-sight component. The same is observed in the absence of a strong line-of-sight component if instead there is a dominant propagation path. Systems with elements of both polarizations perform better in terms of capacity in both environments.

In this paper we propose a pipelined structure for multistage parallel interference canceler for CDMA. This structure is easy to implement, has small processing delay, and applies to multipath fading channels and asynchronous users. Bit error rate (BER) performance with realistic channel estimation is simulated and compared to the performance with ideal channel estimation. Different methods are implemented to improve the performance, such as damping the update of the soft decisions, and performing partial interference cancellation.

Multiple input multiple output (MIMO) systems can provide high data rate wireless services in a rich scattering environment. We study one of the proposals for MIMO systems, the Bell Labs Layered Space-Time (BLAST) architecture. Channel estimation using training sequences is required for coherent detection in BLAST. We apply the maximum likelihood (ML) channel estimator and the optimal training sequences for block flat fading channels to continuous flat fading channels, and analyze the estimation error. The optimal training length and training interval that maximize the throughput for a given target bit error rate are found by computer simulations as functions of the Doppler frequency and the number of antennas.

In the context of heterogeneous networks and diverse communication devices, real person-to-person communication can be achieved with a universal personal identification (UPI) that uniquely identifies an individual and is independent of the access networks and communication devices. Hierarchical mobility management techniques (MMTs) have been proposed to support UPI. Traditional replication methods for improving the performance of such MMTs are threshold-based. We present optimal replication algorithms that minimize the network messaging cost based on network structure, communication link cost, and user calling and mobility statistics. We develop our algorithms for both unicast and multicast replica update strategies. The performance of these replication algorithms is studied via large scale network simulations and compared to results obtained from the threshold-based method.

We analyze the behavior of a multiple element antenna system in an indoor environment based on the measurements taken in the Lucent Bell Labs building in Crawford Hill, NJ, with a system of 12 transmitters and 15 receivers. In particular, we investigate the capacity behavior with respect to the polarization of the transmitting/receiving elements and the distance between the transmitting and the receiving arrays. The analysis of the power rolloff versus distance clearly demonstrates the different propagation characteristics of the horizontally versus the vertically polarized electric fields. Under strong line-of-sight (LOS) conditions (hallway environment), the power of the horizontally polarized waves falls off faster with distance than that of the vertically polarized fields. Also, the cross-polarization coupling is about -15 dB. Under nonline-of-sight (NLOS) conditions (labs), both polarizations display similar rolloff behavior with distance and the cross-polarization coupling is about 0 dB. There is a power loss of at least 15 dB under NLOS conditions relative to LOS conditions. The average received signal power affects the system capacity. In the hallway, horizontally polarized systems achieve lower capacities than their vertically polarized counterparts. Also the achievable capacity in the labs is much lower than that in the hallway, because of the lower average received power. The comparison of single polarization systems to hybrid polarization systems shows that combined polarization systems perform better in terms of achievable capacity. Therefore, there lies an advantage in using both electric field polarizations. However, under strong LOS conditions the channel itself inherently limits the capacity behavior of the system.

Multiple-input-multiple-output (MIMO) systems can provide high data rate wireless services in a rich scattering environment. We study one of the proposals for MIMO systems, the Bell Labs Layered Space-Time (BLAST) architecture. Channel estimation using training sequences is required for coherent detection in BLAST. We apply the maximum-likelihood channel estimator and the optimal training sequences for block flat fading channels to continuous flat fading channels and analyze the estimation error. The optimal training length and training interval that maximize the throughput for a given target bit error-rate are found by computer simulations as functions of the Doppler frequency and the number of antennas.

Systems with multiple element transmitter and receiver arrays have been shown to achieve very high spectral efficiencies. This is especially important in wireless applications that are power, bandwidth and complexity limited. The achievable capacity depends on the channel between the transmitters and the receivers, and a common assumption is that of a rich scattering environment. However an environment such as a hallway is more closely approximated as a waveguide, and we look into the limitations this poses on the achievable rates. We first show that the number of allowable propagation modes in a lossless waveguide limits the number of equivalent spatial sub-channels. We then apply a similar analysis to a lossy waveguide, where the losses are due to imperfectly conducting boundaries and imperfections in the dielectric. For demonstration purposes we show these effects for a waveguide of small dimensions. A hallway, which is a more practical environment for wireless applications, can be treated as an overmoded waveguide. In a hallway the losses are higher because the boundaries are dielectric materials that allow high power leakage. We again demonstrate the above effects, taking into account the losses off the walls.

This paper presents measurement results that demonstrate the different propagation characteristics of the vertical and horizontal polarizations and their crosscorrelation coupling under line-of-sight and non-line-of-sight conditions. We also present a simplified model for the propagation environment that uses an approach similar to the method of images, and achieves results that match the measurements.

In an outdoor communication environment, a percentage of bandwidth is wasted sending a training sequence for channel estimation and equalization. For instance in GSM, 17.93% of bandwidth is dedicated to the transmission of such a known sequence. Therefore, if a blind algorithm, using only the knowledge of input constellation and/or correlation, can achieve an acceptable performance, it will save bandwidth considerably. If the receiver does not have this information, blind adaptation would not be feasible with reasonable complexity. We present a robust blind adaptation structure for TDMA-based communication systems. Also, we include implementation issues such as differential coding and oversampling in our system modeling.

We consider fundamental limits on the symbol rate in frequency selective continuous fading channels. We show the relationship between the upper and lower bounds on the symbol rate set by the delay spread and the Doppler frequency of the channel respectively. We assume training-based channel estimation without tracking, and coherent detection without equalization. An approximation is derived for simulating channels with small delay spread using symbol period spaced samples. Simulation results with a two-path channel model are presented, including bit error rate (BER) and maximum throughput as functions of normalized delay spread for fixed products of delay spread and Doppler frequency. We also discuss the general implications of these results.

A new method is presented for estimating the mobile speed and the average received power in general wireless propagation environments. The locally stationary received signal is expanded in a local exponential basis using best basis methods of wavelet analysis. An estimate of the time-varying Doppler power spectrum is obtained together with an estimate of the maximum Doppler frequency, which is proportional to the mobile speed. The average received power is then estimated by integrating the time-varying spectrum. Simulations demonstrate good tracking of variable mobile speed and average received power for a wide range of angular distributions of incident power. The estimator is shown to perform significantly better than an adaptive averaging method described in the literature. The speed and average power estimates are also used to detect the corner effect in urban cellular systems to improve handoff performance and reduce the call dropping rate.

In the context of heterogeneous networks and diverse communication devices, real person-to person communication can be achieved with a universal personal identification (UPI) that uniquely identifies an individual and is independent of the access networks and communication devices. In this paper, we propose two location management techniques (LMTs) to support integrated personal communication services from heterogeneous networks. We first modify the existing HLR/VLR scheme with a meta-HLR database to map a UPI to multiple HLR addresses. To solve the scalability problem in this centralized approach, we propose a shared hierarchical LMT that not only scales better but also takes into account the locality of a user's call and movement patterns. The performance of these LMTs is studied via large scale network simulations.

A new technique is described for estimating the nonstationary mean signal received at a mobile station in a Rayleigh fading environment. The estimate is based on samples taken at the midpoints between the local minima of the received envelope. The continuous wavelet transform is used to estimate the local minima. An estimate of the mean signal is obtained using a fixed number of local minima. This technique requires neither an estimate of the mobile speed nor an adaptive temporal averaging window in contrast to other estimators. Simulations show that the mean signal is estimated well in a nonstationary environment with variable mobile speed.

This paper presents a new technique for estimating the speed of a mobile station in a wireless system. The mobile speed maps the characteristic spatial scale of the received signal into a characteristic temporal scale. The continuous wavelet transform tracks changes in the temporal scale to estimate the mobile speed as a function of time. This technique requires neither knowledge of the average received power of the nonstationary signal nor adaptation of a temporal observation window, in contrast to other speed estimators given in the literature. Simulation results indicate the ability to track a variable speed profile.

With the evolution of analog mobile wireless communications systems into digital second- and third-generation systems, there is growing interest in finding more efficient ways of managing the available resources, in particular radio spectrum and power. In the context of time division multiple access (TDMA) systems, this interest has led to the development of a variety of dynamic channel assignment (DCA) and power control (PC) algorithms. Despite the intense activity in both the DCA and the PC arenas and some proposals for combined DCA with PC, not enough work has been devoted to effectively integrating them. In this paper, the integration of DCA and PC is investigated and a family of integrated algorithms is presented. These algorithms, fully distributed and cost adaptive, achieve capacity levels significantly higher than those of a system with only DCA or PC. With respect to a system without any DCA or PC, several-fold capacity increases are obtained. Furthermore, these capacity levels are attained with user mobility included in the analysis.

A new technique is described for estimating the speed of a mobile station in a wireless system. The mobile speed maps the characteristic spatial scale of the received signal into a characteristic temporal scale. The continuous wavelet transform tracks changes in the temporal scale to estimate the mobile speed as a function of time. This technique requires neither knowledge of the average received power of the nonstationary signal nor adaptation of a temporal observation window, in contrast to other speed estimators given in the literature. Simulations show the tracking of a variable speed profile.

Increasing demand for wireless mobile communications, coupled with limited network resources, has motivated investigation into alternative efficient mobility management solutions. We consider wireless PCS call setup protocols. We propose a lightweight location lookup protocol, which significantly reduces the network signaling and setup delay during call setup. As an alternative, we consider a reverse connection setup protocol that also performs efficient call setup, while minimizing the cost of failed call attempts. The two protocols have different advantages, in particular when combined with location management techniques including HLR/VLR, HLR/VLR with partial replication, and HLR/VLR with caching. We study the combinations of call setup protocols with location management techniques and compare their performance to find the most efficient call setup schemes. Both analytical and simulation results are presented.

A new handoff algorithm based on recognition of patterns in the received signal power is presented. The handoff algorithm uses the constancy of the large scale signal variation with respect to a base station to improve handoff performance in comparison with the conventional hysteresis rule. A probabilistic neural network (PNN) is used for a pattern classifier. For a given environment, a training run is performed. A set of training patterns consists of averaged signal power samples (in dB) from nearby base stations within adjacent spatial windows. A probability of failure based on signal strength is defined and determines the possible serving base station(s) for each training class. A handoff is performed if a nearby class in the sequence of classes requires a handoff. Performance comparison of this handoff algorithm and the hysteresis rule is made for environments with four and five base stations. Simulation results indicate that for a fixed probability of failure, the pattern recognition-based handoff algorithm results in considerably fewer handoffs in comparison to the hysteresis rule. This reduction of unnecessary handoffs decreases the signalling load.

Indoor high-speed wireless data networks encounter signal fading and delay-spread multipath propagation. Hence, the realization of low error rate transmission requires measures to combat the performance degradation due to both signal fading and intersymbol interference (ISI). Receiver diversity has been known to be an efficient way of coping with the former problem, while adaptive equalization could be used to mitigate the effects of the latter. Incorporation of receiver diversity with adaptive equalization is therefore desirable. We propose a novel selection-diversity approach with an adaptive decision-feedback equalizer (DFE). In this method, selection is done on a symbol-by-symbol basis such that the output of the branch with the lowest estimated a posteriori probability of error is used as the final decision. This final (and hence more reliable) decision is used to adapt the DFE for all diversity branches. It is shown in this paper that the proposed selection rule is optimal for selection-diversity in the maximum a posteriori probability (MAP) sense. A very simple selection metric can be derived from this selection rule and practical ways of computing the selection metric are also presented. Simulation results show that the proposed method is very efficient. It is capable of achieving almost the same performance as an optimal [least squares (LS)], but computationally intensive, combining diversity approach. Furthermore, at an average bit error rate (BER) of 10/sup -4/, a gain of approximately 1.25 dB can be achieved over a previously proposed selection-diversity equalization approach.

The world of wireless telecommunications is rapidly changing. The capabilities of wireless networks are improving at a steady pace. This paper presents two possible protocols for implementing mobility for wireless users in an asynchronous transfer mode (ATM) network. The vision of the authors is of one "wireless ATM telecommunications network" that is capable of supporting a variety of today's applications with room to grow for advanced applications of the future. We first visit database architectures that can support mobility in a wireless ATM network. We then discuss one of two signaling architecture alternatives, the "overlay signaling", for overlay support of mobile users in the ATM-based wireless telecommunications network. "Overlay signaling" aims at minimizing the modification needed to the existing ATM protocols. We then describe a native "migratory signaling" approach that further integrates wireless and wireline users into one global wireless ATM network at the expense of requiring some modifications to the existing ATM protocols. A performance analysis of the proposed signaling architecture alternatives is also presented. We conclude by pointing out some challenges in merging ATM with wireless telecommunications.

We propose a method to eliminate the bias term present in the timing-error estimator employed in digital receivers where the input signal is sampled by a fixed clock which is not synchronized to the transmitter clock. This bias error results from the nonideal interpolation that precedes the timing-error estimator. We show that it can be derived as a function of the previously estimated symbol timings. An unbiased timing-error estimate can then be obtained by subtracting this bias term from the output of the timing-error detector. Simulation results are included to show the performance improvement realizable by employing this method.

We propose an optimal interpolation filter for symbol timing recovery in a digital receiver where the input analog-to-digital (A-D) conversion sampling clock is not synchronized to the transmitter symbol clock. The optimal filter is designed by minimizing the mean-square error (MSE) at the output of the receiver, assuming that the correct timing difference between the nonsynchronized input samples and the correct strobe synchronized to the symbol is given. The MSE minimization procedure results in a system of linear equations which can easily be solved to yield the optimal filter coefficients. We also analyze the symbol tracking performance of a symbol synchronizer that employs the proposed optimal interpolation filter. Although the proposed optimal filter is designed assuming the availability of the correct timing difference, we show that it minimizes the timing estimation error variance even when the computed timing difference values deviate from the ideal, we also show that the timing estimation is virtually unbiased if the length of the interpolation filter is greater than or equal to 4. Simulation results are included to show the performance improvement realizable by employing the optimal interpolation filter.

A rapid synchronization method is presented for an orthogonal frequency-division multiplexing (OFDM) system using either a continuous transmission or a burst operation over a frequency-selective channel. The presence of a signal can be detected upon the receipt of just one training sequence of two symbols. The start of the frame and the beginning of the symbol can be found, and carrier frequency offsets of many subchannels spacings can be corrected. The algorithms operate near the Cramer-Rao lower bound for the variance of the frequency offset estimate, and the inherent averaging over many subcarriers allows acquisition at very low signal-to-noise ratios (SNRs).

The world of wireless telecommunications is rapidly changing. The capabilities of wireless networks are improving at a steady pace. This paper presents a protocol implementation for providing mobility in an asynchronous transfer mode (ATM) network. The vision of the authors is of one "ATM-based wireless telecommunications network" that is capable of supporting a variety of today's applications with room to grow for advanced applications of the future. The next generation wireless networks are called personal communication services (PCS) networks and will provide for global user mobility, a life-long user identification numbering system and the ability to accommodate varying data rates.

We consider the problem of managing the information required to locate users in a wireless communication system, with a focus on designing and evaluating location management techniques that are efficient, scalable, and flexible. The three key contributions of this paper are: (1) A family of location management techniques, HiPER (for Hierarchical ProfilE Replication), that efficiently provide life-long (non-geographic) numbering with fast location lookup; (2) Pleiades, a scalable event-driven wireless system simulator with realistic calling and mobility patterns derived from several months of real traffic traces; and (3) multi-day simulations comparing our proposed location management techniques with current and previously proposed techniques on a realistic geographical and network topology. Research supported by the Center for Telecommunications and the Center for Integrated Systems at Stanford University, and by equipment grants from Digital and IBM Corporations. 1 Introduction I...

A new method for estimating the frequency of a complex sinusoid in complex white Gaussian noise is proposed. Its computational complexity is comparable to Kay's method [1989], but it attains the Cramer-Rao bound (CRB) down to lower signal-to-noise ratio (SNR) values. Simulation results are included to demonstrate the performance of the proposed method.

In a high capacity personal communication system (PCS), for a given bandwidth, co-channel interference (CCI) limits the system capacity. Low-complexity diversity combining algorithms and circuit architectures for co-channel interference cancellation and frequency-selective fading mitigation, which do not require training sequences, are introduced. Results obtained from computer simulation of hardware show that two-antenna diversity combining gives wireless communication systems a signal to interference ratio improvement of at least 3 dB over conventional two-antenna selection diversity. The technique is also effective in mitigating frequency-selective fading without using conventional equalization-an average irreducible word error rate (WER) of 2.4% is obtained in computer simulation of hardware for radio channels with normalized delay spread of 0.3. In contrast, for the same WER, selection diversity and single antenna without diversity can sustain normalized delay spread up to about 0.16 and 0.06 respectively.

It is suggested that low-power digital radio as an access technology could be integrated into a local exchange network to provide a ubiquitous personal communications network (PCN). High-quality tetherless communications services that could be provided by such an exchange-network-based PCN are described. A possible low-power exchange access digital radio system for providing these exchange-network based PCS services is discussed. The radio system uses a spectrum-efficient time-division multiple-access (TDMA) architecture made possible by advanced digital signal processing techniques. Control of the frequency reuse system is described, and frequency spectrum needs are indicated.>

In our highly mobile society, the provision of voice and data communications to a person away from his/her wireline telephone has become a major communications frontier. The early penetration of this frontier has been based on very limited portable communications approaches, e.g., cordless telephones, mobile radio telephones, and radio paging. Each of these approaches only partially satisfies portable communications needs. This paper describes an approach to providing universal digital portable communications integrated into telephone networks. A system configuration employing time-division multiple-access radio link architecture and frequency reuse is described. Issues affecting radio link transmission rates and radio system coverage are discussed. Characteristics and parameters of a possible system to supplement the wire (or fiber) loop are indicated.

Providing voice and data communications to people away from their wireline telephones has become a major communication frontier. This frontier is being penetrated by evolving approaches to portable communications, e.g., cordless telephones, mobile radiotelephone, and radio paging. However, these approaches have many limitations; none can provide universal portable communications services. This paper discusses limitations of the evolving approaches and considers objectives and approaches for providing more universal digital portable communications as an integrated part of telephone exchange networks. These more universal communications could be accomplished by using demand-assigned radio links for the last thousand feet or so of telephone loops and sharing the remainder of the fixed distribution facilities. Fixed radio ports as integrated parts of telephone distribution networks could be placed throughout service regions. Efficient use of the radio spectrum could be insured by the planned reuse of radio frequencies throughout the regions. The severe multipath radio propagation environment within and around buildings that strongly influences the design of portable communications systems is described in the paper. System configurations and radio link techniques, that can provide reliable communications in the multipath environment, are discussed. Radio system calculations are illustrated for radio ports with 30-ft-high antennas in residential areas. The calculations indicate that radio link availability would be greater than 99 percent for 2000-ft port separations and 5-mW portable transmitters. Reuse of frequencies would require dividing the allocated frequency band into segments for use at adjacent ports. Calculations suggest that link availability in the cochannel interference environment would be greater than 99 percent, if 25 to 35 segments were used in residential areas.

Antenna diversity can mitigate signal impairments caused by random angular orientation and multipath radio propagation when using portable radiotelephones. Cumulative distributions of signal-to-noise ratio (S/N) were determined for antenna diversity using realistic orientation and multipath propagation models. In a random orientation and multipath propagation environment with -6 dB average crosspolarization coupling, two-branch selection diversity with two perpendicular antennas yields anS/Ndistribution with the same slope as two-branch selection diversity in the fixed-oriented mobile radio environment. The distribution for random orientation is about 4.5 dB worse, however, than the mobile radio distribution.

Signal attenuation due to rain is the primary source of outages on earth-satellite radio links operating above 10 GHz. Measurements of the durations of rain fades are also useful in assessing link performance since the effect of many short outages is clearly different than that of fewer long outages. In this paper the authors present a set of measurements of the number and duration of rain fades measured at 19.04 GHz along an earth-satellite path with 18.5 degree elevation. Since rain attenuation is seasonally dependent, one would expect some seasonal variation in at least the rate of occurrence, and possibly the durations of rain fades. This paper briefly examines this dependence. Another factor affecting the performance of earth-satellite RF links is the distribution of time intervals between rain fades, called here `interfade intervals'. If the probability of short interfade intervals were high, this could adversely influence the link performance for services with long holding times. This paper presents measured distributions of these intervals.

Frequency reuse small-coverage-area radio systems having hexagonal and square coverage areas are compared. Comparison is made on the basis of average signal to average interference (S̅/I̅) in the corners of the areas and on the basis of the expected probability ofS/Iexceeding some system threshold for at least one base station that is eligible to provide service. The difference in performance between square and hexagonal systems is small, smaller than the usual uncertainties in the propagation parameters needed in the performance estimates. Results suggest that, even if the signal strength decreases as slowly as the inverse cube of the distance and the standard deviation of the large scale signal variation is as large as 10 dB, good service probabilities (on the order of 99 percent) can be provided in small-coverage-area radio systems using 30-40 channel sets.

Relationships between rain and ice attenuation and depolarization at 19 GHz have been determined for linear incident polarizations ± 4 and\pm 8\degfrom vertical. These relationships are based on measurements made using beacon transmissions from a COMSTAR satellite. A technique is used to bound depolarization magnitude at any incident polarization from a depolarization magnitude measurement made at another polarization. This technique makes use of previous observations that depolarization is minimal at vertical and horizontal incident polarizations. The bounds are compared with the results from the COMSTAR measurements.

Three experimental programs at Bell Laboratories, Crawford Hill, NJ, have collected rain attenuation data at three frequencies in the 10-30 GHz range for over 2 years. As expected from a simple geometrical model, rain attenuation statistics scale approximately as the cosecant of the path elevation angle. Thus, greater attenuation is encountered on lower elevation angle paths. The increased rain attenuation at higher frequencies is illustrated by comparing cumulative rain attenuation distributions at three widely separated frequencies. Typical year-to-year variations in these distributions are presented. Periods of severe rain attenuation are shown to occur more frequently during the summer months and during the afternoon hours. Cumulative attenuation distributions are compared for these and other time periods.

Cumulative attenuation distributions for one year and relationships between attenuations at 19 GHz and 28 GHz are presented for an earth-space propagation path terminating at Crawford Hill, New Jersey. An example from the distributions indicates that 20 dB attenuation was exceeded for 0.012% of the year at 19.04 GHz and for 0.06% at 28.56 GHz. For both frequencies the polarization was linear and oriented 21° from vertical at the receiver. When the attenuation at 19 GHz was less than 20 dB, the attenuation in dB at 28 GHz averaged 2.1 times that at 19 GHz. This ratio was lower at higher attenuations.

The performance of large scale multicell mobile radio systems using dynamic channel assignment and having limitations on the number of servers available in each coverage cell was investigated by computer simulation. At low system blocking the traffic carried is determined solely by the server limitations while at higher blocking the traffic carried is determined by channel limitations.

A simple technique, time division circular polarization (TDCP), is described which permits the simultaneous measurement of attenuation and depolarization for two orthogonal linearly polarized waves and two orthogonal circularly polarized waves in a slowly varying linear propagation medium using only a single frequency beacon switched between two linearly polarized antenna ports.

LINC is a technique that uses signal processing to produce Linear amplification of bandpass signals with grossly Nonlinear circuit Components. The important signal processing functions in LINC are forming two constant envelope phase modulated signal components from the bandpass input signal and recombining the amplified components to produce an amplified replica of the input signal. Two- and three-tone laboratory tests of a complete LINC amplifier show that, at full output, spurious levels 40 dB below tone level are achievable for tone spacings of 100 kHz. Because the laboratory model is operated at relatively low frequencies (hundreds of MHz), scaling up in frequency should result in a LINC with\sin^{-1}phase modulator that is less sensitive to delay in a feedback loop and a wider-bandwidth lower-distortion input limiter.

A new technique for linear amplification of envelopevarying bandpass signals using sampling techniques (LIST) and devices having nonlinear input-output relationships is described. Two component signals are produced by phase-reversal modulating two quadrature carriers with two delta-coded low-pass envelopes derived from the quadrature components of a bandpass signal. The component signals either are produced at a high level or are amplified in nonlinear amplifiers. An amplified replica of the input signal results from passively summing and bandpass filtering these two component signals. A laboratory demonstration shows that a LIST amplifier with spurious 40 dB below desired signals in a two-tone test is easily achieved.

Distributions of delay spread and correlation bandwidth at 0.9 and 0.5 correlation for Gaussian wide-sense stationary uncorrelated scattering (GWSSUS) channels associated with 100 small-scale areas at different locations within a 2 × 2.5 km region of New York City are presented. For delay spread the maximum value observed was3\frac{1}{2};\mus and l0 percent of the areas exceeded2\frac{1}{2}\mus; for correlation bandwidth at 0.9 correlation the minimum was 20 kHz and 10 percent of the areas were less than 30 kHz; for correlation bandwidth at 0.5 correlation the minimum was 55 kHz and 10 percent of the areas were less than 130 kHz. The region is representative of the heavily built-up areas of many large cities in the United States.

LINC is a technique that uses signal processing to produce linear amplification of bandpass signals with grossly nonlinear circuit components. Two important signal-processing functions of LINC are 1) forming two constant envelope phase-modulated signal components from the bandpass input signal and 2) recombining the amplified components to produce an amplified replica of the input signal. Two-tone laboratory tests of a LINC component separator and combiner, not including a limiter and envelope detector, show that, at full output, spurious levels 40 dB below tone level are achievable over a 1-MHz band. Because the laboratory model operated at relatively low frequencies (hundreds of megahertz), scaling up in frequency should result in a LINC with-1phase modulator that is less sensitive to delay in a feedback loop.

A technique for producing bandpass linear amplification with nonlinear components (LINC) is described. The bandpass signal first is separated into two constant envelope component signals. All of the amplitude and phase information of the original bandpass signal is contained in phase modulation on the component signals. These constant envelope signals can be amplified or translated in frequency by amplifiers or mixers which have nonlinear input-output amplitude transfer characteristics. Passive linear combining of the amplified and/or translated component signals produces an amplified and/or translated replica of the original signal.

Small scale statistics of multipath propagation in a heavily built-up urban mobile radio environment are presented. The statistics cover vehicle travel distances on the order of 30 m along streets. Measuring equipment time delay resolution is about 0.1 μs. In some locations, paths with significant amplitudes are observed with excess delays of 9 to 10 μs. The delay spreads (\sqrt{second central moment of power delay profile}) in this environment are on the order of 2 μs. Often the signal at fixed delays has a Rayleighdistributed amplitude but large departures from the Rayleigh distribution also occur. From the measurements it appears reasonable to model the urban mobile radio channel as a Gaussian quasi-wide-sense stationary uncorrelated scattering channel within a bandwidth of 10 MHz and for intervals along the street of up to 30 m.

Large-scale mobile radio systems using dynamic channel REassignment switch the channels assigned to some of the calls in progress to maintain a nearly optimum separation between coverage areas simultaneously using the same channel. Dynamic channel REassignment produces a significant increase in channel occupancy. For example, in the systems studied, which have a uniform distribution of their fixed channels and were operated against a uniform spatial distribution of offered traffic, the channel occupancy was increased by\frac23over a pure fixed channel assignment system at a blocking rate of one percent. This corresponds to a channel savings of about 40 percent for the same traffic carried at one percent blocking by the hybrid systems that were studied.

When more than one radio channel is available to serve a mobile customer in a large-scale mobile communications system the strategy used to make the assignment has an effect on system performance. Performance characteristics for five different assignment strategies obtained from a computer simulation are compared in this paper. It was found that the strategy that chooses a channel to be assigned at a base station so that the distance to the next base station using the same channel is a minimum, performs better than a strategy that minimizes the mean square distance to the next "in use" base stations on both sides. Also, it was found that choosing the first available channel encountered in a channel search is a strategy that produces reasonably good system performance.

Dynamic channel assignment in mobile communications systems is able to cope with varying nonuniform spatial user demand. It also readily handles the normal randomness in offered traffic. This study presents system performance data for radio systems having 8, 20, 40, 80, and 120 radio channels available and a radio-channel reuse interval of 4 base-station coverage areas. Data are also presented that compare performance of a 30 radiochannel system with a reuse interval of 3 base stations, a 40 radiochannel system with a reuse interval of 4 base stations, and a 50 radio-channel system with a reuse interval of 5 base stations. The results are obtained from a computer simulation of a onedimensional network laid out along a line. As the number of radio channels available to a communication system increases, the system becomes more efficient in handling the fluctuations in offered traffic. Changing the radio-channel reuse interval also affects the per-channel traffic carried by a dynamic-channel-assignment system.