In this paper, we present a feasibility of wideband analog self-interference cancellation with a newly proposed digitally controlled analog adaptive filter. We analyze the key existing issues and concerns in analog interference cancellation and propose novel solutions to cancel the self-interference with a simple 1-tap cancellation circuit. The proposed solution conducts a blind estimation of optimal analog filter coefficients in the digital domain without requiring a pilot signal. We present an end-to-end system analysis and 1-tap adaptive filter circuit analysis to show the feasibility of wideband cancellation. We then propose a few novel blind coefficient estimation algorithms and show their cancellation performance by both simulations and measurements from an actual prototype. The results suggest that our self-interference approach enables the simultaneous operation of the licensed and unlicensed technologies in the adjacent spectrum within a single device, which, in turn, facilitates graceful spectrum sharing and bandwidth aggregation between the licensed and unlicensed technologies.

Energy efficiency of mobile data communication systems is of significant importance. In this paper, we propose an adaptive radio connection management technique that can effectively reduce the mobile battery consumptions by intelligently reducing the radio connection tail overhead. This adaptive fast dormancy (AFD) algorithm learns the device-specific traffic behavior over time and determines the optimal traffic inactivity timer for the device. In order to maximize the battery saving gains while precisely controlling the false alarm probability of the fast dormancy decision by the mobile, we formulate a stochastic approximation problem and develop a low-complexity adaptive algorithm. Simulation results for UMTS networks demonstrate the efficacy and convergence of the AFD algorithm.

Active hand-in refers to the ability for a voice or data connection for a mobile on the macrocell to be seamlessly handed off to the target femtocell. The challenge for active hand-in is that a large number of femtocells share limited PN offsets in a geographical region and hence the target femtocell cannot be identified uniquely by its PN offset. An ideal solution would be one that requires no hardware or software changes to macro network or legacy mobiles. This paper discusses a method of using an out-of-band (OOB) radio integrated with the femtocell to assist in active hand-in by uniquely identifying the target femtocell with the help of OOB link identifiers.

In this paper, we consider a network scaling approach to realize network energy savings (NES). As cellular networks are typically deployed to meet the high traffic demand during the busy hour, a scaled down version of the network should be able to effectively handle the low traffic load in the off-peak hours without compromising the network performance. In order to demonstrate the realistic utility of this idea, we conducted a network energy savings analysis for a commercially deployed dual-carrier UMTS network. In an urban area, it is found that 20% - 35% of the cells can be switched off during the off-peak hours even after turning off one carrier entirely. Based on the base station energy consumption model developed from the measurements in commercial base stations, this translates to 11% network energy savings. In a suburban area, on the other hand, it is found that approximately 12% of the sites can be switched off after disabling one carrier. This still offers 8% network energy savings.

For UWB to be a competitive solution in mobile devices, power consumption of the modem has to be optimized. The focus of this paper is to decrease the power consumption of the modem at physical layer by trading frequency diversity when operating at data rates up to 200 Mbps and by changing the constraint length of convolutional code and using QPSK when operating at data rates 320 Mbps and up. We show more than 50% and 25% gains in power consumption at physical layer with additional 1 dB and 3 dB received SNR at data rates 53.3 Mbps and 480 Mbps, respectively.

Modern digital communication systems use quadrature modulation (IQ) such as QAM. An imbalance between the I and Q paths in the RF module results in self-interference that affects both channel estimation and data demodulation. We propose unbiased training sequences that automatically eliminate the interference from the channel estimate. Strong gains are achieved for high data rates. For additional gains, we also propose a relatively inexpensive extension to compensate for data distortion in OFDM. End-to-end (transmitter and receiver) and frequency-dependent imbalances are covered.

In this paper, we investigate channel estimation for a single-carrier direct-sequence ultra-wideband (DS-UWB) system based on the realistic IEEE P802.15.3a UWB channel model. In the preamble of each packet, pilot symbols are spread by the pseudorandom acquisition code and sent through the channel. At the receiver, the well known sliding window algorithm is applied to estimate the multipath channel coefficients. For channel models CM1 through CM4, we determine the length of the acquisition code to mitigate the impact of intersymbol interference on channel estimation with negligible redundant search space. We also develop a generic approach to design the pilot sequence length based on the UWB channel statistics. Furthermore, a general analytical framework is developed to accurately evaluate the symbol error rate performance with imperfect Rake tap coefficients. Simulation results are shown to verify the theoretical evaluation.

Abstract Wireless Service Providers have become increasingly interested in developing position location services for commercial application of the evolving Third Generation (3G) communication systems. The initial and driving force behind this interest is a regulation in the United States promulgated by the Federal Communications Commission, requiring wireless carriers to be capable of delivering the position of a wireless device (or mobile station) making an emergency call to emergency authorities by October 2001. Similarly in Europe, the 1999 Communications Review (COM 1999/539) set the date of 1 January 2003 for the carriers to make location information available for emergency authorities. The cellular mobile station geolocation problem can be solved using either network‐based methods or handset‐based methods. A hybrid position location method that combines measurements from the communication system's forward and/or reverse links with measurements from the Global Positioning System (GPS) can improve position service availability. Given the ability to accurately determine a user's location means that position location service applications can and most certainly will provide value‐added features such as navigation, fleet management, real‐time traffic updates, location sensitive billing, etc. In this paper we present different geolocation technologies, list some of the important applications, and review the status of the location services in the 3G Standards Development Organizations. Copyright © 2002 John Wiley & Sons, Ltd.

A maximum-likelihood decision-directed receiver structure is derived for a data transmission system which employs quadrature-quadrature phase shift keying (Q/sup 2/PSK) as the modulation scheme. The receiver is derived by treating the problem of sequence decision, sample timing, and carrier phase as a composite estimation problem. It is shown that the resulting maximum-likelihood estimates (MLEs) are asymptotically unbiased, and the variances of these estimates are derived. In this receiver, symbol detection is carried out by means of a modified Viterbi algorithm (MVA) while a parallel synchronizer is used for carrier phase and delay recovery. As a measure of performance, an error-event probability bound is derived for the MVA. The effect of intersymbol interference, (ISI) present at the two matched filters' outputs, on the error bound is analyzed and illustrated. It is shown that the resulting MVA and the parallel estimator are well suited for systolic array implementation.>

An automatic modulation classification algorithm utilizing the statistical moments of the signal phase is developed and used to classify the modulation type of general M-ary PSK signals. It is shown that the nth moment (n even) of the phase of the signal is a monotonic increasing function of M. On the basis of this property, the authors formulate a general hypothesis test, develop a decision rule, and derive an analytic expression for the probability of misclassification. Two examples are given to demonstrate the performance of the algorithm. The algorithm is compared with the quasi-log-likelihood radio (qLLRC), square-law (SLC), and phase-based (PBC) classifiers. The algorithm is outperformed by qLLRC at low CNR but is comparable to SLC and is better than PBC. The qLLRC algorithm is only valid at CNR>

The authors consider a number of topics concerned with coding schemes over fading dispersive channels. They begin with comparing the performance of the energy receiver with the optimum and envelope receivers. Different coding schemes are used to improve the performance. The effect of channel spread on the performance of single-state Reed-Solomon (RS) codes is investigated. Channel information (CI) is discussed as a means for further improvement. The simple erasure criterion used is suboptimal, but gives a measure of the gain that can be achieved by using CI. Of special interest is the use of a concatenated coding technique for forming codes of large distance. Concatenated codes using short RS codes with CI as inner codes and RS outer codes are compared. It has been shown that the improvement obtained by concatenation is significant for overspread channels. This is attributed to the implicit diversity introduced by the channel. Furthermore, a concatenated system using convolutional codes as inner codes performs better than the system using RS inner codes.>

An optimum closed-loop symbol synchronizer operating over fading dispersive channels is developed. It is shown that the optimum tracking synchronizer is a multiplier that forms the product of the received signal and a filtered version of it, the filter impulse response is the first derivative with respect to time (or frequency) of the autocorrelation function of the signal part of the received waveform. It is proved that for small misalignment the control signal vector is an incrementally linear function of the error vector. For a Gaussian-shaped scattering function, a simple closed-loop structure is obtained and expressions for tracking error statistics are derived. The dependency of such statistics on the signal-to-noise ratio and on the spread of the channel is investigated.>

An investigation is reported of an acquisition system for digital communication over fading dispersive channels that is optimum in the maximum-likelihood sense. A performance measure for the synchronizer in the acquisition mode is defined. Upper and lower bounds on the performance measure are derived. The effective bounds on parameters, such as signal-to-noise ratio, pulse width and modulation, number of pulses, spread of the channel, and the size of the resolution cell, on the performance of the synchronizer is investigated.>

A generalized spread ambiguity function is defined, and its relationship to P.M. Woodward's (1953) ambiguity function for different models is developed. A derivation and discussion of the properties of the spread ambiguity function are presented. It is shown that smoothing the ambiguity function by the scattering function of the channel does not alter the properties of the ambiguity function. Measures of RMS correlation time and RMS correlation bandwidth of nonstationary processes are defined. An example is given to demonstrate how spread ambiguity functions can be used to help in signal design when dealing with doubly spread targets.>