E.G. Strom, S. Parkvall, S.L. Miller, and B.E. Ottersten, "Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems," IEEE Trans. on Comm., 44(1):84-93, Jan 1996.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....However, these multiuser detectors all need perfect time delay information for all the users at the basestation receiver. In a practical CDMA communication system, time delays have to be estimated. Existing delay estimation methods include the sliding correlator [6] subspace based algorithms [7] [8] single user maximum likelihood [9] and the large sample maximum likelihood (LSML) algorithm [10] For an observation length of 100 symbols, current sliding correlator delay estimation methods [6] can achieve a delay estimation error within 0.2 T c , and the subspace based MUSIC algorithm ....

....sample maximum likelihood (LSML) algorithm [10] For an observation length of 100 symbols, current sliding correlator delay estimation methods [6] can achieve a delay estimation error within 0.2 T c , and the subspace based MUSIC algorithm can achieve 0. 03 T c , where T c is the chip duration [7]. The e#ect of imperfect time delay estimation, i.e. delay mismatch, on the performance of multiuser detectors has been investigated in [11] 12] 13] 14] 15] It is shown that these multiuser detectors are sensitive to delay mismatch: bit error rate (BER) performance degrades greatly W. Zha ....

[Article contains additional citation context not shown here]

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersten, "Propagation delay estimation in asynchronous direct-sequence codedivision multiple access systems," IEEE Trans. Commun., vol. 44, pp. 84-93, Jan. 1996.

....channels in [94] and [55] In [36] a channel parameter estimation method is proposed for antenna array CDMA systems, which is not near far resistant. Joint parameter estimation and multiuser signal detection approaches are studied for single antenna CDMA systems in [33] 81] and [108] In [2] [83] and [89] subspace based channel parameter estimators are proposed for multi user CDMA systems. Comparative studies for blind channel estimation schemes are provided for multipath CDMA channels in [51] Recently proposed channel estimation techniques for TDMA systems can be found in [61] 86] and ....

....still suffers from timing errors. Therefore, higher precise approaches to time delay estimation are required to achieve acceptable performance when the time delays are not available. Time delay estimation for multi user CDMA systems using subspace based techniques have been proposed in [2] and [83]. In [47] it is observed that using a basestation antenna array can increase dimension of the noise subspace in [2] and hence increase the number of users in the system whose parameters can be correctly estimated. The application of the EM algorithm to time delay estimation is first reported in ....

E. Strom, S. Parkvall, S. Miller and B. Ottersten, "Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems", IEEE Transactions on Communications, vol. 44, no. 1, pp. 84--93, Jan. 1996.

....each reference s position. The estimation of reference code time of arrival is hindered by the near far effect due to the design of the HP, power control is not feasible. The joint estimation of code delays has previously been considered in the CDMA literature, using algorithms such as MUSIC [2] and channel subspace identification [3] Here, a new algorithm for multiuser code delay estimation (MCDE) based on expectationmaximization [4] is presented. It is shown that systematic application of the EM technique to MCDE leads to a parallel interference cancellation (PIC) structure. The ....

E.G. Strom, S. Parkvall, S.L. Miller, and B.E. Ottersten, "Propagation delay estimation in asynchronous directsequence code-division multiple-access systems," IEEE Transactions on Communications, vol. 44, pp. 84--93, Jan. 1996.

....between multipaths closer than 1.5 chips distance. In order to increase the resolution, an alternative to the code tracking loops are the superresolution methods. The most known superresolution algorithm for delay estimation is the Multiple Signal Classification (MUSIC) algorithm [6] 7] [9], 8] 10] One of the main advantages of MUSIC algorithm is its near far resistance [6] On the other hand, the drawbacks of MUSIC algorithm are its high complexity [9] and its sensitivity to high levels of noise [7] More recently, the authors introduced a Teager Kaiser (TK) operator based ....

.... The most known superresolution algorithm for delay estimation is the Multiple Signal Classification (MUSIC) algorithm [6] 7] 9] 8] 10] One of the main advantages of MUSIC algorithm is its near far resistance [6] On the other hand, the drawbacks of MUSIC algorithm are its high complexity [9] and its sensitivity to high levels of noise [7] More recently, the authors introduced a Teager Kaiser (TK) operator based algorithm for increased resolution in CDMA systems [11] 12] It was also shown that this algorithm is near far resistant [12] and has the advantage of a very simple ....

E.G. Strom, S. Parkvall, S.L. Miller, B.E. Ottersten, " Propagation delay estimation in asynchronous directsequence code-division multiple access systems", IEEE Trans. on Communications, vol. 44, no.1, pp. 84-93, Jan. 1996.

....that work and develop subspace based blind techniques to combat both MAI and ISI in the highrate dispersive CDMA systems. Several recent works have addressed the use of the subspace based MUSIC type of methods for parameter estimation in CDMA systems, such as delay and channel estimation [1] [17]. The major contribution of this work is the development of blind techniques for blind joint suppression of MAI and ISI in dispersive CDMA channels. 0090 6778 98 10.00 1998 IEEE 92 IEEE TRANSACTIONS ON COMMUNICATIONS, VOL. 46, NO. 1, JANUARY 1998 In this paper, the communication channels are ....

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersten, "Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems," IEEE Trans. Commun., vol. 44, pp. 84--93, Jan. 1996.

.... or cyclic methods with fractional sampling [141] Especially for code division multiple access (CDMA) systems entailing asynchronous users, correct timing acquisition affects performance considerably, and a variety of methods (including subspace approaches) have been proposed recently (see e.g. [383], 25] Blind equalizers of FIR frequency selective channels do not need to acquire timing because they absorb it in the channel itself. In addition, methods that rely on the constant modulus (restoral) algorithm (CMA) do not require frequency offset estimation [143] 416] CMA is basically a ....

.... packets (e.g. in distributed networks and wireless PCS prototypes) or, when the propagation medium is rapidly varying (e.g. in large cells with considerable delays and high data rate time varying wireless environments) Such self recovering CDMA receivers were proposed recently in [420] 474] [383], 241] 25] and [419] They capitalize on code diversity offered by the user code(s) and the received data but have relatively high complexity especially because they adopt subspace decomposition (via the SVD) of large matrices for signature waveform estimation. Inverse filtering criteria and ....

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersten, "Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems," IEEE Trans. on Comm., pp. 84--93, Jan. 1996.

....In this paper, subspace type techniques, such as MUSIC [9, 11] and different variants of it, are studied in CDMA reception. It has already been demonstrated that subspace methods provide useful solutions of intermediate complexity and performance to the parameter estimation problem discussed above [1, 8, 10]. We use a modification of the ACM MUSIC [3, 7] method for delay estimation. Hierarchic ACM MUSIC appoach has been studied in [5] We use ACM MUSIC type method for tracking the slowly varying delays. We also introduce an optimal reference approach, which assumes the prior knowledge about the ....

....the prior knowledge about the fadings and symbols. In the simulations, we compare the new methods to the MUSIC. 2 Data model Collect C vectors r m from subsequent discretized data samples r[n] r m = Theta r[mC] r[mC 1] Delta Delta Delta r[ m 1)C Gamma 1] T (2) They have the form [1, 8, 10] r m = K X k=1 L X l=1 [ff kl;m Gamma1 g E kl ff klm g L kl ] nm (3) where nm denotes noise vector, ff klm = a klm b km , and g E kl = Theta s k [C Gamma d kl 1] Delta Delta Delta s k [C] 0 Delta Delta Delta 0 T (4) g L kl = Theta 0 Delta Delta Delta 0 s k ....

E. Strom, S. Parkvall, S. Miller, and B. Ottersten, "Propagation Delay Estimation in Asynchronous DirectSequence Code Division Multiple Access Systems", IEEE Transactions on Communications, vol. 44, January 1996, pp. 84-93.

....studied extensively, because such systems are highly useful in mobile radio communications and other applications. In the uplink signal processing, the codes of the users are known but their delays are unknown. For delay estimation, one can use e.g. mathced filter [9] subspace approaches [1, 8, 10], or maximum likelihood method [11, 9] or its approximations [5] When the delays have been estimated, one can estimate the other parameters such as fading process and symbols. A downlink or mobile phone signal processing differs dramatically from the uplink or base station signal processing. ....

....approach is a GTM algoithm applied to the binary data. It allows more users than the code length. 2 Data model Collect C vectors r m from subsequent discretized equispaced data samples r[n] r m = Theta r[mC] r[mC 1] Delta Delta Delta r[ m 1)C Gamma 1] T (3) They have the form [1, 8, 10] r m = K X k=1 b k;m Gamma1 L X l=1 a l g E kl b km L X l=1 a l g L kl # nm (4) where nm denotes noise vector, and the early and late parts of the code vectors are g E kl = Theta s k [C Gamma d l 1] Delta Delta Delta s k [C] 0 Delta Delta Delta 0 T ....

E. Strom, S. Parkvall, S. Miller, and B. Ottersten, "Propagation Delay Estimation in Asynchronous Direct-Sequence Code Division Multiple Access Systems ", IEEE Transactions on Communications, vol. 44, January 1996, pp. 84-93.

....straightforward approaches. In addition, the algorithm yields very good estimates to the codes. 2 Subspace model Collect C vectors r m from subsequent discretized equispaced data samples r[n] r m = Theta r[mC] r[mC 1] Delta Delta Delta r[ m 1)C Gamma 1] T (2) They have the form [1, 10, 12] r m = K X k=1 L X l=1 [ff kl;m Gamma1 g E kl ff klm g L kl ] nm (3) where nm denotes noise vector, ff klm = alm b km , and g E kl = Theta s k [C Gamma d l 1] Delta Delta Delta s k [C] 0 Delta Delta Delta 0 T (4) g L kl = Theta 0 Delta Delta Delta 0 s k [1] ....

....H m ) GR ff G H oe 2 I (12) R ff is the autocorrelation matrix of the amplitudes ff klm . The variance of the noise is oe 2 . In the uplink signal processing, the codes are known but their delays are unknown. For delay estimation, one can use e.g. mathced filter [11] subspace approaches [1, 4, 5, 10, 12], or maximum likelihood method [13, 11] or its approximations [3] The problem in the downlink is that the mobile phone knows only the code of its own. Because the matrix G contains the code information, it should be directly estimated without any parametric fitting as in matched filter, subspace ....

E. Strom, S. Parkvall, S. Miller, and B. Ottersten, "Propagation Delay Estimation in Asynchronous DirectSequence Code Division Multiple Access Systems", IEEE Transactions on Communications, vol. 44, January 1996, pp. 84-93.

....can later be used by the multiuser detectors. Multiuser decoding is required since every practical system is coded for error control. Optimal and suboptimal multiuser algorithms for decoding have been proposed in [11, 12, 13, 14, 15, 4 16] while those for channel estimation have been proposed in [17, 18, 19, 20, 21, 22]. These algorithms are developed for CDMA systems with short spreading codes for each user that repeat every symbol. However, current and next generation wireless systems use long spreading codes whose period is much larger than the symbol duration. Therefore, many of these algorithms are ....

....interference limited since they allow multiple users to access the same bandwidth at the same time. Receivers can combat multiple access interference by using multiuser channel estimation, detection and decoding algorithms. Several multiuser algorithms have been proposed for channel estimation in [17, 18, 19, 20, 21, 22]. These algorithms are developed for CDMA systems with short spreading codes for the various users that repeat every symbol. However, spreading codes used in practical CDMA systems have a period much larger than the symbol duration and are called long spreading codes. Therefore, most of the ....

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersten, "Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems," IEEE Transactions on Communications, vol. 44, no. 1, pp. 84--93, Jan 1996.

....have to be acquired first. In the PN code acquisition the purpose is to roughly estimate the timing of the PN code. Acquisition is declared if the timing error is less than half a chip duration. Conventional matched filter [6] being inadequate in near far scenario, subspace methods, e.g. [1, 7], are ofter proposed. They achieve near far resistance with tolerable computation, but are far too sub optimal in low SNR s or as the system is heavily loaded [4] In this paper, we consider computationally very simple near far resistant code acquisition. The underlying observation is the huge ....

E. Strom, S. Parkvall, S. Miller, and B. Ottersten, "Propagation Delay Estimation in Asynchronous Direct-Sequence Code Division Multiple Access Systems", IEEE Transactions on Communications, vol. 44, January 1996, pp. 84-93.

....Code division multiple access (CDMA) systems are inherently interference limited. Receivers can combat multiple access interference (MAI) by using multiuser channel estimation, detection and decoding algorithms. Several multiuser algorithms have been proposed for channel estimation in [1] 2] [3], 4] 5] 6] These algorithms are developed for CDMA systems with short spreading codes for the various users that repeat every symbol. However, spreading codes used in practical CDMA systems have a period much larger than the symbol duration and are called long spreading codes. Therefore, ....

....to the i th and i 1 th symbols of each user in the observation window. Thus, U i is of the form [U R 1;i U L 1;i 1 U R 2;i U L 2;i 1 : U R K;i U L K;i 1 ] where U R k;i = 2 6 6 6 6 6 6 6 6 6 6 6 4 c k;i [1] c k;i [2] c k;i [N ] 0 c k;i [2] c k;i [3] 0 0 . c k;i [N 1] c k;i [N ] 0 0 c k;i [N ] 0 0 0 3 7 7 7 7 7 7 7 7 7 7 7 5 ; is constructed with the right part of the spreading code of user k corresponding to symbol i and U L k;i 1 = 2 6 6 6 6 6 6 6 6 6 6 6 ....

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersten, \Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems," IEEE Transactions on Communications, vol. 44, no. 1, pp. 84-93, Jan 1996. DRAFT 24

....Code division multiple access (CDMA) systems are inherently interference limited. Receivers can combat multiple access interference (MAI) by using multiuser channel estimation, detection and decoding algorithms. Several multiuser algorithms have been proposed for channel estimation in [1] 2] [3], 4] 5] 6] These algorithms are developed for CDMA systems with short repeating (every symbol) spreading codes for the various users. However, spreading codes used in practical CDMA systems have a period much larger than the symbol duration (called long spreading codes) Therefore, most of ....

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersten, "Propagation delay estimation in asynchronous direct-sequence codedivision multiple access systems," IEEE Transactions on Communications, vol. 44, no. 1, pp. 84--93, Jan 1996.

....of the code of the user of interest only. 3. Algorithm The algorithm consists of two components: separation with matched filter followed by fixed point iteration. For clarity, the latter component is now considered first. First, the data is sphered. It is shown that under rather mild conditions [1, 8] the data have the subspace structure. The new data is thus of form Y = # 1 2 s U T s X, where # s and U s corresponds to 2K principal eigenvalues and eigenvectors of the correlation matrix estimate XX T N , respectively. In [3] a fast ICA algorithm was introduced to estimate ....

E. Strom, S. Parkvall, S. Miller, and B. Ottersten, "Propagation Delay Estimation in Asynchronous Direct-Sequence Code Division Multiple Access Systems ", IEEE Transactions on Communications, vol. 44, January 1996, pp. 84-93.

....unequal in the reception (so called near far situation) the performance can totally collapse. Therefore, power control by the uplink (e.g. basestation) would be needed. In any scenario, maximum likelihood method [2, 1] would be optimal but it is computationally exhausting. Subspace methods (e.g. [3, 4, 5]) are therefore often proposed, having sub optimal performance but intermediate complexity. However, their performance is also found to be inadequate in highly loaded systems (e.g. 6] Most promising approaches are di#erentially coherent methods which are studied e.g in [7, 8, 9] The methods ....

....(or at most the code length C) which is rank(G) min 2KL,C . That is also the rank of the signal part GR a G T of the correlation matrix R (with time lag zero) Furthermore, the rank of the signal subspace characterized by largest eigenvectors of R equals 2KL. Therefore, eigenvector based MUSIC [3, 4, 5] does not perform well in the highly loaded systems, when 2KL # C. We compare MUSIC with the new correlation based method in the simulations. It has the form d = arg max d U T s g(d) 2 (22) where U s consists of 2KL (or C 1, when 2KL # C) largest eigenvectors of estimated ....

E. Strom, S. Parkvall, S. Miller, and B. Ottersten, "Propagation Delay Estimation in Asynchronous DirectSequence Code Division Multiple Access Systems", IEEE Transactions on Communications, vol. 44, January 1996, pp. 84-93.

....It was demonstrated in [1] that PIC based SCCL loop has satisfactory tracking performance both in steady and transient state. The major problem with PIC based delay trackers is the complexity, which limits also the employment of other recent proposals for delay tracking in multiuser receivers [4, 5]. One of the simplest delay trackers for multiuser receivers is a standard coherent DLL. In this paper the performance of a decision directed coherent single user DLL and a PIC based delay tracker is studied. The loops are used in conjunction with a PIC based detector [2] 2. System Model The ....

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersen. Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems. IEEE Trans. Commun., 44(1):84--93, Jan. 1996.

....I Introduction Delay estimation is one of the most demanding problems which one has to solve when implementing spread spectrum receivers. In multiuser code division multipleaccess (CDMA) receivers delay estimation is even more critical, which has initiated a lot of research activity recently. In [1, 2] subspace methods have been applied to delay estimation. Subspace delay estimators estimate the projection matrix to signal subspace from the input signal sample covariance matrix. The maximum projection of the desired user signature vector onto signal subspace gives the delay estimate for the ....

....is a well defined procedure to solve (8) regardless of the rank deficiency of the code matrix Sd . Therefore, the Hestenes Stiefel CG algorithm can be used for implementing the iterative blind ML delay estimator. IV Numerical Results The numerical results presented earlier for delay estimators [1, 2, 3, 4] have been produced by using such assumptions about the system which are not necessarily valid in mobile cellular systems. The channel model has usually been single path AWGN and the number of users has been quite small for practical interest. We would like to know the performance of the ML delay ....

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersen. Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems. IEEE Trans. Commun., 44(1):84--93, Jan. 1996.

....optimization problem to several problems of less dimensions. The maximum likelihood type delay estimators for multiuser CDMA systems are very complex [2] 3] 4] One popular research topic in recent years has been subspace based estimators developed originally for antenna arrays [5] [6]. However, it has been shown that the subspace based delay estimators are not efficient in highly loaded systems [7] The solution for the delay acquisition scheme depends on the receiver structure itself. The blind least squares single user receivers have shown promising performance [8] 9] ....

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersen, "Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems," IEEE Trans. Commun. , vol. 44, no. 1, pp. 84--93, Jan. 1996.

....since delay estimation has been found to be one of the most demanding problems in practical systems. Maximum likelihood type delay estimators [36] 37] 38] for multiuser CDMA systems are very complex. One popular research topic in recent years has been subspace based delay estimators [39] [40] derived originally for spectrum estimation [41] 42] However, it has been shown that the subspace based delay estimators are not efficient in highly loaded systems [43] The blind least squares single user receivers can be used for delay acquisition by using the minimum variance principle. The ....

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersen, "Propagation delay estimation in asynchronous directsequence code-division multiple access systems," IEEE Trans. Commun., vol. 44, no. 1, pp. 84--93, Jan. 1996.

....Code division multiple access (CDMA) systems are inherently interference limited. Receivers can combat multiple access interference (MAI) by using multiuser channel estimation, detection and decoding algorithms. Several multiuser algorithms have been proposed for channel estimation in [1] 2] [3], 4] 5] 6] These algorithms are developed for CDMA systems with short repeating (every symbol) spreading codes for the various users. However, spreading codes used in practical CDMA systems have a period much larger than the symbol duration (called long spreading codes) Therefore, most of ....

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersten, "Propagation delay estimation in asynchronous direct-sequence codedivision multiple access systems," IEEE Transactions on Communications, vol. 44, no. 1, pp. 84--93, Jan 1996.

....due to the near far effect in a multi user environment. This problem is not just restricted to the demodulator, but also affects the performance of delay time acquisition and tracking for synchronization between the transmitter and receiver. Attempts to solve this problem can be seen in [1, 2, 3, 4, 5, 6] and the references therein. The most popular resistant near far delay time estimators are based on subspace techniques as in [3, 5, 6] The basic principle of subspace techniques is to estimate the signal noise subspace, normally through Eigendecomposition of the covariance matrix of a received ....

....of delay time acquisition and tracking for synchronization between the transmitter and receiver. Attempts to solve this problem can be seen in [1, 2, 3, 4, 5, 6] and the references therein. The most popular resistant near far delay time estimators are based on subspace techniques as in [3, 5, 6]. The basic principle of subspace techniques is to estimate the signal noise subspace, normally through Eigendecomposition of the covariance matrix of a received vector input. Assuming that the desired user code is known, the delay associated with the desired transmission can be obtained. Since ....

[Article contains additional citation context not shown here]

E. G. Strom, S. Parkvall, S. L. Miller and B. E. Ottersten, "Propagation Delay Estimation in Asynchronous DirectSequence Code-Division Multiple Access Systems" IEEE Trans. on Comms., Vol. 44, No. 1, pp. 84-93, Jan 1996.

....to a different user. We assume that perfect knowledge of the desired user s signature sequence, timing and DoA is available to the receiver. Knowledge of timing can be obtained using a joint timing acquisition and demodulation algorithm as in [12, 13] or using subspace based methods as in [14, 15]. Knowledge of DoA is less critical since the array response is insensitive to small changes in DoA. Thus, when the DoA is unknown, 3 d Antenna Array Signal q Figure 1: The antenna array with the defining parameters. it suffices to use a sectorization technique in which a small number of ....

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersten, "Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems", preprint, 1994.

....Code division multiple access (CDMA) systems are inherently interference limited. Receivers can combat multiple access interference (MAI) by using multiuser channel estimation, detection and decoding algorithms. Several multiuser algorithms have been proposed for channel estimation in [1] 2] [3], 4] 5] 6] These algorithms are developed for CDMA systems with short repeating (every symbol) spreading codes for the various users. However, spreading codes used in practical CDMA systems have a period much larger than the symbol duration (called long spreading codes) Therefore, most of ....

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersten, "Propagation delay estimation in asynchronous direct-sequence codedivision multiple access systems," IEEE Transactions on Communications, vol. 44, no. 1, pp. 84--93, Jan 1996.

....This is particularly true for channels with large delay spreads. We note that the first approach offers the advantage of yielding a more easily parameterized channel model at the expense of over parametrization. Much of the work performed for the estimation of the continuous delay of one path [2, 3, 7] or the continuous delays of a multi path channel has [1, 14] focused on rectangular chip pulse shapes. When rectangular pulse shapes are used, the correlation of the pulse is piecewise linear (a triangular pulse) and both data aided maximum likelihood and blind subspace algorithms yield closed ....

....do not exist and we need to perform numerical optimizations. To our knowledge, 19, 6] may be the only work which considers multiuser concepts with channel estimation for systems with bandwidth efficient pulses. In [6] it is shown 1 that the single path blind subspace based algorithms of [3, 7] can be extended for the case with bandlimited pulses. However, blind subspace algorithms with nonlinear pulse shapes result in high error variance and require a large number of symbols to achieve reasonable performance. In this work, we consider bandlimited pulses, but in contrast to [6] ....

[Article contains additional citation context not shown here]

E.G.Strom, S.Parkvall, S.L. Miller, and B.E.Ottersten. Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems. IEEE Transactions on Communications, 44(1):84--93, January 1996.

....over the same frequency band simultaneously. The identi cation of a particular user is done by his unique code. In the uplink (e.g. base station) signal processing the codes of the users are known but their delays are not known. For delay estimation matched lter [12] subspace approaches [2, 15] or maximum likelihood method [12] can be used. Once the delays are estimated, the other parameters such as the fading process or the symbols can be determined. The signal processing in the downlink (e.g. mobile phone) di ers from that of the uplink. First, only the own code of the mobile phone ....

E. Strom, S. Parkvall, S. Miller, and B. Ottersten. Propagation delay estimation in asynchronous direct-sequence code division multiple access. IEEE Trans. on Communications, (44):8493, 1996.

....all the signals sent by different users must be detected, but there is also much more signal processing capacity available. The codes of the all the users are known but their delays are unknown. For delay estimation, one can use for example the simple matched filter [12, 14] subspace approaches [1, 13], or the optimal but computationally highly demanding maximum likelihood method [12, 14] When the delays have been estimated, one can estimate the other parameters such as fading process and symbols [14] In downlink (mobile phone) signal processing, each user knows only their own code, while ....

E. Strom, S. Parkvall, S. Miller, and B. Ottersten. Propagation delay estimation in asynchronous direct-sequence code division multiple access systems. IEEE Trans. Communications, 44:84--93, January 1996.

....unknown coefficients [3] and no a priori information about the channels is exploited. We refer to this approach as unstructured channel estimation. More recently, a priori knowledge of the structure of the overall use channel impulse responses has been exploited in order to improve estimation [6, 7, 8, 9, 10, 11]. In some works, the special feature of rectangular chipshaping pulses is exploited to estimate directly the channel physical parameters (delays and path gains) In this way, a minimal number of unknowns needs to be estimated. On the other hand, these methods do not generalize easily to arbitrary ....

E. Strom, S. Parkvall, S. Miller, B. Ottersten, "Propagation delay estimation in asynchronous Direct-Sequence Code-Division Multiple-Access systems," IEEE Trans. on Commun., Vol. 44, No. 1, pp. 84 -- 93, Jan. 1996.

No context found.

E. Strom, S. Parkvall, S. Miller, B. Ottersten. Propagation delay estimation in asynchronous direct- sequence code-division multiple access systems. IEEE Trans. Commun., vol. 44, pages 84-93, Jan. 1996.

No context found.

E.G. Strom, S. Parkvall, S.L. Miller, B.E. Ottersten. Propagation delay estimation in asynchronous direct- sequence code-division multiple access systems. IEEE Trans. Commun. , pages 84-93, vol.44, Jan. 1996.

No context found.

E. Strom, S. Parkvall, S. Miller, B. Ottersten. Propagation delay estimation in asynchronous direct-sequence codedivision multiple access systems. IEEE Trans. Commun., vol. 44, pages 84-93, Jan. 1996.

....of Technology (KTH) S 100 44 Stockholm, Sweden E mail: ostman s3.kth. se Abstract An algorithm for time delay estimation in an asynchronous direct sequence code division multiple access (DS CDMA ) system which exploits the structure of the system better than previously known algorithms [1], is presented. The different algorithms are compared through simulation and asymptotic analysis (large number of vector samples) For a typical scenario it is shown that the proposed algorithm decreases the standard deviation of the time delay errors with a factor of approximately 2. 1 ....

....deviation of the time delay errors with a factor of approximately 2. 1 Introduction The aim of this work is to present an algorithm for time delay estimation in an asynchronous direct sequence code division multiple access (DS CDMA) system which exploits the structure of the system better than [1], and hence has better performance. The basic observation is that in [ 1] only the correlation matrix of lag 0 is considered. However, since the algorithm uses vector samples which are asynchronous to the sent data, two symbols will contribute to each vector sample and the sequence will be ....

[Article contains additional citation context not shown here]

E.G. StrOm, S. Parkvail, S. L. Miller, and B. E. Otter- sten, "Propagation Delay Estimation in Asynchronous Direct-Sequence Code-Division Multiple Access Sys- tems" IEEE Transactions on Communications, vol. 44, pp. 84-93, January 1996.

....the accuracy is of course much more convenient than having to resort to simulations. Recall that a natural measure of accuracy (for unbiased estimators) is the estimator variance. In this paper, we derive an analytical expression for the covariance matrix of the MUSIC propagation delay estimator [3]. II. System Model The system under consideration is modeled as an asynchronous K user DS CDMA system operating over an additive white Gaussian noise (AWGN) channel. The modulation scheme is BPSK with bit duration T and chip duration T c = T=N , where N is an integer. The code waveforms are ....

.... k ) As seen from this equation, the kth user contributes d k (m)fi k a 2k Gamma1 ( k ) if d k (m) d k (m Gamma 1) or d k (m)fi k a 2k ( k ) if d k (m) Gammad k (m Gamma 1) The vectors fa 2k Gamma1 ( k ) a 2k ( k )g are defined by the kth user s propagation delay and code waveform [3]. For convenience, we will use the notation a 2k Gammai = a 2k Gammai ( k ) for i = 0; 1, k = 1; 2; K. The received vector may be written as r(m) n(m) K X k=1 r k (m) ABz(m) n(m) where 3 A = a 1 a 2 Delta Delta Delta a 2K ] 2 IR QN Theta2K , B = diag(fi 1 ; fi 1 ; fi ....

[Article contains additional citation context not shown here]

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersten. "Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems." IEEE Transactions on Communications , vol. 44, no. 1,

....interest to study near far resistant methods for synchronization and detection. Lately, there has been an increased interest in subspace based methods for near far robust synchronization and channel estimation [1, 5] In this paper, we will modify the synchronization procedure proposed in [5] and [4] to take advantage of training sequences. The new algorithms are derived for an additive white Gaussian noise (AWGN) channel; however, they can be modified for time varying fading channels [3] II. System Model The system under consideration is modeled as an asynchronous K user DS CDMA system ....

.... k ) As seen from this equation, the kth user contributes d k (m)fi k a 2k Gamma1 ( k ) if d k (m) d k (m Gamma 1) or d k (m)fi k a 2k ( k ) if d k (m) Gammad k (m Gamma 1) The vectors fa 2k Gamma1 ( k ) a 2k ( k )g are defined by the kth user s propagation delay and code waveform [4]. The received vector may be written as r(m) n(m) K X k=1 r k (m) ABz(m) n(m) 1 Matrices and vectors are typeset in bold face. The (i; j)th element of a matrix A is denoted by A(i; j) A T and A denotes transpose and conjugate transpose, respectively. 2 ffi K (p) denotes the ....

[Article contains additional citation context not shown here]

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersten. "Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems." IEEE Trans. Commun., vol. 44, no. 1, pp. 84--93, January 1996.

....[1 4] All these receivers assume knowledge of the propagation delays for all users and may be sensitive to propagation delay estimation errors [5] especially in a near far scenario. The propagation delays must be estimated in a near far resistant manner, e.g. with the technique proposed in [6]. Since the algorithm proposed in [6] is able to estimate the propagation delays during data transmission, it is of interest to find techniques which facilitates data demodulation without prior knowledge of the propagation delays. Applications for the proposed receiver could be a soft handover ....

....knowledge of the propagation delays for all users and may be sensitive to propagation delay estimation errors [5] especially in a near far scenario. The propagation delays must be estimated in a near far resistant manner, e.g. with the technique proposed in [6] Since the algorithm proposed in [6] is able to estimate the propagation delays during data transmission, it is of interest to find techniques which facilitates data demodulation without prior knowledge of the propagation delays. Applications for the proposed receiver could be a soft handover scenario in a mobile communication ....

[Article contains additional citation context not shown here]

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersten, "Propagation Delay Estimation in Asynchronous Direct-Sequence Code-Division Multiple Access Systems" IEEE Transactions on Communications, vol. 44, pp. 84--93, January 1996.

.... motivated much research into subspace methods [11] and performance analysis [3, 5, 16, 17, 23, 24] Subspace methods have been applied to a number of other application areas as well, system identification [7, 9, 21, 22] image processing [1] blind channel estimation [8, 19] timing estimation [18] and signal separation in wireless communications [20] Common to all these application areas is a parameterized low rank signal model and the use of a decomposition technique to extract a dominant subspace or its orthogonal complement. The signal parameters may be estimated by matching the ....

....Multiple Access (DS CDMA) communication systems, the near far problem in the acquisition phase, i.e. the problem of estimating the propagation delay for a weak user in presence of strong interfering users, is of great concern. Subspace based delay estimation provides a solution to this problem [18], whereas conventional techniques fail. The received signal is sampled at a multiple Q of the chip rate and the samples are stacked in vectors, r(t) Each vector, consisting of QN scalar samples, corresponds to one 6 Ottersten et al. bit time, where each bit consists of N chips. The received ....

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersten, Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems, IEEE Transactions on Communications, 44 (1996), pp. 84--93.

....A modified correlator like technique was developed based on the MMSE (minimum mean squared errors) receiver [10] The MMSE timing estimator attains better near far resistance at the cost of some increased complexity. An alternative near far resistant code timing estimator is the MUSIC algorithm [11] [12] originally proposed by Schmidt [13] for direction of arrival estimation in array processing. Unlike the other techniques discussed herein, the MUSIC algorithm is blind in the sense that it needs no training for the purpose of code acquisition. However, the MUSIC timing estimator is ....

....delay will be within one bit interval (see, e.g. 19] and the references therein) using a side channel for initial synchronization is unnecessary. The receiver front end consists of an in phase quadrature (IQ) mixer followed by an integrateand dump filter with integration time T c (see, e.g. [11]) The equivalent received complex sequence, fy(l)g, can be expressed as (with double frequency terms ignored) y(l) K X k=1 a k p P k e j k 1 T c Z lTc (l Gamma1)T c s k (t Gamma k )dt n(l) l = 0; 1; MN Gamma 1; 5) 3 where k = 0 k Gamma c k , n(l) denotes ....

[Article contains additional citation context not shown here]

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersten, "Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems," IEEE Transactions on Communications, vol. 44, pp. 84--93, January 1996.

No context found.

E.G. Strom, S. Parkvall, S.L. Miller, and B.E. Ottersten, "Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems," IEEE Trans. on Comm., 44(1):84-93, Jan 1996.

No context found.

E. Strom, S. Parkvall, S. Miller, and B. Ottersten. Propagation delay estimation in asynchronous direct-sequence code division multiple access. IEEE Trans. on Communications, (44):8493, 1996.

No context found.

E. Strom, S. Parkvall, S. Miller, and B. Ottersten. Propagation delay estimation in asynchronous directsequence code division multiple access. IEEE Trans. Communications, 44:84--93, January 1996.

No context found.

E. G. Strom, S. Parkvall, S. L. Miller, B. E. Ottersten, "Propagation Delay Estimation in Asynchronous Direct-Sequence Code-Division Multiple Access Systems," IEEE Trans. Commun. , Vol. 44, pp. 84-93, Jan. 1996.

No context found.

E.G. Strom, S. Parkvall, S.L. Miller, and B.E. Ottersten, "Propagation Delay Estimation in Asynchronous Direct-sequence Code-division Multiple Access Systems" IEEE Trans. on Comm., Vol. 44 n. 1 , pp. 84-93, Jan. 1996.

No context found.

E. Strom, S. Parkvall, S. Miller and B. Ottersten, "Propagation Delay Estimation in Asynchronous Direct-Sequence Code-Division Multiple Access Systems ", IEEE Trans. on Comm., vol.44, no.1, pp.84-93, Jan. 1996

No context found.

E.G. Strom, S. Parkvall, S.L. Miller, and B. Ottersten, "Propagation delay estimation in asynchronous direct-sequence codedivision multiple access systems," IEEE Transactions on Communications, vol. 44, no. 1, pp. 84--93, Jan. 1996.

No context found.

E. G. Strom, S. Parkvall, S. L. Miller, B. E. Ottersten, Propagation Delay Estimation in Asynchronous Direct-Sequence Code-Division Multiple Access Systems, IEEE Transactions on Communications, January 1996, pp. 84-93

No context found.

E. G. Strom, S. Parkvall, S. L. Miller and B. E. Ottersten, "Propagation delay estimation in asynchronous direct-sequence codedivision multiple access systems," IEEE Trans. Commun., Jan. 1996.

No context found.

E. G. Strom, S. Parkvall, S. L. Miller and B. E. Ottersten, "Propagation delay estimation in asynchronous direct-sequence codedivision multiple access systems," IEEE Trans. Commun., Jan. 1996.

No context found.

E. G. Strom, S. Parkvall, S. L. Miller, B. E. Ottersten, Propagation Delay Estimation in Asynchronous Direct-Sequence Code-

No context found.

E. Strom, S. Parkvall, S. Miller, and B. Ottersten, "Propagation Delay Estimation in Asynchronous Direct-Sequence Code-Division Multiple Access Systems ", IEEE Trans. Commun., pages 84--93, January 1996.

No context found.

E. G. Strom, S. Parkval, S. L. Miller, and B. E. Ottersten, "Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems," IEEE Transactions on Communications, vol. 44, no. 1, pp. 84--93, January 1996.

No context found.

E. G. Strom, S. Parkvall, S. L. Miller, and B. E. Ottersen. Propagation delay estimation in asynchronous direct-sequence code-division multiple access systems. IEEE Transactions on Communications, 44(1):84-93, January 1996.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC